From 9a2ecce0f5d284ba44f68f5f49a436dcc6f60227 Mon Sep 17 00:00:00 2001
From: Lando-L <hello.lando@icloud.com>
Date: Wed, 10 Apr 2024 09:20:30 +0100
Subject: [PATCH 1/9] feat: update optimizer to return acquisition value

---
 boax/optimization/optimizers/alias.py         |  20 +-
 boax/optimization/optimizers/base.py          |   6 +-
 docs/source/guides/Batched_Optimization.ipynb |   2 +-
 .../guides/Constrained_Optimization.ipynb     |   2 +-
 docs/source/guides/Fitting_With_Priors.ipynb  |   2 +-
 docs/source/guides/Getting_Started.ipynb      |   2 +-
 docs/source/guides/Untitled.ipynb             | 494 ++++++++++++++++++
 .../optimizers/optimizers_test.py             |   6 +-
 8 files changed, 519 insertions(+), 15 deletions(-)
 create mode 100644 docs/source/guides/Untitled.ipynb

diff --git a/boax/optimization/optimizers/alias.py b/boax/optimization/optimizers/alias.py
index 185d06d..f83e5e4 100644
--- a/boax/optimization/optimizers/alias.py
+++ b/boax/optimization/optimizers/alias.py
@@ -14,6 +14,8 @@
 
 """Alias for optimizers."""
 
+from functools import partial
+
 from jax import numpy as jnp
 from jax import random
 
@@ -21,6 +23,7 @@
 from boax.optimization.optimizers.base import Optimizer
 from boax.optimization.optimizers.initializers.base import Initializer
 from boax.optimization.optimizers.solvers.base import Solver
+from boax.utils.functools import compose
 
 
 def batch(initializer: Initializer, solver: Solver) -> Optimizer:
@@ -53,7 +56,9 @@ def optimizer(key, fun, bounds, q, num_samples, num_restarts):
     candidates = initializer(key2, x, y, num_restarts)
     next_candidates, values = solver(fun, bounds, candidates)
 
-    return next_candidates[jnp.argmax(values)]
+    idx = jnp.argmax(values)
+
+    return next_candidates[idx], values[idx]
 
   return optimizer
 
@@ -77,11 +82,14 @@ def sequential(initializer: Initializer, solver: Solver) -> Optimizer:
   inner = batch(initializer, solver)
 
   def optimizer(key, fun, bounds, q, num_samples, num_restarts):
-    return jnp.concatenate(
-      [
-        inner(random.fold_in(key, i), fun, bounds, 1, num_samples, num_restarts)
-        for i in range(q)
-      ]
+    next_candidates, values = zip(*(
+      inner(random.fold_in(key, i), fun, bounds, 1, num_samples, num_restarts)
+      for i in range(q)
+    ))
+
+    return (
+      jnp.concatenate(list(next_candidates)),
+      jnp.array(list(values))
     )
 
   return optimizer
diff --git a/boax/optimization/optimizers/base.py b/boax/optimization/optimizers/base.py
index f68eac5..e5bca01 100644
--- a/boax/optimization/optimizers/base.py
+++ b/boax/optimization/optimizers/base.py
@@ -14,7 +14,7 @@
 
 """Base interface for optimizers."""
 
-from typing import Callable, Protocol
+from typing import Callable, Protocol, Tuple
 
 from boax.utils.typing import Array, PRNGKey
 
@@ -32,7 +32,7 @@ def __call__(
     q: int,
     num_samples: int,
     num_restarts: int,
-  ) -> Array:
+  ) -> Tuple[Array, Array]:
     """
     The optimization function.
 
@@ -45,5 +45,5 @@ def __call__(
       num_restarts: The number of restarts.
 
     Returns:
-      The maxima resulting of the optimization.
+      A tuple of the maxima and their acquisition values.
     """
diff --git a/docs/source/guides/Batched_Optimization.ipynb b/docs/source/guides/Batched_Optimization.ipynb
index 1bf8543..a4da54a 100644
--- a/docs/source/guides/Batched_Optimization.ipynb
+++ b/docs/source/guides/Batched_Optimization.ipynb
@@ -438,7 +438,7 @@
     "                    solver=optimizers.solvers.scipy(method='bfgs'),\n",
     "                )\n",
     "    \n",
-    "        next_x = bfgs(\n",
+    "        next_x, _ = bfgs(\n",
     "            random.fold_in(optimizer_key, i),\n",
     "            acqf,\n",
     "            bounds,\n",
diff --git a/docs/source/guides/Constrained_Optimization.ipynb b/docs/source/guides/Constrained_Optimization.ipynb
index b42e861..4f3a374 100644
--- a/docs/source/guides/Constrained_Optimization.ipynb
+++ b/docs/source/guides/Constrained_Optimization.ipynb
@@ -497,7 +497,7 @@
     "                    )\n",
     "                )\n",
     "    \n",
-    "        next_x = bfgs(\n",
+    "        next_x, _ = bfgs(\n",
     "            random.fold_in(optimizer_key, i),\n",
     "            acqf,\n",
     "            bounds,\n",
diff --git a/docs/source/guides/Fitting_With_Priors.ipynb b/docs/source/guides/Fitting_With_Priors.ipynb
index 3f8a043..53ff3b2 100644
--- a/docs/source/guides/Fitting_With_Priors.ipynb
+++ b/docs/source/guides/Fitting_With_Priors.ipynb
@@ -556,7 +556,7 @@
     "    # Optimizing\n",
     "    acqf = acquisition_fn(model, beta)\n",
     "    \n",
-    "    next_x = bfgs(\n",
+    "    next_x, _ = bfgs(\n",
     "        random.fold_in(key, i),\n",
     "        acqf,\n",
     "        bounds,\n",
diff --git a/docs/source/guides/Getting_Started.ipynb b/docs/source/guides/Getting_Started.ipynb
index 3869b7e..bb93ddc 100644
--- a/docs/source/guides/Getting_Started.ipynb
+++ b/docs/source/guides/Getting_Started.ipynb
@@ -537,7 +537,7 @@
     "    # Selecting\n",
     "    acqf = acquisition_fn(model)\n",
     "    \n",
-    "    next_x = bfgs(\n",
+    "    next_x, _ = bfgs(\n",
     "        random.fold_in(key, i),\n",
     "        acqf,\n",
     "        bounds,\n",
diff --git a/docs/source/guides/Untitled.ipynb b/docs/source/guides/Untitled.ipynb
new file mode 100644
index 0000000..20cc81c
--- /dev/null
+++ b/docs/source/guides/Untitled.ipynb
@@ -0,0 +1,494 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "61b0cf2e-48b7-425f-b32b-e90c7e301f89",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from jax import config\n",
+    "\n",
+    "config.update(\"jax_enable_x64\", True)\n",
+    "\n",
+    "from jax import numpy as jnp\n",
+    "from jax import jit, lax, nn, random, value_and_grad, vmap\n",
+    "\n",
+    "import optax\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.style.use('bmh')\n",
+    "\n",
+    "from boax.core import distributions, samplers\n",
+    "from boax.prediction import models, objectives\n",
+    "from boax.optimization import acquisitions, optimizers"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "55860138-87d3-4ce2-8ddc-bf7faed007ba",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data_key, optimizer_key = random.split(random.key(0))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "087ac062-9584-40e3-ad0f-3d1e707d9bd5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def objective(x):\n",
+    "    return -((x[..., 0] + 1) ** 2) * jnp.sin(2 * x[..., 0] + 2) / 5 + 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "a23cb1db-ca8d-4568-b391-4038352632b4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def approximate(x):\n",
+    "    return 0.5 * objective(x) + x[..., 0] / 4 + 2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "dc827755-2117-4d1a-ae36-f3b54eb29ce9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "bounds = jnp.array([[-5.0, 5.0]])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "36aeac5b-ede2-49fc-845e-1b4bd87970e8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fidelities = jnp.array([0.5, 1.0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "4c59393e-f1e3-46a5-83a9-1707ccbc363c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "x_train_values = random.uniform(random.fold_in(data_key, 0), minval=bounds[:, 0], maxval=bounds[:, 1], shape=(10, 1))\n",
+    "x_train_fidelities = random.randint(random.fold_in(data_key, 1), minval=0, maxval=2, shape=(10, 1))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "99e42b33-fc9c-4945-ae87-b876593257b8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "x_train = jnp.concatenate(\n",
+    "    [x_train_values, fidelities[x_train_fidelities]], axis=-1,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "e3a5325a-9dbb-4f8e-8d9f-28be117b1656",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_train = jnp.where(\n",
+    "    x_train_fidelities[..., 0],\n",
+    "    objective(x_train_values),\n",
+    "    approximate(x_train_values)\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "94fb7e2e-b4b1-4e29-a32e-5ea7b1ec4afe",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "xs = jnp.linspace(bounds[:, 0], bounds[:, 1], 501)\n",
+    "ys_true = objective(xs)\n",
+    "ys_approx = approximate(xs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "b99309b2-51cb-44bd-a915-607e0b33607c",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots(figsize=(12, 4))\n",
+    "\n",
+    "ax.plot(xs, ys_true, c='r', label='objective')\n",
+    "ax.plot(xs, ys_approx, c='k', linestyle='--', label='approximation')\n",
+    "ax.scatter(x_train_values, y_train, marker='x', c='k', label='observations')\n",
+    "ax.legend()\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "c94c2231-34fb-42de-b29d-5863ba4c5864",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def model_fn(params, x_train, y_train):\n",
+    "    return models.multi_fidelity(\n",
+    "        models.means.zero(),\n",
+    "        lambda fid1, fid2: models.kernels.scaled(\n",
+    "            models.kernels.linear_truncated(\n",
+    "                models.kernels.matern_five_halves(nn.softplus(params['unbiased'])),\n",
+    "                models.kernels.matern_five_halves(nn.softplus(params['biased'])),\n",
+    "                nn.softplus(params['power']),\n",
+    "            )(fid1, fid2),\n",
+    "            nn.softplus(params['amplitude']),\n",
+    "        ),\n",
+    "        models.likelihoods.gaussian(nn.softplus(params['noise'])),\n",
+    "        x_train,\n",
+    "        y_train,\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "3059ac02-6e3e-4553-9445-9a1a2aa28b69",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def loss_fn(params, x_train, y_train):\n",
+    "    y_hat = model_fn(params, None, None)(x_train)\n",
+    "    \n",
+    "    objective = objectives.penalized(\n",
+    "        objectives.negative_log_likelihood(\n",
+    "            distributions.multivariate_normal.logpdf\n",
+    "        ),\n",
+    "        -jnp.sum(\n",
+    "            distributions.gamma.logpdf(\n",
+    "                distributions.gamma.gamma(2.0, 0.15),\n",
+    "                nn.softplus(params['amplitude']),\n",
+    "            )\n",
+    "        ),\n",
+    "        -jnp.sum(\n",
+    "            distributions.gamma.logpdf(\n",
+    "                distributions.gamma.gamma(3.0, 3.0),\n",
+    "                nn.softplus(params['power']),\n",
+    "            )\n",
+    "        ),\n",
+    "        -jnp.sum(\n",
+    "            distributions.gamma.logpdf(\n",
+    "                distributions.gamma.gamma(3.0, 6.0),\n",
+    "                nn.softplus(params['unbiased']),\n",
+    "            )\n",
+    "        ),\n",
+    "        -jnp.sum(\n",
+    "            distributions.gamma.logpdf(\n",
+    "                distributions.gamma.gamma(6.0, 2.0),\n",
+    "                nn.softplus(params['biased']),\n",
+    "            )\n",
+    "        )\n",
+    "    )\n",
+    "\n",
+    "    return objective(y_hat, y_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "c6ab3460-015c-4963-99e0-d24fa45fcc83",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "params = {\n",
+    "    'biased': jnp.zeros(()),\n",
+    "    'unbiased': jnp.zeros(()),\n",
+    "    'power': jnp.zeros(()),\n",
+    "    'amplitude': jnp.zeros(()),\n",
+    "    'noise': jnp.zeros(()),\n",
+    "}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "bf78868e-dbe1-444d-b71b-e3d2f6cb8248",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "adam = optax.adam(0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "ba5b1bd4-c2b0-4f48-83de-baf72512866c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def fit(x_train, y_train):\n",
+    "    def step(state, i):\n",
+    "        loss, grads = value_and_grad(loss_fn)(state[0], x_train, y_train)\n",
+    "        updates, opt_state = adam.update(grads, state[1])\n",
+    "        params = optax.apply_updates(state[0], updates)\n",
+    "\n",
+    "        return (params, opt_state), loss\n",
+    "\n",
+    "    (next_params, _), _ = lax.scan(\n",
+    "        jit(step),\n",
+    "        (params, adam.init(params)),\n",
+    "        jnp.arange(500),\n",
+    "    )\n",
+    "\n",
+    "    return next_params"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "d403af5d-872a-4cac-b0eb-e1be0c33fc73",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_mean, y_var = jnp.mean(y_train), jnp.var(y_train)\n",
+    "y_norm = nn.standardize(y_train, mean=y_mean, variance=y_var)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "4b41a926-7235-4335-8767-11164e14ac58",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "next_params = fit(x_train, y_norm)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "44291f9c-614c-4b3b-97aa-5a84cd3f9dca",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model = models.scaled(\n",
+    "    models.outcome_transformed(\n",
+    "        model_fn(next_params, x_train, y_norm),\n",
+    "        distributions.mvn_to_norm,\n",
+    "    ),\n",
+    "    distributions.normal.scale,\n",
+    "    loc=y_mean,\n",
+    "    scale=jnp.sqrt(y_var),\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "05a0bc3d-151b-488f-ada7-a7a1eba3f7a6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_hat = model(jnp.hstack([xs, jnp.ones((501, 1))]))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "633e5615-7028-4d1d-81f2-f9a516ca6ee4",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots(figsize=(12, 4))\n",
+    "\n",
+    "ax.plot(xs, ys_true, c='r', label='objective')\n",
+    "ax.plot(xs, ys_approx, c='k', linestyle='--', label='approximation')\n",
+    "ax.scatter(x_train_values, y_train, marker='x', c='k', label='observations')\n",
+    "\n",
+    "ax.plot(xs, y_hat.loc, label='mean')\n",
+    "ax.fill_between(\n",
+    "    xs.flatten(),\n",
+    "    y_hat.loc - 2 * y_hat.scale,\n",
+    "    y_hat.loc + 2 * y_hat.scale,\n",
+    "    alpha=0.3,\n",
+    "    label='95% CI'\n",
+    ")\n",
+    "\n",
+    "ax.legend(loc='upper left')\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "b9d81dbd-2ea0-4e15-a346-e7f7117e31b0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def cost(x):\n",
+    "    return 10 + x[..., 1]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "id": "d90d542b-4e0c-4194-bf8d-0a6dd1b505c6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model_cost = models.joined(\n",
+    "    model,\n",
+    "    cost,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "id": "c9e415c2-e969-45e5-9de5-42b9f27c2b8d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "key1, key2 = random.split(random.key(0))\n",
+    "s, n = 32, 10\n",
+    "\n",
+    "best = 0.0\n",
+    "loc, scale = random.uniform(key1, (2, n, s, 1))\n",
+    "cost = random.uniform(key2, (n,))\n",
+    "preds = distributions.normal.normal(loc, scale)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "id": "8d532346-278c-4cbb-a5b7-4b17ef0430af",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(10,)"
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "cost.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "id": "4f79deff-d624-4c47-a390-40bfc6561301",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(10, 32)"
+      ]
+     },
+     "execution_count": 46,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "jnp.squeeze(preds.loc - best, axis=-1).shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "id": "4968d650-3b19-4b22-ba3f-ac32940c69b0",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Array([0.73711553, 1.26476817, 0.92278978, 0.68727337, 0.50606529,\n",
+       "       0.67834267, 0.86699063, 0.61253229, 1.11807063, 2.28338359],      dtype=float64)"
+      ]
+     },
+     "execution_count": 53,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "jnp.mean((jnp.squeeze(preds.loc - best, axis=-1) / cost[..., jnp.newaxis]), axis=-1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "286b4fba-4de9-4eee-8112-3883bd7505e1",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/tests/optimization/optimizers/optimizers_test.py b/tests/optimization/optimizers/optimizers_test.py
index 71a1ce0..0c60ecf 100644
--- a/tests/optimization/optimizers/optimizers_test.py
+++ b/tests/optimization/optimizers/optimizers_test.py
@@ -18,7 +18,7 @@ def test_batch(self):
     acqf = itemgetter((..., 0, 0))
     bounds = jnp.array([[-1.0, 1.0]])
 
-    next_x = optimizers.batch(initializer, solver)(
+    next_x, next_a = optimizers.batch(initializer, solver)(
       key,
       acqf,
       bounds,
@@ -28,6 +28,7 @@ def test_batch(self):
     )
 
     self.assertEqual(next_x.shape, (q, d))
+    self.assertEqual(next_a.shape, ())
 
   def test_sequential(self):
     key = random.key(0)
@@ -39,7 +40,7 @@ def test_sequential(self):
     acqf = itemgetter((..., 0, 0))
     bounds = jnp.array([[-1.0, 1.0]])
 
-    next_x = optimizers.sequential(initializer, solver)(
+    next_x, next_a = optimizers.sequential(initializer, solver)(
       key,
       acqf,
       bounds,
@@ -49,6 +50,7 @@ def test_sequential(self):
     )
 
     self.assertEqual(next_x.shape, (q, d))
+    self.assertEqual(next_a.shape, (q,))
 
 
 if __name__ == '__main__':

From 6ce2be77c561fe6462cb814591918bdbf18f7a67 Mon Sep 17 00:00:00 2001
From: Lando-L <hello.lando@icloud.com>
Date: Wed, 10 Apr 2024 09:04:18 +0100
Subject: [PATCH 2/9] feat: add multi-fidelity optimization

---
 boax/optimization/acquisitions/__init__.py    |  1 +
 boax/optimization/acquisitions/alias.py       | 62 +++++++++---
 boax/prediction/models/__init__.py            |  1 +
 boax/prediction/models/transformed.py         | 33 ++++++-
 .../acquisitions/acquisitions_test.py         | 14 +++
 tests/prediction/models/models_test.py        | 97 ++++++++++---------
 6 files changed, 147 insertions(+), 61 deletions(-)

diff --git a/boax/optimization/acquisitions/__init__.py b/boax/optimization/acquisitions/__init__.py
index 94b5449..8648f5e 100644
--- a/boax/optimization/acquisitions/__init__.py
+++ b/boax/optimization/acquisitions/__init__.py
@@ -28,6 +28,7 @@
 from .alias import q_probability_of_improvement as q_probability_of_improvement
 from .alias import q_upper_confidence_bound as q_upper_confidence_bound
 from .alias import upper_confidence_bound as upper_confidence_bound
+from .alias import q_multi_fidelity_knowledge_gradient as q_multi_fidelity_knowledge_gradient
 from .base import Acquisition as Acquisition
 from .transformed import constrained as constrained
 from .transformed import log_constrained as log_constrained
diff --git a/boax/optimization/acquisitions/alias.py b/boax/optimization/acquisitions/alias.py
index 0873508..90888bc 100644
--- a/boax/optimization/acquisitions/alias.py
+++ b/boax/optimization/acquisitions/alias.py
@@ -16,7 +16,8 @@
 
 import math
 from functools import partial
-from operator import attrgetter
+from operator import attrgetter, itemgetter
+from typing import Callable, Tuple
 
 from jax import jit, lax, scipy
 from jax import numpy as jnp
@@ -25,7 +26,7 @@
 from boax.core.distributions.normal import Normal
 from boax.optimization.acquisitions import functions
 from boax.optimization.acquisitions.base import Acquisition
-from boax.utils.functools import compose
+from boax.utils.functools import apply, compose, unwrap
 from boax.utils.typing import Array, Numeric
 
 
@@ -41,7 +42,7 @@ def probability_of_improvement(
 
   Example:
     >>> acqf = probability_of_improvement(0.2)
-    >>> poi = acqf(model(xs))
+    >>> poi = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -71,7 +72,7 @@ def log_probability_of_improvement(
 
   Example:
     >>> acqf = log_probability_of_improvement(0.2)
-    >>> log_poi = acqf(model(xs))
+    >>> log_poi = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -103,7 +104,7 @@ def expected_improvement(
 
   Example:
     >>> acqf = expected_improvement(0.2)
-    >>> ei = acqf(model(xs))
+    >>> ei = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -138,7 +139,7 @@ def log_expected_improvement(
 
   Example:
     >>> acqf = log_expected_improvement(0.2)
-    >>> log_ei = acqf(model(xs))
+    >>> log_ei = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -168,7 +169,7 @@ def upper_confidence_bound(
 
   Example:
     >>> acqf = upper_confidence_bound(2.0)
-    >>> ucb = acqf(model(xs))
+    >>> ucb = acqf(vmap(model)(xs))
 
   Args:
     beta: The mean and covariance trade-off parameter.
@@ -191,7 +192,7 @@ def posterior_mean() -> Acquisition:
 
   Example:
     >>> acqf = posterior_mean()
-    >>> mean = acqf(model(xs))
+    >>> mean = acqf(vmap(model)(xs))
 
   Args:
     model: A gaussian process regression surrogate model.
@@ -214,7 +215,7 @@ def posterior_scale() -> Acquisition:
 
   Example:
     >>> acqf = posterior_scale()
-    >>> scale = acqf(model(xs))
+    >>> scale = acqf(vmap(model)(xs))
 
   Args:
     model: A gaussian process regression surrogate model.
@@ -248,7 +249,7 @@ def q_probability_of_improvement(
 
   Example:
     >>> acqf = q_probability_of_improvement(1.0, 0.2)
-    >>> qpoi = acqf(model(xs))
+    >>> qpoi = acqf(vmap(model)(xs))
 
   Args:
     tau: The temperature parameter.
@@ -281,7 +282,7 @@ def q_expected_improvement(
 
   Example:
     >>> acqf = q_expected_improvement(0.2)
-    >>> qei = acqf(model(xs))
+    >>> qei = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -311,7 +312,7 @@ def q_upper_confidence_bound(
 
   Example:
     >>> acqf = q_upper_confidence_bound(2.0)
-    >>> qucb = acqf(model(xs))
+    >>> qucb = acqf(vmap(model)(xs))
 
   Args:
     beta: The mean and covariance trade-off parameter.
@@ -339,7 +340,7 @@ def q_knowledge_gradient(
 
   Example:
     >>> acqf = q_knowledge_gradient(0.2)
-    >>> qucb = acqf(model(xs))
+    >>> qkg = acqf(vmap(model)(xs))
 
   Args:
     best: The best function value observed so far.
@@ -356,3 +357,38 @@ def q_knowledge_gradient(
       attrgetter('loc'),
     )
   )
+
+
+def q_multi_fidelity_knowledge_gradient(
+  best: Numeric,
+  cost_fn: Callable,
+) -> Acquisition[Tuple[Normal, Array]]:
+  """
+  MC-based batch multi-fidelity Knowledge Gradient acquisition function.
+
+  Example:
+    >>> acqf = q_knowledge_gradient(0.2, cost_fn)
+    >>> qmfkg = acqf(vmap(model)(xs))
+
+  Args:
+    best: The best function value observed so far.
+
+  Returns:
+    The corresponding `Acquisition`.
+  """
+
+  return jit(
+    compose(
+      partial(jnp.mean, axis=-1),
+      apply(
+        unwrap(cost_fn),
+        compose(
+          partial(jnp.squeeze, axis=-1),
+          partial(lax.sub, y=best),
+          attrgetter('loc'),
+          itemgetter(0),
+        ),
+        itemgetter(1)
+      )
+    )
+  )
diff --git a/boax/prediction/models/__init__.py b/boax/prediction/models/__init__.py
index 4996c4e..c09bd65 100644
--- a/boax/prediction/models/__init__.py
+++ b/boax/prediction/models/__init__.py
@@ -25,3 +25,4 @@
 from .transformed import outcome_transformed as outcome_transformed
 from .transformed import sampled as sampled
 from .transformed import scaled as scaled
+from .transformed import fantasized as fantasized
diff --git a/boax/prediction/models/transformed.py b/boax/prediction/models/transformed.py
index 745e810..9f3ca1c 100644
--- a/boax/prediction/models/transformed.py
+++ b/boax/prediction/models/transformed.py
@@ -20,7 +20,7 @@
 from jax import vmap
 
 from boax.prediction.models.base import Model
-from boax.utils.functools import apply, call, compose
+from boax.utils.functools import apply, call, compose, identity, unwrap
 from boax.utils.typing import Array
 
 A = TypeVar('A')
@@ -152,3 +152,34 @@ def sampled(
     partial(partial, sample_fn),
     model,
   )
+
+
+def fantasized(
+  model: Model[Array],
+  fantasy_fn: Callable[[Array, Array], Model[T]],
+  fantasy_samples: Array,
+) -> Model[T]:
+  """
+  Constructs a fantasy model.
+
+  Example:
+    >>> transformed = fantasized(model, fantasy_fn, fantasy_samples)
+    >>> result = transformed(xs)
+
+  Args:
+    model: The base model.
+    fantasy_fn: The fantasy function.
+    fantasy_samples: The fantasy samples.
+
+  Returns:
+    The transformed `Model` function.
+  """
+
+  return compose(
+    call(fantasy_samples),
+    apply(
+      unwrap(fantasy_fn),
+      identity,
+      model,
+    ),
+  )
diff --git a/tests/optimization/acquisitions/acquisitions_test.py b/tests/optimization/acquisitions/acquisitions_test.py
index 3570aad..1a1fe6b 100644
--- a/tests/optimization/acquisitions/acquisitions_test.py
+++ b/tests/optimization/acquisitions/acquisitions_test.py
@@ -113,6 +113,20 @@ def test_q_knowledge_gradient(self):
     qkg = acquisitions.q_knowledge_gradient(best)(preds)
 
     self.assertEqual(qkg.shape, (n,))
+  
+  def test_q_multi_fidelity_knowledge_gradient(self):
+    key1, key2 = random.split(random.key(0))
+    s, n = 32, 10
+
+    best = 0.0
+    cost_fn = lambda a, b: a / b[..., jnp.newaxis]
+    loc, scale = random.uniform(key1, (2, n, s, 1))
+    preds = distributions.normal.normal(loc, scale)
+    costs = random.uniform(key2, (n,))
+
+    qmfkg = acquisitions.q_multi_fidelity_knowledge_gradient(best, cost_fn)((preds, costs))
+
+    self.assertEqual(qmfkg.shape, (n,))
 
   def test_constrained(self):
     key = random.key(0)
diff --git a/tests/prediction/models/models_test.py b/tests/prediction/models/models_test.py
index 1515b06..3efe417 100644
--- a/tests/prediction/models/models_test.py
+++ b/tests/prediction/models/models_test.py
@@ -1,3 +1,4 @@
+import numpy as np
 from absl.testing import absltest, parameterized
 from jax import numpy as jnp
 from jax import random
@@ -5,6 +6,7 @@
 from boax.core import distributions
 from boax.prediction import models
 from boax.prediction.models import kernels, likelihoods, means
+from boax.utils.functools import const
 
 
 class ProcessesTest(parameterized.TestCase):
@@ -44,7 +46,7 @@ def test_gaussian_process(self):
     self.assertEqual(mean.shape, (10,))
     self.assertEqual(cov.shape, (10, 10))
 
-  def test_multi_fidelity_regression(self):
+  def test_multi_fidelity(self):
     key1, key2 = random.split(random.key(0))
 
     index_points = random.uniform(key1, shape=(10, 2), minval=-1, maxval=1)
@@ -91,74 +93,75 @@ def test_multi_fidelity_regression(self):
   def test_scaled(self):
     key1, key2 = random.split(random.key(0))
 
-    index_points = random.uniform(key1, shape=(10, 1), minval=-1, maxval=1)
-    loc, scale = random.uniform(key2, shape=(2, 1), minval=0, maxval=1)
+    preds = distributions.normal.normal(*random.uniform(key1, shape=(2, 10)))
+    scaled = random.uniform(key2, shape=(2, 1))
 
     model = models.scaled(
-      models.gaussian_process(
-        means.zero(),
-        kernels.rbf(jnp.array(0.2)),
-        likelihoods.gaussian(1e-4),
-      ),
-      distributions.multivariate_normal.scale,
-      loc=loc,
-      scale=scale,
+      const(preds),
+      distributions.normal.scale,
+      loc=scaled[0],
+      scale=scaled[1],
     )
 
-    mean, cov = model(index_points)
+    result = model(jnp.empty((10,)))
 
-    self.assertEqual(mean.shape, (10,))
-    self.assertEqual(cov.shape, (10, 10))
+    expected = distributions.normal.scale(
+      preds,
+      scaled[0],
+      scaled[1],
+    )
+
+    np.testing.assert_allclose(result.loc, expected.loc, atol=1e-4)
+    np.testing.assert_allclose(result.scale, expected.scale, atol=1e-4)
 
   def test_sampled(self):
     key1, key2 = random.split(random.key(0))
 
-    index_points = random.uniform(key1, shape=(10, 1), minval=-1, maxval=1)
+    preds = distributions.normal.normal(*random.uniform(key1, shape=(2, 10)))
+    samples = random.normal(key2, shape=(5, 10))
 
     model = models.sampled(
-      models.gaussian_process(
-        means.zero(),
-        kernels.rbf(jnp.array(0.2)),
-        likelihoods.gaussian(1e-4),
-      ),
-      distributions.multivariate_normal.sample,
-      random.normal(key2, shape=(5, 10)),
+      const(preds),
+      lambda _, s: s,
+      samples,
     )
 
-    samples = model(index_points)
+    result = model(jnp.empty((10,)))
 
-    self.assertEqual(
-      samples.shape,
-      (
-        5,
-        10,
-      ),
-    )
+    np.testing.assert_allclose(result, samples, atol=1e-4)
 
   def test_joined(self):
     key = random.key(0)
 
-    index_points = random.uniform(key, shape=(10, 1), minval=-1, maxval=1)
+    samples = random.uniform(key, shape=(4, 10))
+    preds1 = distributions.normal.normal(samples[0], samples[1])
+    preds2 = distributions.normal.normal(samples[2], samples[3])
 
-    model = models.joined(
-      models.gaussian_process(
-        means.zero(),
-        kernels.rbf(jnp.array(0.2)),
-        likelihoods.gaussian(1e-4),
-      ),
-      models.gaussian_process(
-        means.zero(),
-        kernels.rbf(jnp.array(0.5)),
-        likelihoods.gaussian(1e-4),
-      ),
+    model = models.joined(const(preds1), const(preds2))
+
+    result1, result2 = model(jnp.empty((10,)))
+
+    np.testing.assert_allclose(result1.loc, preds1.loc, atol=1e-4)
+    np.testing.assert_allclose(result1.scale, preds1.scale, atol=1e-4)
+    np.testing.assert_allclose(result2.loc, preds2.loc, atol=1e-4)
+    np.testing.assert_allclose(result2.scale, preds2.scale, atol=1e-4)
+  
+  def test_fantazised(self):
+    key1, key2 = random.split(random.key(0))
+
+    samples = random.uniform(key1, shape=(10, 3, 1))
+    preds = distributions.normal.normal(*random.uniform(key2, shape=(2, 10, 3)))
+    
+    model = models.fantasized(
+      const(samples),
+      const(const(preds)),
+      jnp.empty((10, 3))
     )
 
-    mvn_a, mvn_b = model(index_points)
+    result = model(jnp.empty((10,)))
 
-    self.assertEqual(mvn_a.mean.shape, (10,))
-    self.assertEqual(mvn_a.cov.shape, (10, 10))
-    self.assertEqual(mvn_b.mean.shape, (10,))
-    self.assertEqual(mvn_b.cov.shape, (10, 10))
+    np.testing.assert_allclose(result.loc, preds.loc, atol=1e-4)
+    np.testing.assert_allclose(result.scale, preds.scale, atol=1e-4)
 
 
 if __name__ == '__main__':

From a4c62bfe83ba035c8272de30579715b03f0085ea Mon Sep 17 00:00:00 2001
From: Lando-L <hello.lando@icloud.com>
Date: Wed, 10 Apr 2024 09:36:09 +0100
Subject: [PATCH 3/9] feat: refactor tests to use chex assertions

---
 tests/core/samplers_test.py                   |  5 +-
 .../acquisitions/acquisitions_test.py         | 48 +++++++++----------
 .../acquisitions/constraints_test.py          |  6 +--
 .../optimizers/initializers_test.py           |  5 +-
 .../optimizers/optimizers_test.py             |  9 ++--
 tests/optimization/optimizers/solvers_test.py |  5 +-
 tests/prediction/models/kernels_test.py       | 30 ++++++------
 tests/prediction/models/likelihoods_test.py   |  6 +--
 tests/prediction/models/means_test.py         |  8 ++--
 tests/prediction/models/models_test.py        | 36 +++++++-------
 .../prediction/objectives/objectives_test.py  |  6 +--
 11 files changed, 84 insertions(+), 80 deletions(-)

diff --git a/tests/core/samplers_test.py b/tests/core/samplers_test.py
index b31dddf..701776d 100644
--- a/tests/core/samplers_test.py
+++ b/tests/core/samplers_test.py
@@ -1,4 +1,5 @@
 from absl.testing import absltest, parameterized
+from chex import assert_shape
 from jax import random
 
 from boax.core import distributions, samplers
@@ -14,7 +15,7 @@ def test_halton_normal(self):
 
     result = samplers.halton_normal(normal)(key3, 5)
 
-    self.assertEqual(result.shape, (5, 10))
+    assert_shape(result, (5, 10))
 
   def test_halton_uniform(self):
     key1, key2, key3 = random.split(random.key(0), 3)
@@ -25,7 +26,7 @@ def test_halton_uniform(self):
 
     result = samplers.halton_uniform(uniform)(key3, 5)
 
-    self.assertEqual(result.shape, (5, 10))
+    assert_shape(result, (5, 10))
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/acquisitions/acquisitions_test.py b/tests/optimization/acquisitions/acquisitions_test.py
index 1a1fe6b..86de411 100644
--- a/tests/optimization/acquisitions/acquisitions_test.py
+++ b/tests/optimization/acquisitions/acquisitions_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_shape, assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -17,12 +17,12 @@ def test_probability_of_improvement(self):
     loc, scale = random.uniform(key, (2, n, q))
     preds = distributions.normal.normal(loc, scale)
 
-    pi = acquisitions.probability_of_improvement(best)(preds)
-    lpi = acquisitions.log_probability_of_improvement(best)(preds)
+    poi = acquisitions.probability_of_improvement(best)(preds)
+    lpoi = acquisitions.log_probability_of_improvement(best)(preds)
 
-    self.assertEqual(pi.shape, (n,))
-    self.assertEqual(lpi.shape, (n,))
-    np.testing.assert_allclose(jnp.log(pi), lpi, atol=1e-4)
+    assert_shape(poi, (n,))
+    assert_shape(lpoi, (n,))
+    assert_trees_all_close(jnp.log(poi), lpoi, atol=1e-4)
 
   def test_expected_improvement(self):
     key = random.key(0)
@@ -35,9 +35,9 @@ def test_expected_improvement(self):
     ei = acquisitions.expected_improvement(best)(preds)
     lei = acquisitions.log_expected_improvement(best)(preds)
 
-    self.assertEqual(ei.shape, (n,))
-    self.assertEqual(lei.shape, (n,))
-    np.testing.assert_allclose(jnp.log(ei), lei, atol=1e-4)
+    assert_shape(ei, (n,))
+    assert_shape(lei, (n,))
+    assert_trees_all_close(jnp.log(ei), lei, atol=1e-4)
 
   def test_upper_confidence_bound(self):
     key = random.key(0)
@@ -50,8 +50,8 @@ def test_upper_confidence_bound(self):
     ucb = acquisitions.upper_confidence_bound(beta)(preds)
     expected = jnp.squeeze(loc + jnp.sqrt(beta) * scale)
 
-    self.assertEqual(ucb.shape, (n,))
-    np.testing.assert_allclose(ucb, expected, atol=1e-4)
+    assert_shape(ucb, (n,))
+    assert_trees_all_close(ucb, expected, atol=1e-4)
 
   def test_posterior(self):
     key = random.key(0)
@@ -63,11 +63,10 @@ def test_posterior(self):
     posterior_mean = acquisitions.posterior_mean()(preds)
     posterior_scale = acquisitions.posterior_scale()(preds)
 
-    self.assertEqual(posterior_mean.shape, (n,))
-    self.assertEqual(posterior_scale.shape, (n,))
-
-    np.testing.assert_allclose(posterior_mean, jnp.squeeze(loc), atol=1e-4)
-    np.testing.assert_allclose(posterior_scale, jnp.squeeze(scale), atol=1e-4)
+    assert_shape(posterior_mean, (n,))
+    assert_shape(posterior_scale, (n,))
+    assert_trees_all_close(posterior_mean, jnp.squeeze(loc), atol=1e-4)
+    assert_trees_all_close(posterior_scale, jnp.squeeze(scale), atol=1e-4)
 
   def test_q_probability_of_improvement(self):
     key = random.key(0)
@@ -78,7 +77,7 @@ def test_q_probability_of_improvement(self):
 
     qpoi = acquisitions.q_probability_of_improvement(best)(preds)
 
-    self.assertEqual(qpoi.shape, (n,))
+    assert_shape(qpoi, (n,))
 
   def test_q_expected_improvement(self):
     key = random.key(0)
@@ -89,7 +88,7 @@ def test_q_expected_improvement(self):
 
     qei = acquisitions.q_expected_improvement(best)(preds)
 
-    self.assertEqual(qei.shape, (n,))
+    assert_shape(qei, (n,))
 
   def test_q_upper_confidence_bound(self):
     key = random.key(0)
@@ -100,7 +99,7 @@ def test_q_upper_confidence_bound(self):
 
     qucb = acquisitions.q_upper_confidence_bound(beta)(preds)
 
-    self.assertEqual(qucb.shape, (n,))
+    assert_shape(qucb, (n,))
 
   def test_q_knowledge_gradient(self):
     key = random.key(0)
@@ -112,7 +111,7 @@ def test_q_knowledge_gradient(self):
 
     qkg = acquisitions.q_knowledge_gradient(best)(preds)
 
-    self.assertEqual(qkg.shape, (n,))
+    assert_shape(qkg, (n,))
   
   def test_q_multi_fidelity_knowledge_gradient(self):
     key1, key2 = random.split(random.key(0))
@@ -126,7 +125,7 @@ def test_q_multi_fidelity_knowledge_gradient(self):
 
     qmfkg = acquisitions.q_multi_fidelity_knowledge_gradient(best, cost_fn)((preds, costs))
 
-    self.assertEqual(qmfkg.shape, (n,))
+    assert_shape(qmfkg, (n,))
 
   def test_constrained(self):
     key = random.key(0)
@@ -148,9 +147,10 @@ def test_constrained(self):
       constraints.log_less_or_equal(1.0),
     )(model)
 
-    self.assertEqual(cei.shape, (n,))
-    self.assertEqual(clei.shape, (n,))
-    np.testing.assert_allclose(jnp.log(cei), clei, atol=1e-4)
+
+    assert_shape(cei, (n,))
+    assert_shape(clei, (n,))
+    assert_trees_all_close(jnp.log(cei), clei, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/acquisitions/constraints_test.py b/tests/optimization/acquisitions/constraints_test.py
index 48873ff..f7736f7 100644
--- a/tests/optimization/acquisitions/constraints_test.py
+++ b/tests/optimization/acquisitions/constraints_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -24,8 +24,8 @@ def test_range(self):
     ge = constraints.greater_or_equal(upper)(preds)
     lge = constraints.log_greater_or_equal(upper)(preds)
 
-    np.testing.assert_allclose(jnp.log(le), lle, atol=1e-4)
-    np.testing.assert_allclose(jnp.log(ge), lge, atol=1e-4)
+    assert_trees_all_close(jnp.log(le), lle, atol=1e-4)
+    assert_trees_all_close(jnp.log(ge), lge, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/optimizers/initializers_test.py b/tests/optimization/optimizers/initializers_test.py
index 69be209..a3576b1 100644
--- a/tests/optimization/optimizers/initializers_test.py
+++ b/tests/optimization/optimizers/initializers_test.py
@@ -1,4 +1,5 @@
 from absl.testing import absltest, parameterized
+from chex import assert_shape
 from jax import random
 
 from boax.optimization import optimizers
@@ -18,7 +19,7 @@ def test_q_batch(self):
       num_restarts,
     )
 
-    self.assertEqual(result.shape, (10, 1))
+    assert_shape(result, (10, 1))
 
   def test_q_nonnegative(self):
     key1, key2 = random.split(random.key(0))
@@ -33,7 +34,7 @@ def test_q_nonnegative(self):
       num_restarts,
     )
 
-    self.assertEqual(result.shape, (10, 1))
+    assert_shape(result, (10, 1))
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/optimizers/optimizers_test.py b/tests/optimization/optimizers/optimizers_test.py
index 0c60ecf..9f56c45 100644
--- a/tests/optimization/optimizers/optimizers_test.py
+++ b/tests/optimization/optimizers/optimizers_test.py
@@ -1,6 +1,7 @@
 from operator import itemgetter
 
 from absl.testing import absltest, parameterized
+from chex import assert_shape
 from jax import numpy as jnp
 from jax import random
 
@@ -27,8 +28,8 @@ def test_batch(self):
       num_restarts,
     )
 
-    self.assertEqual(next_x.shape, (q, d))
-    self.assertEqual(next_a.shape, ())
+    assert_shape(next_x, (q, d))
+    assert_shape(next_a, ())
 
   def test_sequential(self):
     key = random.key(0)
@@ -49,8 +50,8 @@ def test_sequential(self):
       num_restarts,
     )
 
-    self.assertEqual(next_x.shape, (q, d))
-    self.assertEqual(next_a.shape, (q,))
+    assert_shape(next_x, (q, d))
+    assert_shape(next_a, (q,))
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/optimizers/solvers_test.py b/tests/optimization/optimizers/solvers_test.py
index caf71c0..c23ac9a 100644
--- a/tests/optimization/optimizers/solvers_test.py
+++ b/tests/optimization/optimizers/solvers_test.py
@@ -1,6 +1,7 @@
 from operator import itemgetter
 
 from absl.testing import absltest, parameterized
+from chex import assert_shape, assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -24,8 +25,8 @@ def test_scipy(self):
       candidates,
     )
 
-    self.assertEqual(next_candidates.shape, (n, q, d))
-    self.assertEqual(values.shape, (n,))
+    assert_shape(next_candidates, (n, q, d))
+    assert_shape(values, (n,))
 
 
 if __name__ == '__main__':
diff --git a/tests/prediction/models/kernels_test.py b/tests/prediction/models/kernels_test.py
index 76381a5..0cc1d00 100644
--- a/tests/prediction/models/kernels_test.py
+++ b/tests/prediction/models/kernels_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_shape, assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -17,7 +17,7 @@ def test_rbf_function(self):
     result = functions.rbf.rbf(x, y, length_scale)
     expected = jnp.exp(-jnp.linalg.norm((x - y) ** 2) / (2 * length_scale**2))
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_rbf_kernel(self):
     length_scale = jnp.array([0.2, 0.5])
@@ -26,7 +26,7 @@ def test_rbf_kernel(self):
 
     result = kernels.rbf(length_scale)(x, y)
 
-    self.assertEqual(result.shape, (10, 10))
+    assert_shape(result, (10, 10))
 
   def test_matern_one_half_function(self):
     key = random.key(0)
@@ -37,7 +37,7 @@ def test_matern_one_half_function(self):
     result = functions.matern.one_half(x, y, length_scale)
     expected = jnp.exp(-jnp.linalg.norm(x - y) / length_scale)
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_matern_one_half_kernel(self):
     length_scale = jnp.array([0.2, 0.5])
@@ -46,7 +46,7 @@ def test_matern_one_half_kernel(self):
 
     result = kernels.matern_one_half(length_scale)(x, y)
 
-    self.assertEqual(result.shape, (10, 10))
+    assert_shape(result, (10, 10))
 
   def test_matern_three_halves_function(self):
     key = random.key(0)
@@ -58,7 +58,7 @@ def test_matern_three_halves_function(self):
     result = functions.matern.three_halves(x, y, length_scale)
     expected = (1 + z) * jnp.exp(-z)
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_matern_three_halves_kernel(self):
     length_scale = jnp.array([0.2, 0.5])
@@ -67,7 +67,7 @@ def test_matern_three_halves_kernel(self):
 
     result = kernels.matern_three_halves(length_scale)(x, y)
 
-    self.assertEqual(result.shape, (10, 10))
+    assert_shape(result, (10, 10))
 
   def test_matern_five_halves_function(self):
     key = random.key(0)
@@ -79,7 +79,7 @@ def test_matern_five_halves_function(self):
     result = functions.matern.five_halves(x, y, length_scale)
     expected = (1 + z + z**2 / 3) * jnp.exp(-z)
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_matern_three_halves_kernel(self):
     length_scale = jnp.array([0.2, 0.5])
@@ -88,7 +88,7 @@ def test_matern_three_halves_kernel(self):
 
     result = kernels.matern_five_halves(length_scale)(x, y)
 
-    self.assertEqual(result.shape, (10, 10))
+    assert_shape(result, (10, 10))
 
   def test_periodic_function(self):
     key = random.key(0)
@@ -102,7 +102,7 @@ def test_periodic_function(self):
     result = functions.periodic.periodic(x, y, length_scale, variance, period)
     expected = variance * jnp.exp(-0.5 * jnp.sum(z, axis=0))
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_periodic_kernel(self):
     length_scale = jnp.array([0.2, 0.5])
@@ -113,7 +113,7 @@ def test_periodic_kernel(self):
 
     result = kernels.periodic(length_scale, variance, period)(x, y)
 
-    self.assertEqual(result.shape, (10, 10))
+    assert_shape(result, (10, 10))
 
   def test_scaled(self):
     key = random.key(0)
@@ -127,7 +127,7 @@ def test_scaled(self):
     result = kernels.scaled(inner, amplitude)(x, y)
     expected = amplitude * inner(x, y)
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_linear_truncated(self):
     key = random.key(0)
@@ -142,7 +142,7 @@ def test_linear_truncated(self):
     factor = (1 - x_fid) * (1 - y_fid.T) * (1 + x_fid * y_fid.T)
     expected = inner(x, y) + inner(x, y) * factor
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_additive(self):
     key = random.key(0)
@@ -154,7 +154,7 @@ def test_additive(self):
       kernels.rbf(0.2)(x, y) + kernels.rbf(0.3)(x, y) + kernels.rbf(0.4)(x, y)
     )
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_product(self):
     key = random.key(0)
@@ -166,7 +166,7 @@ def test_product(self):
       kernels.rbf(0.2)(x, y) * kernels.rbf(0.3)(x, y) * kernels.rbf(0.4)(x, y)
     )
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/prediction/models/likelihoods_test.py b/tests/prediction/models/likelihoods_test.py
index cc0678a..51e25c1 100644
--- a/tests/prediction/models/likelihoods_test.py
+++ b/tests/prediction/models/likelihoods_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -20,8 +20,8 @@ def test_gaussian(self):
       mean, cov + 1e-4 * jnp.identity(10)
     )
 
-    np.testing.assert_allclose(result.mean, expected.mean, atol=1e-4)
-    np.testing.assert_allclose(result.cov, expected.cov, atol=1e-4)
+    assert_trees_all_close(result.mean, expected.mean, atol=1e-4)
+    assert_trees_all_close(result.cov, expected.cov, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/prediction/models/means_test.py b/tests/prediction/models/means_test.py
index e051e3e..1bc9bd6 100644
--- a/tests/prediction/models/means_test.py
+++ b/tests/prediction/models/means_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -13,7 +13,7 @@ def test_zero(self):
     result = means.zero()(value)
     expected = jnp.zeros(())
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_constant(self):
     x = jnp.array(2.0)
@@ -22,7 +22,7 @@ def test_constant(self):
     result = means.constant(x)(value)
     expected = x
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
   def test_linear(self):
     scale = jnp.array(2.0)
@@ -33,7 +33,7 @@ def test_linear(self):
     result = means.linear(scale, bias)(value)
     expected = scale * value + bias
 
-    np.testing.assert_allclose(result, expected, atol=1e-4)
+    assert_trees_all_close(result, expected, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/prediction/models/models_test.py b/tests/prediction/models/models_test.py
index 3efe417..e50502f 100644
--- a/tests/prediction/models/models_test.py
+++ b/tests/prediction/models/models_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_shape, assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -23,8 +23,8 @@ def test_gaussian_process(self):
 
     mean, cov = model(index_points)
 
-    self.assertEqual(mean.shape, (10,))
-    self.assertEqual(cov.shape, (10, 10))
+    assert_shape(mean, (10,))
+    assert_shape(cov, (10, 10))
 
     observation_index_points = random.uniform(
       key2, shape=(5, 1), minval=-1, maxval=1
@@ -43,8 +43,8 @@ def test_gaussian_process(self):
 
     mean, cov = model(index_points)
 
-    self.assertEqual(mean.shape, (10,))
-    self.assertEqual(cov.shape, (10, 10))
+    assert_shape(mean, (10,))
+    assert_shape(cov, (10, 10))
 
   def test_multi_fidelity(self):
     key1, key2 = random.split(random.key(0))
@@ -63,8 +63,8 @@ def test_multi_fidelity(self):
 
     mean, cov = model(index_points)
 
-    self.assertEqual(mean.shape, (10,))
-    self.assertEqual(cov.shape, (10, 10))
+    assert_shape(mean, (10,))
+    assert_shape(cov, (10, 10))
 
     observation_index_points = random.uniform(
       key2, shape=(5, 2), minval=-1, maxval=1
@@ -87,8 +87,8 @@ def test_multi_fidelity(self):
 
     mean, cov = model(index_points)
 
-    self.assertEqual(mean.shape, (10,))
-    self.assertEqual(cov.shape, (10, 10))
+    assert_shape(mean, (10,))
+    assert_shape(cov, (10, 10))
 
   def test_scaled(self):
     key1, key2 = random.split(random.key(0))
@@ -111,8 +111,8 @@ def test_scaled(self):
       scaled[1],
     )
 
-    np.testing.assert_allclose(result.loc, expected.loc, atol=1e-4)
-    np.testing.assert_allclose(result.scale, expected.scale, atol=1e-4)
+    assert_trees_all_close(result.loc, expected.loc, atol=1e-4)
+    assert_trees_all_close(result.scale, expected.scale, atol=1e-4)
 
   def test_sampled(self):
     key1, key2 = random.split(random.key(0))
@@ -128,7 +128,7 @@ def test_sampled(self):
 
     result = model(jnp.empty((10,)))
 
-    np.testing.assert_allclose(result, samples, atol=1e-4)
+    assert_trees_all_close(result, samples, atol=1e-4)
 
   def test_joined(self):
     key = random.key(0)
@@ -141,10 +141,10 @@ def test_joined(self):
 
     result1, result2 = model(jnp.empty((10,)))
 
-    np.testing.assert_allclose(result1.loc, preds1.loc, atol=1e-4)
-    np.testing.assert_allclose(result1.scale, preds1.scale, atol=1e-4)
-    np.testing.assert_allclose(result2.loc, preds2.loc, atol=1e-4)
-    np.testing.assert_allclose(result2.scale, preds2.scale, atol=1e-4)
+    assert_trees_all_close(result1.loc, preds1.loc, atol=1e-4)
+    assert_trees_all_close(result1.scale, preds1.scale, atol=1e-4)
+    assert_trees_all_close(result2.loc, preds2.loc, atol=1e-4)
+    assert_trees_all_close(result2.scale, preds2.scale, atol=1e-4)
   
   def test_fantazised(self):
     key1, key2 = random.split(random.key(0))
@@ -160,8 +160,8 @@ def test_fantazised(self):
 
     result = model(jnp.empty((10,)))
 
-    np.testing.assert_allclose(result.loc, preds.loc, atol=1e-4)
-    np.testing.assert_allclose(result.scale, preds.scale, atol=1e-4)
+    assert_trees_all_close(result.loc, preds.loc, atol=1e-4)
+    assert_trees_all_close(result.scale, preds.scale, atol=1e-4)
 
 
 if __name__ == '__main__':
diff --git a/tests/prediction/objectives/objectives_test.py b/tests/prediction/objectives/objectives_test.py
index c7e79bc..df9e790 100644
--- a/tests/prediction/objectives/objectives_test.py
+++ b/tests/prediction/objectives/objectives_test.py
@@ -1,5 +1,5 @@
-import numpy as np
 from absl.testing import absltest, parameterized
+from chex import assert_trees_all_close
 from jax import numpy as jnp
 from jax import random
 
@@ -28,7 +28,7 @@ def test_negative_log_likelihood(self):
       targets,
     )
 
-    np.testing.assert_allclose(result, -expected, atol=1e-4)
+    assert_trees_all_close(result, -expected, atol=1e-4)
 
   def test_penalized(self):
     key1, key2, key3, key4 = random.split(random.key(0), 4)
@@ -55,7 +55,7 @@ def test_penalized(self):
       targets,
     )
 
-    np.testing.assert_allclose(result, -expected + penalization, atol=1e-4)
+    assert_trees_all_close(result, -expected + penalization, atol=1e-4)
 
 
 if __name__ == '__main__':

From 50b0e9c282547b65b5ce524a5c54d8e4a9fe7e2c Mon Sep 17 00:00:00 2001
From: Lando Loeper <lando.loeper@deutschebahn.com>
Date: Sun, 21 Apr 2024 16:07:19 +0200
Subject: [PATCH 4/9] feat: add iid samplers

---
 boax/core/distributions/gamma.py             |  2 +-
 boax/core/distributions/poisson.py           |  2 +-
 boax/core/samplers/__init__.py               |  2 +
 boax/core/samplers/alias.py                  | 58 ++++++++++++++++++--
 boax/core/samplers/base.py                   |  6 +-
 boax/core/samplers/functions/__init__.py     |  1 +
 boax/core/samplers/functions/iid.py          | 29 ++++++++++
 boax/core/samplers/functions/quasi_random.py | 10 +++-
 tests/core/samplers_test.py                  | 30 ++++++++--
 9 files changed, 124 insertions(+), 16 deletions(-)
 create mode 100644 boax/core/samplers/functions/iid.py

diff --git a/boax/core/distributions/gamma.py b/boax/core/distributions/gamma.py
index 66ba508..770024e 100644
--- a/boax/core/distributions/gamma.py
+++ b/boax/core/distributions/gamma.py
@@ -37,7 +37,7 @@ class Gamma(NamedTuple):
 
 def gamma(a: Array, b: Array = jnp.ones(())) -> Gamma:
   """
-  Smart constructor for the beta distribution.
+  Smart constructor for the gamma distribution.
 
   Args:
     a: The shape parameter.
diff --git a/boax/core/distributions/poisson.py b/boax/core/distributions/poisson.py
index 0197565..f4da7b4 100644
--- a/boax/core/distributions/poisson.py
+++ b/boax/core/distributions/poisson.py
@@ -34,7 +34,7 @@ class Poisson(NamedTuple):
 
 def poisson(mu: Array) -> Poisson:
   """
-  Smart constructor for the beta distribution.
+  Smart constructor for the poisson distribution.
 
   Args:
     mu: The rate parameter.
diff --git a/boax/core/samplers/__init__.py b/boax/core/samplers/__init__.py
index 9778448..499384a 100644
--- a/boax/core/samplers/__init__.py
+++ b/boax/core/samplers/__init__.py
@@ -14,6 +14,8 @@
 
 """The samplers sub-package."""
 
+from .alias import normal as normal
+from .alias import uniform as uniform
 from .alias import halton_normal as halton_normal
 from .alias import halton_uniform as halton_uniform
 from .base import Sampler as Sampler
diff --git a/boax/core/samplers/alias.py b/boax/core/samplers/alias.py
index 20c3a28..7bb1f31 100644
--- a/boax/core/samplers/alias.py
+++ b/boax/core/samplers/alias.py
@@ -16,7 +16,7 @@
 
 from functools import partial
 
-from jax import lax
+from jax import lax, random
 from jax import numpy as jnp
 
 from boax.core import distributions
@@ -27,6 +27,56 @@
 from boax.utils.functools import compose
 
 
+def uniform(
+  uniform: Uniform = Uniform(jnp.zeros((1,)), jnp.ones((1,))),
+) -> Sampler:
+  """
+  The i.i.d. uniform sampler.
+
+  Example:
+    >>> sampler = uniform()
+    >>> base_samples =  sampler(key, (128,))
+
+  Args:
+    uniform: The base uniform distribution.
+
+  Returns:
+    The corresponding `Sampler`.
+  """
+  
+  out_shape = lax.broadcast_shapes(uniform.a.shape, uniform.b.shape)
+
+  return compose(
+    partial(partial, distributions.uniform.sample)(uniform),
+    partial(functions.iid.uniform, ndims=out_shape[0])
+  )
+
+
+def normal(
+  normal: Normal = Normal(jnp.zeros((1,)), jnp.ones((1,))),
+) -> Sampler:
+  """
+  The i.i.d. normal sampler.
+
+  Example:
+    >>> sampler = normal()
+    >>> base_samples = sampler(key, (128,))
+
+  Args:
+    normal: The base normal distribution.
+
+  Returns:
+    The corresponding `Sampler`.
+  """
+
+  out_shape = lax.broadcast_shapes(normal.loc.shape, normal.scale.shape)
+
+  return compose(
+    partial(partial, distributions.normal.sample)(normal),
+    partial(functions.iid.normal, ndims=out_shape[0])
+  )
+
+
 def halton_uniform(
   uniform: Uniform = Uniform(jnp.zeros((1,)), jnp.ones((1,))),
 ) -> Sampler:
@@ -34,8 +84,8 @@ def halton_uniform(
   The quasi-MC uniform sampler based on halton sequences.
 
   Example:
-    >>> sampler = halton_uniform(uniform)
-    >>> base_samples = sampler(key, 128)
+    >>> sampler = halton_uniform()
+    >>> base_samples = sampler(key, (128,))
 
   Args:
     uniform: The base uniform distribution.
@@ -66,7 +116,7 @@ def halton_normal(
   The quasi-MC normal sampler based on halton sequences.
 
   Example:
-    >>> sampler = halton_normal(normal)
+    >>> sampler = halton_normal()
     >>> base_samples = sampler(key, 128)
 
   Args:
diff --git a/boax/core/samplers/base.py b/boax/core/samplers/base.py
index 696457a..9e8c37f 100644
--- a/boax/core/samplers/base.py
+++ b/boax/core/samplers/base.py
@@ -14,7 +14,7 @@
 
 """Base interface for samplers."""
 
-from typing import Protocol
+from typing import Protocol, Sequence
 
 from boax.utils.typing import Array, PRNGKey
 
@@ -27,13 +27,13 @@ class Sampler(Protocol):
   and returns `num_results` samples.
   """
 
-  def __call__(self, key: PRNGKey, num_results: int) -> Array:
+  def __call__(self, key: PRNGKey, shape: Sequence[int]) -> Array:
     """
     Draws `num_results` of samples.
 
     Args:
       key: The pseudo-random number generator key.
-      candidates: The number of results to return.
+      shape: The sample shape.
 
     Returns:
       A set of `num_results` samples.
diff --git a/boax/core/samplers/functions/__init__.py b/boax/core/samplers/functions/__init__.py
index 328c4d9..769f6e0 100644
--- a/boax/core/samplers/functions/__init__.py
+++ b/boax/core/samplers/functions/__init__.py
@@ -14,5 +14,6 @@
 
 """The sampler functions sub-package."""
 
+from . import iid as iid
 from . import quasi_random as quasi_random
 from . import utils as utils
diff --git a/boax/core/samplers/functions/iid.py b/boax/core/samplers/functions/iid.py
new file mode 100644
index 0000000..3355b91
--- /dev/null
+++ b/boax/core/samplers/functions/iid.py
@@ -0,0 +1,29 @@
+# Copyright 2023 The Boax Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""IID sampling functions."""
+
+from typing import Sequence
+
+from jax import random
+
+from boax.utils.typing import PRNGKey, Array
+
+
+def uniform(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Array:
+  return random.uniform(key, sample_shape + (ndims,))
+
+
+def normal(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Array:
+  return random.normal(key, sample_shape + (ndims,))
diff --git a/boax/core/samplers/functions/quasi_random.py b/boax/core/samplers/functions/quasi_random.py
index 33eaef1..624557c 100644
--- a/boax/core/samplers/functions/quasi_random.py
+++ b/boax/core/samplers/functions/quasi_random.py
@@ -14,6 +14,8 @@
 
 """Quasi Random sampling functions."""
 
+from typing import Sequence
+
 from jax import numpy as jnp
 from jax import random
 
@@ -26,8 +28,9 @@
 assert len(PRIMES) == MAX_DIMENSION
 
 
-def halton_sequence(key: PRNGKey, num_samples: int, ndims: int) -> Array:
+def halton_sequence(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Array:
   shuffle_key, correction_key = random.split(key)
+  num_samples = jnp.prod(jnp.asarray(sample_shape))
 
   radixes = PRIMES[0:ndims][..., jnp.newaxis]
   indices = jnp.reshape(jnp.arange(num_samples) + 1, (-1, 1, 1))
@@ -48,8 +51,9 @@ def halton_sequence(key: PRNGKey, num_samples: int, ndims: int) -> Array:
   base_values = jnp.sum(shuffled / (radixes * weights), axis=-1)
   zero_correction = random.uniform(correction_key, (ndims, 1))
 
-  return (
-    base_values + (zero_correction / (radixes**max_sizes_by_axes)).flatten()
+  return jnp.reshape(
+    base_values + (zero_correction / (radixes**max_sizes_by_axes)).flatten(),
+    sample_shape + (ndims,)
   )
 
 
diff --git a/tests/core/samplers_test.py b/tests/core/samplers_test.py
index 701776d..7413e4e 100644
--- a/tests/core/samplers_test.py
+++ b/tests/core/samplers_test.py
@@ -6,6 +6,28 @@
 
 
 class SamplersTest(parameterized.TestCase):
+  def test_normal(self):
+    key1, key2, key3 = random.split(random.key(0), 3)
+
+    loc = random.uniform(key1, (10,))
+    scale = random.uniform(key2, (10,))
+    normal = distributions.normal.normal(loc, scale)
+
+    result = samplers.normal(normal)(key3, (2, 5, 3))
+
+    assert_shape(result, (2, 5, 3, 10))
+
+  def test_uniform(self):
+    key1, key2, key3 = random.split(random.key(0), 3)
+
+    a = random.uniform(key1, (10,))
+    b = a + random.uniform(key2, (10,))
+    uniform = distributions.uniform.uniform(a, b)
+
+    result = samplers.uniform(uniform)(key3, (2, 5, 3))
+
+    assert_shape(result, (2, 5, 3, 10))
+  
   def test_halton_normal(self):
     key1, key2, key3 = random.split(random.key(0), 3)
 
@@ -13,9 +35,9 @@ def test_halton_normal(self):
     scale = random.uniform(key2, (10,))
     normal = distributions.normal.normal(loc, scale)
 
-    result = samplers.halton_normal(normal)(key3, 5)
+    result = samplers.halton_normal(normal)(key3, (2, 5, 3))
 
-    assert_shape(result, (5, 10))
+    assert_shape(result, (2, 5, 3, 10))
 
   def test_halton_uniform(self):
     key1, key2, key3 = random.split(random.key(0), 3)
@@ -24,9 +46,9 @@ def test_halton_uniform(self):
     b = a + random.uniform(key2, (10,))
     uniform = distributions.uniform.uniform(a, b)
 
-    result = samplers.halton_uniform(uniform)(key3, 5)
+    result = samplers.halton_uniform(uniform)(key3, (2, 5, 3))
 
-    assert_shape(result, (5, 10))
+    assert_shape(result, (2, 5, 3, 10))
 
 
 if __name__ == '__main__':

From 084efc2e7b0533bc4e8ad29532a4637334a94d66 Mon Sep 17 00:00:00 2001
From: Lando Loeper <lando.loeper@deutschebahn.com>
Date: Sun, 21 Apr 2024 16:08:33 +0200
Subject: [PATCH 5/9] feat: refactor optimizers interface

---
 boax/optimization/optimizers/alias.py         | 29 +++---------
 boax/optimization/optimizers/base.py          | 15 +-----
 .../optimizers/initializers/alias.py          | 44 ++++++++++++++---
 .../optimizers/initializers/base.py           |  7 +--
 boax/optimization/optimizers/solvers/alias.py | 15 ++++--
 boax/optimization/optimizers/solvers/base.py  |  8 +---
 .../optimizers/initializers_test.py           | 47 ++++++++++---------
 .../optimizers/optimizers_test.py             | 43 ++++++-----------
 tests/optimization/optimizers/solvers_test.py | 15 +++---
 9 files changed, 104 insertions(+), 119 deletions(-)

diff --git a/boax/optimization/optimizers/alias.py b/boax/optimization/optimizers/alias.py
index f83e5e4..3091635 100644
--- a/boax/optimization/optimizers/alias.py
+++ b/boax/optimization/optimizers/alias.py
@@ -14,16 +14,12 @@
 
 """Alias for optimizers."""
 
-from functools import partial
-
 from jax import numpy as jnp
 from jax import random
 
-from boax.core import distributions, samplers
 from boax.optimization.optimizers.base import Optimizer
 from boax.optimization.optimizers.initializers.base import Initializer
 from boax.optimization.optimizers.solvers.base import Solver
-from boax.utils.functools import compose
 
 
 def batch(initializer: Initializer, solver: Solver) -> Optimizer:
@@ -32,7 +28,7 @@ def batch(initializer: Initializer, solver: Solver) -> Optimizer:
 
   Example:
     >>> optimizer = batch(initializer, solver)
-    >>> next_candidates = optimizer(key, fun, bounds, q, num_samples, num_restarts)
+    >>> next_candidates = optimizer(key)
 
   Args:
     initializer: The initializer function.
@@ -42,20 +38,9 @@ def batch(initializer: Initializer, solver: Solver) -> Optimizer:
     The batch `Optimizer`.
   """
 
-  def optimizer(key, fun, bounds, q, num_samples, num_restarts):
-    key1, key2 = random.split(key)
-
-    x = jnp.reshape(
-      samplers.halton_uniform(
-        distributions.uniform.uniform(bounds[:, 0], bounds[:, 1])
-      )(key1, num_samples * q),
-      (num_samples, q, -1),
-    )
-    y = fun(x)
-
-    candidates = initializer(key2, x, y, num_restarts)
-    next_candidates, values = solver(fun, bounds, candidates)
-
+  def optimizer(key):
+    candidates = initializer(key)
+    next_candidates, values = solver(candidates)
     idx = jnp.argmax(values)
 
     return next_candidates[idx], values[idx]
@@ -63,7 +48,7 @@ def optimizer(key, fun, bounds, q, num_samples, num_restarts):
   return optimizer
 
 
-def sequential(initializer: Initializer, solver: Solver) -> Optimizer:
+def sequential(initializer: Initializer, solver: Solver, q: int) -> Optimizer:
   """
   Sequential optimizer.
 
@@ -81,9 +66,9 @@ def sequential(initializer: Initializer, solver: Solver) -> Optimizer:
 
   inner = batch(initializer, solver)
 
-  def optimizer(key, fun, bounds, q, num_samples, num_restarts):
+  def optimizer(key):
     next_candidates, values = zip(*(
-      inner(random.fold_in(key, i), fun, bounds, 1, num_samples, num_restarts)
+      inner(random.fold_in(key, i))
       for i in range(q)
     ))
 
diff --git a/boax/optimization/optimizers/base.py b/boax/optimization/optimizers/base.py
index e5bca01..37bb568 100644
--- a/boax/optimization/optimizers/base.py
+++ b/boax/optimization/optimizers/base.py
@@ -24,25 +24,12 @@ class Optimizer(Protocol):
   A callable type for the optimization functions.
   """
 
-  def __call__(
-    self,
-    key: PRNGKey,
-    fun: Callable[[Array], Array],
-    bounds: Array,
-    q: int,
-    num_samples: int,
-    num_restarts: int,
-  ) -> Tuple[Array, Array]:
+  def __call__(self, key: PRNGKey) -> Tuple[Array, Array]:
     """
     The optimization function.
 
     Args:
       key: A PRNG key.
-      fun: The function to be optimized.
-      bounds: The bounds of the search space.
-      q: The batch size.
-      num_samples: The number of samples.
-      num_restarts: The number of restarts.
 
     Returns:
       A tuple of the maxima and their acquisition values.
diff --git a/boax/optimization/optimizers/initializers/alias.py b/boax/optimization/optimizers/initializers/alias.py
index 1016804..e38d2b0 100644
--- a/boax/optimization/optimizers/initializers/alias.py
+++ b/boax/optimization/optimizers/initializers/alias.py
@@ -14,33 +14,50 @@
 
 """Alias for initialization functions."""
 
+from typing import Callable
+
 from jax import lax, nn, random
 from jax import numpy as jnp
 
+from boax.core.samplers.base import Sampler
 from boax.optimization.optimizers.initializers.base import Initializer
-from boax.utils.typing import Numeric
+from boax.utils.typing import Array, Numeric
 
 
 def q_batch(
+  fun: Callable[[Array], Array],
+  sampler: Sampler,
+  q: int,
+  num_results: int,
+  num_restarts: int,
   eta: Numeric = 1.0,
 ) -> Initializer:
   """
   Q batch initializer.
 
   Example:
-    >>> initializer = q_batch(eta=2.0)
-    >>> candidates = initializer(key, x, y, num_restarts)
+    >>> initializer = q_batch(fun, sampler, num_restarts, num_restarts)
+    >>> candidates = initializer(key)
 
   Args:
+    fun: The scoring function.
+    sampler: The candidates sampler.
+    num_results: The number of results.
+    num_restarts: The number of restarts.
     eta: The temperature parameter.
 
   Returns:
     The q-batch `Initializer`.
   """
 
-  def initializer(key, x, y, num_restarts):
+  def initializer(key):
+    key1, key2 = random.split(key)
+    
+    x = sampler(key1, (num_results, q))
+    y = fun(x)
+    
     return random.choice(
-      key,
+      key2,
       x,
       (num_restarts,),
       p=jnp.exp(eta * nn.standardize(y, axis=0)),
@@ -50,6 +67,11 @@ def initializer(key, x, y, num_restarts):
 
 
 def q_batch_nonnegative(
+  fun: Callable[[Array], Array],
+  sampler: Sampler,
+  q: int,
+  num_results: int,
+  num_restarts: int,
   eta: Numeric = 1.0,
   alpha: Numeric = 1e-4,
 ) -> Initializer:
@@ -61,6 +83,10 @@ def q_batch_nonnegative(
     >>> candidates = initializer(key, x, y, num_restarts)
 
   Args:
+    fun: The scoring function.
+    sampler: The candidates sampler.
+    num_results: The number of results.
+    num_restarts: The number of restarts.
     eta: The temperature parameter.
     alpha: The alpha parameter.
 
@@ -68,7 +94,11 @@ def q_batch_nonnegative(
     The q-batch non-negative `Initializer`.
   """
 
-  def initializer(key, x, y, num_restarts):
+  def initializer(key):
+    key1, key2 = random.split(key)
+
+    x = sampler(key1, (num_results, q))
+    y = fun(x)
     max_val = jnp.max(y)
 
     def cond(x):
@@ -83,7 +113,7 @@ def body(x):
     _, alpha_pos = lax.while_loop(cond, body, (alpha, y >= alpha * max_val))
 
     return random.choice(
-      key,
+      key2,
       x[alpha_pos],
       (num_restarts,),
       p=jnp.exp(eta * (y[alpha_pos] / max_val - 1)),
diff --git a/boax/optimization/optimizers/initializers/base.py b/boax/optimization/optimizers/initializers/base.py
index 175c2cf..dd94cb5 100644
--- a/boax/optimization/optimizers/initializers/base.py
+++ b/boax/optimization/optimizers/initializers/base.py
@@ -24,17 +24,12 @@ class Initializer(Protocol):
   A callable type for the initialization step of an `Optimizer`.
   """
 
-  def __call__(
-    self, key: PRNGKey, x: Array, y: Array, num_restarts: int
-  ) -> Array:
+  def __call__(self, key: PRNGKey) -> Array:
     """
     The initialization function.
 
     Args:
       key: A PRNG key.
-      x: The initial index points.
-      y: The initial scores.
-      num_restarts: The number of restarts.
 
     Returns:
       The initial set of candidates.
diff --git a/boax/optimization/optimizers/solvers/alias.py b/boax/optimization/optimizers/solvers/alias.py
index 0591e98..ffc5d3b 100644
--- a/boax/optimization/optimizers/solvers/alias.py
+++ b/boax/optimization/optimizers/solvers/alias.py
@@ -15,37 +15,42 @@
 """Alias for solver functions."""
 
 from functools import partial
+from typing import Callable
 
 from jax import numpy as jnp
 from jax.scipy import optimize
 
 from boax.optimization.optimizers.solvers.base import Solver
 from boax.utils.functools import compose
+from boax.utils.typing import Array
 
 
 def scipy(
+  fun: Callable[[Array], Array],
+  bounds: Array,
   method: str = 'bfgs',
 ) -> Solver:
   """
   Scipy solver.
 
   Example:
-    >>> solver = scipy()
-    >>> next_candidates, values = solver(acqf, bounds, candidates)
+    >>> solver = scipy(fun, bounds)
+    >>> next_candidates, values = solver(candidates)
 
   Args:
+    bounds: The bounds of the search space.
     method: The solver method.
 
   Returns:
     The scipy `Solver`.
   """
 
-  def solver(fn, bounds, candidates):
+  def solver(candidates):
     results = optimize.minimize(
       fun=compose(
         jnp.negative,
         jnp.sum,
-        fn,
+        fun,
         partial(jnp.reshape, newshape=candidates.shape),
       ),
       x0=candidates.flatten(),
@@ -58,6 +63,6 @@ def solver(fn, bounds, candidates):
       a_max=bounds[:, 1],
     )
 
-    return clipped, fn(clipped)
+    return clipped, fun(clipped)
 
   return solver
diff --git a/boax/optimization/optimizers/solvers/base.py b/boax/optimization/optimizers/solvers/base.py
index 61535de..e655a30 100644
--- a/boax/optimization/optimizers/solvers/base.py
+++ b/boax/optimization/optimizers/solvers/base.py
@@ -14,7 +14,7 @@
 
 """Base interface for solvers."""
 
-from typing import Callable, Protocol, Tuple
+from typing import Protocol, Tuple
 
 from boax.utils.typing import Array
 
@@ -24,15 +24,11 @@ class Solver(Protocol):
   A callable type for the solving step of an `Optimizer`.
   """
 
-  def __call__(
-    self, fun: Callable, bounds: Array, candidates: Array
-  ) -> Tuple[Array, Array]:
+  def __call__(self, candidates: Array) -> Tuple[Array, Array]:
     """
     The solving function.
 
     Args:
-      fun: The function to be optimized.
-      bounds: The bounds of the search space.
       candidates: The initial guess.
 
     Returns:
diff --git a/tests/optimization/optimizers/initializers_test.py b/tests/optimization/optimizers/initializers_test.py
index a3576b1..d79853b 100644
--- a/tests/optimization/optimizers/initializers_test.py
+++ b/tests/optimization/optimizers/initializers_test.py
@@ -1,40 +1,43 @@
+from operator import itemgetter
+
 from absl.testing import absltest, parameterized
 from chex import assert_shape
 from jax import random
 
+from boax.core import samplers
 from boax.optimization import optimizers
 
 
 class InitializersTest(parameterized.TestCase):
   def test_q_batch(self):
-    key1, key2 = random.split(random.key(0))
-    x = random.uniform(key1, (100, 1))
-    y = x[..., 0]
-    num_restarts = 10
-
-    result = optimizers.initializers.q_batch()(
-      key2,
-      x,
-      y,
-      num_restarts,
+    key = random.key(0)
+
+    fun = itemgetter((..., 0, 0))
+    sampler = samplers.halton_uniform()
+    s, n, q, d = 10, 5, 3, 1
+
+    initializer = optimizers.initializers.q_batch(
+      fun, sampler, q, s, n,
     )
 
-    assert_shape(result, (10, 1))
+    result = initializer(key)
+
+    assert_shape(result, (n, q, d))
 
   def test_q_nonnegative(self):
-    key1, key2 = random.split(random.key(0))
-    x = random.uniform(key1, (100, 1))
-    y = x[..., 0]
-    num_restarts = 10
-
-    result = optimizers.initializers.q_batch_nonnegative()(
-      key2,
-      x,
-      y,
-      num_restarts,
+    key = random.key(0)
+
+    fun = itemgetter((..., 0, 0))
+    sampler = samplers.halton_uniform()
+    s, n, q, d = 10, 5, 3, 1
+
+    initializer = optimizers.initializers.q_batch_nonnegative(
+      fun, sampler, q, s, n,
     )
 
-    assert_shape(result, (10, 1))
+    result = initializer(key)
+
+    assert_shape(result, (n, q, d))
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/optimizers/optimizers_test.py b/tests/optimization/optimizers/optimizers_test.py
index 9f56c45..7a56b35 100644
--- a/tests/optimization/optimizers/optimizers_test.py
+++ b/tests/optimization/optimizers/optimizers_test.py
@@ -2,7 +2,6 @@
 
 from absl.testing import absltest, parameterized
 from chex import assert_shape
-from jax import numpy as jnp
 from jax import random
 
 from boax.optimization import optimizers
@@ -11,47 +10,33 @@
 class OptimizersTest(parameterized.TestCase):
   def test_batch(self):
     key = random.key(0)
-    num_samples, num_restarts, q, d = 100, 10, 3, 1
 
-    initializer = lambda k, x, _, n: random.choice(k, x, (n,))
-    solver = lambda fun, _, c: (c, fun(c))
+    fun = itemgetter((..., 0, 0))
+    n, q, d = 10, 3, 1
 
-    acqf = itemgetter((..., 0, 0))
-    bounds = jnp.array([[-1.0, 1.0]])
+    initializer = lambda k: random.uniform(k, (n, q, d))
+    solver = lambda c: (c, fun(c))
+    optimizer = optimizers.batch(initializer, solver)
 
-    next_x, next_a = optimizers.batch(initializer, solver)(
-      key,
-      acqf,
-      bounds,
-      q,
-      num_samples,
-      num_restarts,
-    )
+    next_x, next_v = optimizer(key)
 
     assert_shape(next_x, (q, d))
-    assert_shape(next_a, ())
+    assert_shape(next_v, ())
 
   def test_sequential(self):
     key = random.key(0)
-    num_samples, num_restarts, q, d = 100, 10, 3, 1
 
-    initializer = lambda k, x, _, n: random.choice(k, x, (n,))
-    solver = lambda fun, _, c: (c, fun(c))
+    fun = itemgetter((..., 0, 0))
+    n, q, d = 10, 3, 1
 
-    acqf = itemgetter((..., 0, 0))
-    bounds = jnp.array([[-1.0, 1.0]])
+    initializer = lambda k: random.uniform(k, (n, 1, d))
+    solver = lambda c: (c, fun(c))
+    optimizer = optimizers.sequential(initializer, solver, q)
 
-    next_x, next_a = optimizers.sequential(initializer, solver)(
-      key,
-      acqf,
-      bounds,
-      q,
-      num_samples,
-      num_restarts,
-    )
+    next_x, next_v = optimizer(key)
 
     assert_shape(next_x, (q, d))
-    assert_shape(next_a, (q,))
+    assert_shape(next_v, (q,))
 
 
 if __name__ == '__main__':
diff --git a/tests/optimization/optimizers/solvers_test.py b/tests/optimization/optimizers/solvers_test.py
index c23ac9a..ed0789f 100644
--- a/tests/optimization/optimizers/solvers_test.py
+++ b/tests/optimization/optimizers/solvers_test.py
@@ -1,7 +1,7 @@
 from operator import itemgetter
 
 from absl.testing import absltest, parameterized
-from chex import assert_shape, assert_trees_all_close
+from chex import assert_shape
 from jax import numpy as jnp
 from jax import random
 
@@ -11,19 +11,18 @@
 class SolversTest(parameterized.TestCase):
   def test_scipy(self):
     key = random.key(0)
-    n, q, d = 10, 3, 1
 
-    acqf = itemgetter((..., 0, 0))
+    fun = itemgetter((..., 0, 0))
     bounds = jnp.array([[-1.0, 1.0]])
+    n, q, d = 5, 3, 1
+
     candidates = random.uniform(
       key, minval=bounds[:, 0], maxval=bounds[:, 1], shape=(n, q, d)
     )
 
-    next_candidates, values = optimizers.solvers.scipy()(
-      acqf,
-      bounds,
-      candidates,
-    )
+    solver = optimizers.solvers.scipy(fun, bounds)
+
+    next_candidates, values = solver(candidates)
 
     assert_shape(next_candidates, (n, q, d))
     assert_shape(values, (n,))

From accf11537e73b2872026b357bda7af06ade77a20 Mon Sep 17 00:00:00 2001
From: Lando-L <hello.lando@icloud.com>
Date: Wed, 10 Apr 2024 10:53:11 +0100
Subject: [PATCH 6/9] feat: renamed model alias for gps

---
 boax/prediction/models/__init__.py            |  4 +-
 boax/prediction/models/base.py                |  2 +-
 .../models/functions/multi_fidelity.py        | 32 ++++----
 .../models/{alias.py => gaussian_process.py}  | 69 +++++++++++++++--
 boax/prediction/models/transformed.py         | 69 +++++------------
 ...odels_test.py => gaussian_process_test.py} | 74 ++++++++-----------
 6 files changed, 132 insertions(+), 118 deletions(-)
 rename boax/prediction/models/{alias.py => gaussian_process.py} (71%)
 rename tests/prediction/models/{models_test.py => gaussian_process_test.py} (79%)

diff --git a/boax/prediction/models/__init__.py b/boax/prediction/models/__init__.py
index c09bd65..b0b0163 100644
--- a/boax/prediction/models/__init__.py
+++ b/boax/prediction/models/__init__.py
@@ -17,12 +17,10 @@
 from . import kernels as kernels
 from . import likelihoods as likelihoods
 from . import means as means
-from .alias import gaussian_process as gaussian_process
-from .alias import multi_fidelity as multi_fidelity
+from . import gaussian_process as gaussian_process
 from .base import Model as Model
 from .transformed import input_transformed as input_transformed
 from .transformed import joined as joined
 from .transformed import outcome_transformed as outcome_transformed
 from .transformed import sampled as sampled
 from .transformed import scaled as scaled
-from .transformed import fantasized as fantasized
diff --git a/boax/prediction/models/base.py b/boax/prediction/models/base.py
index bb29d0b..ecff83b 100644
--- a/boax/prediction/models/base.py
+++ b/boax/prediction/models/base.py
@@ -29,7 +29,7 @@ class Model(Protocol, Generic[T]):
   and returns a posterior prediction of type `T`.
   """
 
-  def __call__(self, index_points: Array) -> T:
+  def __call__(self, index_points: Array, **kwargs) -> T:
     """
     Computes the posterior prediction at the index points.
 
diff --git a/boax/prediction/models/functions/multi_fidelity.py b/boax/prediction/models/functions/multi_fidelity.py
index 8aa2a7f..72ceba5 100644
--- a/boax/prediction/models/functions/multi_fidelity.py
+++ b/boax/prediction/models/functions/multi_fidelity.py
@@ -26,20 +26,15 @@
 from boax.utils.typing import Array, Numeric
 
 
-def split(values: Array) -> Tuple[Array, Array]:
-  return jnp.split(values, [values.shape[-1] - 1], axis=-1)
-
-
 def prior(
   index_points: Array,
+  fidelities: Array,
   mean_fn: Mean,
   kernel_fn: Callable[[Array, Array], Kernel],
   jitter: Numeric,
 ) -> MultivariateNormal:
-  values, fidelities = split(index_points)
-
-  Kxx = kernel_fn(fidelities, fidelities)(values, values)
-  mean = mean_fn(values)
+  Kxx = kernel_fn(fidelities, fidelities)(index_points, index_points)
+  mean = mean_fn(index_points)
   cov = Kxx + jitter * jnp.identity(Kxx.shape[-1])
 
   return distributions.multivariate_normal.multivariate_normal(mean, cov)
@@ -47,21 +42,28 @@ def prior(
 
 def posterior(
   index_points: Array,
+  fidelities: Array,
   observation_index_points: Array,
+  observation_fidelities: Array,
   observations: Array,
   mean_fn: Mean,
   kernel_fn: Callable[[Array, Array], Kernel],
   jitter: Numeric,
 ) -> MultivariateNormal:
-  ivalues, ifidelities = split(index_points)
-  ovalues, ofidelities = split(observation_index_points)
+  mz = mean_fn(index_points)
+  mx = mean_fn(observation_index_points)
 
-  mz = mean_fn(ivalues)
-  mx = mean_fn(ovalues)
+  Kxx = kernel_fn(observation_fidelities, observation_fidelities)(
+    observation_index_points, observation_index_points
+  )
+  
+  Kxz = kernel_fn(observation_fidelities, fidelities)(
+    observation_index_points, index_points
+  )
 
-  Kxx = kernel_fn(ofidelities, ofidelities)(ovalues, ovalues)
-  Kxz = kernel_fn(ofidelities, ifidelities)(ovalues, ivalues)
-  Kzz = kernel_fn(ifidelities, ifidelities)(ivalues, ivalues)
+  Kzz = kernel_fn(fidelities, fidelities)(
+    index_points, index_points
+  )
 
   K = Kxx + jitter * jnp.identity(Kxx.shape[-1])
   chol = scipy.linalg.cholesky(K, lower=True)
diff --git a/boax/prediction/models/alias.py b/boax/prediction/models/gaussian_process.py
similarity index 71%
rename from boax/prediction/models/alias.py
rename to boax/prediction/models/gaussian_process.py
index cfe08a1..43348e7 100644
--- a/boax/prediction/models/alias.py
+++ b/boax/prediction/models/gaussian_process.py
@@ -12,12 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-"""Alias for surrogate models."""
+"""Gaussian for surrogate models."""
 
 from functools import partial
 from typing import Callable, TypeVar
 
-from jax import jit
+from jax import jit, vmap
+from jax import numpy as jnp
 
 from boax.core.distributions.multivariate_normal import MultivariateNormal
 from boax.prediction.models import functions
@@ -31,7 +32,7 @@
 T = TypeVar('T')
 
 
-def gaussian_process(
+def exact(
   mean_fn: Mean,
   kernel_fn: Kernel,
   likelihood_fn: Likelihood[MultivariateNormal, T],
@@ -40,7 +41,7 @@ def gaussian_process(
   jitter: Numeric = 1e-6,
 ) -> Model[T]:
   """
-  The gaussian process model.
+  The exact gaussian process model.
 
   Example:
     >>> model = gaussian_process(mean_fn, kernel_fn)
@@ -84,6 +85,26 @@ def gaussian_process(
         ),
       )
     )
+  
+
+def fantasy(
+  mean_fn: Mean,
+  kernel_fn: Kernel,
+  jitter: Numeric = 1e-6,
+) -> Callable[[Array, Array, Array], MultivariateNormal]:
+  return vmap(
+    vmap(
+      jit(
+        partial(
+          functions.gaussian.posterior,
+          mean_fn=mean_fn,
+          kernel_fn=kernel_fn,
+          jitter=jitter,
+        )
+      ),
+      in_axes=(0, None, 0)
+    )
+  )
 
 
 def multi_fidelity(
@@ -91,6 +112,7 @@ def multi_fidelity(
   kernel_fn: Callable[[Array, Array], Kernel],
   likelihood_fn: Likelihood[MultivariateNormal, T],
   observation_index_points: Array | None = None,
+  observation_fidelities: Array | None = None,
   observations: Array | None = None,
   jitter: Numeric = 1e-6,
 ) -> Model[T]:
@@ -105,6 +127,7 @@ def multi_fidelity(
     mean_fn: The process' mean function.
     kernel_fn: The process' covariance function.
     observation_index_points: The index points of the given observations.
+    observation_fidelities: The fidelities of the given observatons.
     observations: The observed values.
     jitter: The scalar added to the diagonal of the covariance matrix to ensure positive definiteness.
 
@@ -112,7 +135,11 @@ def multi_fidelity(
     The multi fidelity gaussian process `Model`.
   """
 
-  if observation_index_points is None or observations is None:
+  if (
+    observation_index_points is None or
+    observation_fidelities is None or
+    observations is None
+  ):
     return jit(
       compose(
         likelihood_fn,
@@ -132,6 +159,7 @@ def multi_fidelity(
         partial(
           functions.multi_fidelity.posterior,
           observation_index_points=observation_index_points,
+          observation_fidelities=observation_fidelities,
           observations=observations,
           mean_fn=mean_fn,
           kernel_fn=kernel_fn,
@@ -139,3 +167,34 @@ def multi_fidelity(
         ),
       )
     )
+  
+
+def multi_fidelity_fantasy(
+  mean_fn: Mean,
+  kernel_fn: Callable[[Array, Array], Kernel],
+  jitter: Numeric = 1e-6,
+) -> Callable[[Array, Array, Array], MultivariateNormal]:
+  fantasy_fn = vmap(
+    vmap(
+      jit(
+        partial(
+          functions.multi_fidelity.posterior,
+          mean_fn=mean_fn,
+          kernel_fn=kernel_fn,
+          jitter=jitter,
+        )
+      ),
+      in_axes=(0, 0, None, None, 0)
+    )
+  )
+
+  def fn(fantasy_points, observation_index_points, observation_fidelities, observations):
+    return fantasy_fn(
+      fantasy_points,
+      jnp.ones_like(fantasy_points),
+      observation_index_points,
+      observation_fidelities,
+      observations,
+    )
+  
+  return fn
diff --git a/boax/prediction/models/transformed.py b/boax/prediction/models/transformed.py
index 9f3ca1c..4b9a288 100644
--- a/boax/prediction/models/transformed.py
+++ b/boax/prediction/models/transformed.py
@@ -15,7 +15,7 @@
 """Transformation functions for models."""
 
 from functools import partial
-from typing import Callable, Sequence, TypeVar
+from typing import Any, Callable, Sequence, TypeVar
 
 from jax import vmap
 
@@ -28,24 +28,6 @@
 T = TypeVar('T')
 
 
-def joined(*models: Model[T]) -> Model[Sequence[T]]:
-  """
-  Constructs a joined model.
-
-  Example:
-    >>> transformed = joined(objective_model, cost_model)
-    >>> objective_result, cost_result = transformed(xs)
-
-  Args:
-    models: The models to be joined.
-
-  Returns:
-    The transformed `Model` function.
-  """
-
-  return apply(tuple, *models)
-
-
 def outcome_transformed(
   model: Model[A],
   *transformation_fns: Callable[[A], B],
@@ -96,6 +78,24 @@ def input_transformed(
   )
 
 
+def joined(*models: Model[Any]) -> Model[Sequence[Any]]:
+  """
+  Constructs a joined model.
+
+  Example:
+    >>> transformed = joined(objective_model, cost_model)
+    >>> objective_result, cost_result = transformed(xs)
+
+  Args:
+    models: The models to be joined.
+
+  Returns:
+    The transformed `Model` function.
+  """
+
+  return apply(tuple, *models)
+
+
 def scaled(
   model: Model[T],
   scale_fn: Callable[[T, Array, Array], T],
@@ -152,34 +152,3 @@ def sampled(
     partial(partial, sample_fn),
     model,
   )
-
-
-def fantasized(
-  model: Model[Array],
-  fantasy_fn: Callable[[Array, Array], Model[T]],
-  fantasy_samples: Array,
-) -> Model[T]:
-  """
-  Constructs a fantasy model.
-
-  Example:
-    >>> transformed = fantasized(model, fantasy_fn, fantasy_samples)
-    >>> result = transformed(xs)
-
-  Args:
-    model: The base model.
-    fantasy_fn: The fantasy function.
-    fantasy_samples: The fantasy samples.
-
-  Returns:
-    The transformed `Model` function.
-  """
-
-  return compose(
-    call(fantasy_samples),
-    apply(
-      unwrap(fantasy_fn),
-      identity,
-      model,
-    ),
-  )
diff --git a/tests/prediction/models/models_test.py b/tests/prediction/models/gaussian_process_test.py
similarity index 79%
rename from tests/prediction/models/models_test.py
rename to tests/prediction/models/gaussian_process_test.py
index e50502f..70897d7 100644
--- a/tests/prediction/models/models_test.py
+++ b/tests/prediction/models/gaussian_process_test.py
@@ -9,13 +9,13 @@
 from boax.utils.functools import const
 
 
-class ProcessesTest(parameterized.TestCase):
-  def test_gaussian_process(self):
+class GaussianProcessesTest(parameterized.TestCase):
+  def test_exact_gaussian_process(self):
     key1, key2 = random.split(random.key(0))
 
     index_points = random.uniform(key1, shape=(10, 1), minval=-1, maxval=1)
 
-    model = models.gaussian_process(
+    model = models.gaussian_process.exact(
       means.zero(),
       kernels.rbf(jnp.array(0.2)),
       likelihoods.gaussian(1e-4),
@@ -33,7 +33,7 @@ def test_gaussian_process(self):
       observation_index_points[..., 0]
     )
 
-    model = models.gaussian_process(
+    model = models.gaussian_process.exact(
       means.zero(),
       kernels.rbf(jnp.array(0.2)),
       likelihoods.gaussian(1e-4),
@@ -49,9 +49,11 @@ def test_gaussian_process(self):
   def test_multi_fidelity(self):
     key1, key2 = random.split(random.key(0))
 
-    index_points = random.uniform(key1, shape=(10, 2), minval=-1, maxval=1)
+    index_points, fidelities = random.uniform(
+      key1, shape=(2, 10, 1), minval=-1, maxval=1
+    )
 
-    model = models.multi_fidelity(
+    model = models.gaussian_process.multi_fidelity(
       means.zero(),
       kernels.linear_truncated(
         kernels.matern_five_halves(jnp.array(0.2)),
@@ -61,19 +63,19 @@ def test_multi_fidelity(self):
       likelihoods.gaussian(1e-4),
     )
 
-    mean, cov = model(index_points)
+    mean, cov = model(index_points, fidelities)
 
     assert_shape(mean, (10,))
     assert_shape(cov, (10, 10))
 
-    observation_index_points = random.uniform(
-      key2, shape=(5, 2), minval=-1, maxval=1
+    observation_index_points, observation_fidelities = random.uniform(
+      key2, shape=(2, 5, 1), minval=-1, maxval=1
     )
     observations = jnp.sin(observation_index_points[..., 0]) + jnp.cos(
       observation_index_points[..., 0]
     )
 
-    model = models.multi_fidelity(
+    model = models.gaussian_process.multi_fidelity(
       means.zero(),
       kernels.linear_truncated(
         unbiased=kernels.matern_five_halves(jnp.array(0.2)),
@@ -82,14 +84,31 @@ def test_multi_fidelity(self):
       ),
       likelihoods.gaussian(1e-4),
       observation_index_points,
+      observation_fidelities,
       observations,
     )
 
-    mean, cov = model(index_points)
+    mean, cov = model(index_points, fidelities)
 
     assert_shape(mean, (10,))
     assert_shape(cov, (10, 10))
 
+  def test_joined(self):
+    key = random.key(0)
+
+    samples = random.uniform(key, shape=(4, 10))
+    preds1 = distributions.normal.normal(samples[0], samples[1])
+    preds2 = distributions.normal.normal(samples[2], samples[3])
+
+    model = models.joined(const(preds1), const(preds2))
+
+    result1, result2 = model(jnp.empty((10,)))
+
+    assert_trees_all_close(result1.loc, preds1.loc, atol=1e-4)
+    assert_trees_all_close(result1.scale, preds1.scale, atol=1e-4)
+    assert_trees_all_close(result2.loc, preds2.loc, atol=1e-4)
+    assert_trees_all_close(result2.scale, preds2.scale, atol=1e-4)
+
   def test_scaled(self):
     key1, key2 = random.split(random.key(0))
 
@@ -130,39 +149,6 @@ def test_sampled(self):
 
     assert_trees_all_close(result, samples, atol=1e-4)
 
-  def test_joined(self):
-    key = random.key(0)
-
-    samples = random.uniform(key, shape=(4, 10))
-    preds1 = distributions.normal.normal(samples[0], samples[1])
-    preds2 = distributions.normal.normal(samples[2], samples[3])
-
-    model = models.joined(const(preds1), const(preds2))
-
-    result1, result2 = model(jnp.empty((10,)))
-
-    assert_trees_all_close(result1.loc, preds1.loc, atol=1e-4)
-    assert_trees_all_close(result1.scale, preds1.scale, atol=1e-4)
-    assert_trees_all_close(result2.loc, preds2.loc, atol=1e-4)
-    assert_trees_all_close(result2.scale, preds2.scale, atol=1e-4)
-  
-  def test_fantazised(self):
-    key1, key2 = random.split(random.key(0))
-
-    samples = random.uniform(key1, shape=(10, 3, 1))
-    preds = distributions.normal.normal(*random.uniform(key2, shape=(2, 10, 3)))
-    
-    model = models.fantasized(
-      const(samples),
-      const(const(preds)),
-      jnp.empty((10, 3))
-    )
-
-    result = model(jnp.empty((10,)))
-
-    assert_trees_all_close(result.loc, preds.loc, atol=1e-4)
-    assert_trees_all_close(result.scale, preds.scale, atol=1e-4)
-
 
 if __name__ == '__main__':
   absltest.main()

From 4c4cf679e2f2300a90caf1e6fd5011b062900bfe Mon Sep 17 00:00:00 2001
From: Lando Loeper <lando.loeper@deutschebahn.com>
Date: Sun, 21 Apr 2024 16:10:04 +0200
Subject: [PATCH 7/9] feat: refactor acquisitions interface

---
 boax/optimization/acquisitions/alias.py              | 10 +++++-----
 tests/optimization/acquisitions/acquisitions_test.py | 12 ++++++------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/boax/optimization/acquisitions/alias.py b/boax/optimization/acquisitions/alias.py
index 90888bc..2ab9aaa 100644
--- a/boax/optimization/acquisitions/alias.py
+++ b/boax/optimization/acquisitions/alias.py
@@ -261,7 +261,7 @@ def q_probability_of_improvement(
 
   return jit(
     compose(
-      partial(jnp.mean, axis=-1),
+      partial(jnp.mean, axis=0),
       partial(jnp.amax, axis=-1),
       partial(functions.monte_carlo.qpoi, best=best, tau=tau),
     )
@@ -293,7 +293,7 @@ def q_expected_improvement(
 
   return jit(
     compose(
-      partial(jnp.mean, axis=-1),
+      partial(jnp.mean, axis=0),
       partial(jnp.amax, axis=-1),
       partial(functions.monte_carlo.qei, best=best),
     )
@@ -325,7 +325,7 @@ def q_upper_confidence_bound(
 
   return jit(
     compose(
-      partial(jnp.mean, axis=-1),
+      partial(jnp.mean, axis=0),
       partial(jnp.amax, axis=-1),
       partial(functions.monte_carlo.qucb, beta=beta_prime),
     )
@@ -351,7 +351,7 @@ def q_knowledge_gradient(
 
   return jit(
     compose(
-      partial(jnp.mean, axis=-1),
+      partial(jnp.mean, axis=0),
       partial(jnp.squeeze, axis=-1),
       partial(lax.sub, y=best),
       attrgetter('loc'),
@@ -379,7 +379,7 @@ def q_multi_fidelity_knowledge_gradient(
 
   return jit(
     compose(
-      partial(jnp.mean, axis=-1),
+      partial(jnp.mean, axis=0),
       apply(
         unwrap(cost_fn),
         compose(
diff --git a/tests/optimization/acquisitions/acquisitions_test.py b/tests/optimization/acquisitions/acquisitions_test.py
index 86de411..44ced13 100644
--- a/tests/optimization/acquisitions/acquisitions_test.py
+++ b/tests/optimization/acquisitions/acquisitions_test.py
@@ -73,7 +73,7 @@ def test_q_probability_of_improvement(self):
     s, n, q = 32, 10, 5
 
     best = 0.0
-    preds = random.uniform(key, (n, s, q))
+    preds = random.uniform(key, (s, n, q))
 
     qpoi = acquisitions.q_probability_of_improvement(best)(preds)
 
@@ -84,7 +84,7 @@ def test_q_expected_improvement(self):
     s, n, q = 32, 10, 5
 
     best = 0.0
-    preds = random.uniform(key, (n, s, q))
+    preds = random.uniform(key, (s, n, q))
 
     qei = acquisitions.q_expected_improvement(best)(preds)
 
@@ -95,7 +95,7 @@ def test_q_upper_confidence_bound(self):
     s, n, q = 32, 10, 5
 
     beta = 2.0
-    preds = random.uniform(key, (n, s, q))
+    preds = random.uniform(key, (s, n, q))
 
     qucb = acquisitions.q_upper_confidence_bound(beta)(preds)
 
@@ -106,7 +106,7 @@ def test_q_knowledge_gradient(self):
     s, n = 32, 10
 
     best = 0.0
-    loc, scale = random.uniform(key, (2, n, s, 1))
+    loc, scale = random.uniform(key, (2, s, n, 1))
     preds = distributions.normal.normal(loc, scale)
 
     qkg = acquisitions.q_knowledge_gradient(best)(preds)
@@ -119,9 +119,9 @@ def test_q_multi_fidelity_knowledge_gradient(self):
 
     best = 0.0
     cost_fn = lambda a, b: a / b[..., jnp.newaxis]
-    loc, scale = random.uniform(key1, (2, n, s, 1))
+    loc, scale = random.uniform(key1, (2, s, n, 1))
     preds = distributions.normal.normal(loc, scale)
-    costs = random.uniform(key2, (n,))
+    costs = random.uniform(key2, (s,))
 
     qmfkg = acquisitions.q_multi_fidelity_knowledge_gradient(best, cost_fn)((preds, costs))
 

From c34ea6ac6a843bc6f25e74a43f985345246a7656 Mon Sep 17 00:00:00 2001
From: Lando Loeper <lando.loeper@deutschebahn.com>
Date: Sun, 21 Apr 2024 16:11:16 +0200
Subject: [PATCH 8/9] docs: updated docs

---
 docs/source/api_reference/boax.core.rst       |  10 +
 docs/source/api_reference/boax.prediction.rst |  22 +-
 .../generated/boax.core.samplers.normal.rst   |   6 +
 .../generated/boax.core.samplers.uniform.rst  |   6 +
 ...ediction.models.gaussian_process.exact.rst |   6 +
 ...models.gaussian_process.multi_fidelity.rst |   6 +
 ...oax.prediction.models.gaussian_process.rst |   6 -
 .../boax.prediction.models.multi_fidelity.rst |   6 -
 docs/source/guides/Batched_Optimization.ipynb | 184 ++++---
 .../guides/Constrained_Optimization.ipynb     |  58 +-
 docs/source/guides/Fitting_With_Priors.ipynb  |  55 +-
 docs/source/guides/Getting_Started.ipynb      |  57 +-
 docs/source/guides/Untitled.ipynb             | 494 ------------------
 13 files changed, 240 insertions(+), 676 deletions(-)
 create mode 100644 docs/source/api_reference/generated/boax.core.samplers.normal.rst
 create mode 100644 docs/source/api_reference/generated/boax.core.samplers.uniform.rst
 create mode 100644 docs/source/api_reference/generated/boax.prediction.models.gaussian_process.exact.rst
 create mode 100644 docs/source/api_reference/generated/boax.prediction.models.gaussian_process.multi_fidelity.rst
 delete mode 100644 docs/source/api_reference/generated/boax.prediction.models.gaussian_process.rst
 delete mode 100644 docs/source/api_reference/generated/boax.prediction.models.multi_fidelity.rst
 delete mode 100644 docs/source/guides/Untitled.ipynb

diff --git a/docs/source/api_reference/boax.core.rst b/docs/source/api_reference/boax.core.rst
index 44731dc..ee44a49 100644
--- a/docs/source/api_reference/boax.core.rst
+++ b/docs/source/api_reference/boax.core.rst
@@ -145,6 +145,16 @@ Samplers
 ~~~~~~~~
 
 
+I.I.D Samplers
+^^^^^^^^^^^^^^
+
+.. autosummary::
+  :toctree: generated
+
+  normal
+  uniform
+
+
 Quasi-Random Samplers
 ^^^^^^^^^^^^^^^^^^^^^
 
diff --git a/docs/source/api_reference/boax.prediction.rst b/docs/source/api_reference/boax.prediction.rst
index ebdcdd7..6dc828d 100644
--- a/docs/source/api_reference/boax.prediction.rst
+++ b/docs/source/api_reference/boax.prediction.rst
@@ -22,16 +22,6 @@ Models
 ~~~~~~
 
 
-Gaussian Process Models
-^^^^^^^^^^^^^^^^^^^^^^^
-
-.. autosummary::
-  :toctree: generated
-
-  gaussian_process
-  multi_fidelity
-
-
 Transformed Models
 ^^^^^^^^^^^^^^^^^^
 
@@ -45,6 +35,18 @@ Transformed Models
   scaled
 
 
+Gaussian Process Models
+^^^^^^^^^^^^^^^^^^^^^^^
+
+.. currentmodule:: boax.prediction.models.gaussian_process
+
+.. autosummary::
+  :toctree: generated
+
+  exact
+  multi_fidelity
+
+
 boax.prediction.models.means
 ----------------------------
 
diff --git a/docs/source/api_reference/generated/boax.core.samplers.normal.rst b/docs/source/api_reference/generated/boax.core.samplers.normal.rst
new file mode 100644
index 0000000..c8dd3da
--- /dev/null
+++ b/docs/source/api_reference/generated/boax.core.samplers.normal.rst
@@ -0,0 +1,6 @@
+boax.core.samplers.normal
+=========================
+
+.. currentmodule:: boax.core.samplers
+
+.. autofunction:: normal
\ No newline at end of file
diff --git a/docs/source/api_reference/generated/boax.core.samplers.uniform.rst b/docs/source/api_reference/generated/boax.core.samplers.uniform.rst
new file mode 100644
index 0000000..99035fe
--- /dev/null
+++ b/docs/source/api_reference/generated/boax.core.samplers.uniform.rst
@@ -0,0 +1,6 @@
+boax.core.samplers.uniform
+==========================
+
+.. currentmodule:: boax.core.samplers
+
+.. autofunction:: uniform
\ No newline at end of file
diff --git a/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.exact.rst b/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.exact.rst
new file mode 100644
index 0000000..19383e3
--- /dev/null
+++ b/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.exact.rst
@@ -0,0 +1,6 @@
+boax.prediction.models.gaussian\_process.exact
+==============================================
+
+.. currentmodule:: boax.prediction.models.gaussian_process
+
+.. autofunction:: exact
\ No newline at end of file
diff --git a/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.multi_fidelity.rst b/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.multi_fidelity.rst
new file mode 100644
index 0000000..27dfbbb
--- /dev/null
+++ b/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.multi_fidelity.rst
@@ -0,0 +1,6 @@
+boax.prediction.models.gaussian\_process.multi\_fidelity
+========================================================
+
+.. currentmodule:: boax.prediction.models.gaussian_process
+
+.. autofunction:: multi_fidelity
\ No newline at end of file
diff --git a/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.rst b/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.rst
deleted file mode 100644
index 81a8a77..0000000
--- a/docs/source/api_reference/generated/boax.prediction.models.gaussian_process.rst
+++ /dev/null
@@ -1,6 +0,0 @@
-boax.prediction.models.gaussian\_process
-========================================
-
-.. currentmodule:: boax.prediction.models
-
-.. autofunction:: gaussian_process
\ No newline at end of file
diff --git a/docs/source/api_reference/generated/boax.prediction.models.multi_fidelity.rst b/docs/source/api_reference/generated/boax.prediction.models.multi_fidelity.rst
deleted file mode 100644
index 00f4457..0000000
--- a/docs/source/api_reference/generated/boax.prediction.models.multi_fidelity.rst
+++ /dev/null
@@ -1,6 +0,0 @@
-boax.prediction.models.multi\_fidelity
-======================================
-
-.. currentmodule:: boax.prediction.models
-
-.. autofunction:: multi_fidelity
\ No newline at end of file
diff --git a/docs/source/guides/Batched_Optimization.ipynb b/docs/source/guides/Batched_Optimization.ipynb
index a4da54a..3041392 100644
--- a/docs/source/guides/Batched_Optimization.ipynb
+++ b/docs/source/guides/Batched_Optimization.ipynb
@@ -195,7 +195,7 @@
    "outputs": [],
    "source": [
     "def model_fn(params, x_train, y_train):\n",
-    "    return models.gaussian_process(\n",
+    "    return models.gaussian_process.exact(\n",
     "        models.means.constant(params['mean']),\n",
     "        models.kernels.scaled(\n",
     "            models.kernels.matern_five_halves(nn.softplus(params['length_scale'])),\n",
@@ -308,27 +308,50 @@
   {
    "cell_type": "code",
    "execution_count": 13,
-   "id": "37f102fe-f39a-4b9e-887b-efae32c1a1d3",
+   "id": "d51afc94-5300-487b-87aa-62f463c0b883",
    "metadata": {},
    "outputs": [],
    "source": [
-    "def acquisition_fn(model, acquisition):\n",
-    "    def fn(x):\n",
-    "        return acquisition(vmap(model)(x))\n",
-    "\n",
-    "    return fn"
+    "batch_size = 4\n",
+    "num_results = 100\n",
+    "num_restarts = 40\n",
+    "num_samples = 128"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 14,
-   "id": "d51afc94-5300-487b-87aa-62f463c0b883",
+   "id": "7e1d9bc1-c77d-4b4f-8462-0e27d8ed5953",
    "metadata": {},
    "outputs": [],
    "source": [
-    "batch_size = 4\n",
-    "num_samples = 100\n",
-    "num_restarts = 40"
+    "sampler = samplers.halton_uniform(\n",
+    "    distributions.uniform.uniform(bounds[:, 0], bounds[:, 1])\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "44f12bbf-e795-4162-8b24-7ae4dbb5b8fe",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "initializer_samples = random.normal(\n",
+    "    random.fold_in(sampler_key, 0), (num_samples, num_results, batch_size)\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "51c3436a-5f16-4cec-8e1e-76f0e73ff71c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "solver_samples = random.normal(\n",
+    "    random.fold_in(sampler_key, 1), (num_samples, num_restarts, batch_size)\n",
+    ")"
    ]
   },
   {
@@ -343,7 +366,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 17,
    "id": "9ee04819-738c-43b2-96b7-222970fa6083",
    "metadata": {},
    "outputs": [],
@@ -361,23 +384,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
-   "id": "44f12bbf-e795-4162-8b24-7ae4dbb5b8fe",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "base_samples = jnp.reshape(\n",
-    "    samplers.halton_normal()(\n",
-    "        sampler_key,\n",
-    "        128 * 4,\n",
-    "    ),\n",
-    "    (128, 4)\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "id": "7b3f2faf-77a4-4dec-b109-0017e434235f",
    "metadata": {},
    "outputs": [],
@@ -391,61 +398,90 @@
     "    for i in range(5):    \n",
     "        # Fitting\n",
     "        next_params = fit(_x_train, _y_train)\n",
-    "    \n",
-    "        model = models.sampled(\n",
-    "            model_fn(next_params, _x_train, _y_train),\n",
-    "            distributions.multivariate_normal.sample,\n",
-    "            base_samples,\n",
+    "\n",
+    "        initializer_model = models.sampled(\n",
+    "            vmap(model_fn(next_params, _x_train, _y_train)),\n",
+    "            vmap(distributions.multivariate_normal.sample),\n",
+    "            initializer_samples,\n",
+    "        )\n",
+    "\n",
+    "        solver_model = models.sampled(\n",
+    "            vmap(model_fn(next_params, _x_train, _y_train)),\n",
+    "            vmap(distributions.multivariate_normal.sample),\n",
+    "            solver_samples,\n",
     "        )\n",
     "    \n",
-    "        # Optimizing\n",
+    "        # Selecting    \n",
     "        match p:\n",
     "            case 'poi':\n",
-    "                acqf = acquisition_fn(\n",
-    "                    model,\n",
-    "                    acquisitions.q_probability_of_improvement(\n",
-    "                        best=jnp.max(_y_train),\n",
-    "                        tau=1.0,\n",
-    "                    ),\n",
-    "                )\n",
-    "\n",
     "                bfgs = optimizers.batch(\n",
-    "                    initializer=optimizers.initializers.q_batch_nonnegative(),\n",
-    "                    solver=optimizers.solvers.scipy(method='bfgs'),\n",
-    "                )\n",
-    "            case 'ei':\n",
-    "                acqf = acquisition_fn(\n",
-    "                    model,\n",
-    "                    acquisitions.q_expected_improvement(\n",
-    "                        best=jnp.max(_y_train),\n",
+    "                    optimizers.initializers.q_batch_nonnegative(\n",
+    "                        models.outcome_transformed(\n",
+    "                            initializer_model,\n",
+    "                            acquisitions.q_probability_of_improvement(\n",
+    "                                best=jnp.max(_y_train),\n",
+    "                                tau=1.0,\n",
+    "                            )\n",
+    "                            \n",
+    "                        ),\n",
+    "                        sampler, batch_size, num_results, num_restarts,\n",
     "                    ),\n",
+    "                    optimizers.solvers.scipy(\n",
+    "                        models.outcome_transformed(\n",
+    "                            solver_model,\n",
+    "                            acquisitions.q_probability_of_improvement(\n",
+    "                                best=jnp.max(_y_train),\n",
+    "                                tau=1.0,\n",
+    "                            )\n",
+    "                        ),\n",
+    "                        bounds,\n",
+    "                    )\n",
     "                )\n",
-    "\n",
+    "            case 'ei':\n",
     "                bfgs = optimizers.batch(\n",
-    "                    initializer=optimizers.initializers.q_batch_nonnegative(),\n",
-    "                    solver=optimizers.solvers.scipy(method='bfgs'),\n",
-    "                )\n",
-    "            case 'ucb':\n",
-    "                acqf = acquisition_fn(\n",
-    "                    model,\n",
-    "                    acquisitions.q_upper_confidence_bound(\n",
-    "                        beta=2.0,\n",
+    "                    optimizers.initializers.q_batch_nonnegative(\n",
+    "                        models.outcome_transformed(\n",
+    "                            initializer_model,\n",
+    "                            acquisitions.q_expected_improvement(\n",
+    "                                best=jnp.max(_y_train),\n",
+    "                            ),\n",
+    "                            \n",
+    "                        ),\n",
+    "                        sampler, batch_size, num_results, num_restarts,\n",
     "                    ),\n",
+    "                    optimizers.solvers.scipy(\n",
+    "                        models.outcome_transformed(\n",
+    "                            solver_model,\n",
+    "                            acquisitions.q_expected_improvement(\n",
+    "                                best=jnp.max(_y_train),\n",
+    "                            ),\n",
+    "                        ),\n",
+    "                        bounds,\n",
+    "                    )\n",
     "                )\n",
-    "\n",
+    "            case 'ucb':\n",
     "                bfgs = optimizers.batch(\n",
-    "                    initializer=optimizers.initializers.q_batch(),\n",
-    "                    solver=optimizers.solvers.scipy(method='bfgs'),\n",
+    "                    optimizers.initializers.q_batch(\n",
+    "                        models.outcome_transformed(\n",
+    "                            initializer_model,\n",
+    "                            acquisitions.q_upper_confidence_bound(\n",
+    "                                beta=2.0,\n",
+    "                            ),\n",
+    "                        ),\n",
+    "                        sampler, batch_size, num_results, num_restarts,\n",
+    "                    ),\n",
+    "                    optimizers.solvers.scipy(\n",
+    "                        models.outcome_transformed(\n",
+    "                            solver_model,\n",
+    "                            acquisitions.q_upper_confidence_bound(\n",
+    "                                beta=2.0,\n",
+    "                            ),\n",
+    "                        ),\n",
+    "                        bounds,\n",
+    "                    )\n",
     "                )\n",
     "    \n",
-    "        next_x, _ = bfgs(\n",
-    "            random.fold_in(optimizer_key, i),\n",
-    "            acqf,\n",
-    "            bounds,\n",
-    "            batch_size,\n",
-    "            num_samples,\n",
-    "            num_restarts,\n",
-    "        )\n",
+    "        next_x, _ = bfgs(random.fold_in(optimizer_key, i))\n",
     "    \n",
     "        # Evaluating\n",
     "        next_y = objective(next_x)\n",
@@ -460,13 +496,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "id": "398fefbb-faa5-453c-af49-8d9102f9c294",
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 1200x600 with 1 Axes>"
       ]
@@ -509,7 +545,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.7"
   }
  },
  "nbformat": 4,
diff --git a/docs/source/guides/Constrained_Optimization.ipynb b/docs/source/guides/Constrained_Optimization.ipynb
index 4f3a374..361b770 100644
--- a/docs/source/guides/Constrained_Optimization.ipynb
+++ b/docs/source/guides/Constrained_Optimization.ipynb
@@ -269,7 +269,7 @@
    "outputs": [],
    "source": [
     "def model_fn(params, x_train, y_train):\n",
-    "    return models.gaussian_process(\n",
+    "    return models.gaussian_process.exact(\n",
     "        models.means.constant(params['mean']),\n",
     "        models.kernels.scaled(\n",
     "            models.kernels.matern_five_halves(nn.softplus(params['length_scale'])),\n",
@@ -382,40 +382,43 @@
   {
    "cell_type": "code",
    "execution_count": 17,
-   "id": "5efd1762-f5d3-4b7c-a861-1d67fdd7d97e",
+   "id": "e8bff142-5b36-4bc3-a80a-5285f423ea71",
    "metadata": {},
    "outputs": [],
    "source": [
-    "def acquisition_fn(model, acquisition):\n",
-    "    def fn(x):\n",
-    "        return acquisition(vmap(model)(x))\n",
-    "\n",
-    "    return fn"
+    "batch_size = 1\n",
+    "num_results = 500\n",
+    "num_restarts = 100"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 18,
-   "id": "4dd5042f-daed-415d-907a-a4ccb4f9086e",
+   "id": "c1702d8f-d715-48e7-bf2b-f061a9848991",
    "metadata": {},
    "outputs": [],
    "source": [
-    "bfgs = optimizers.batch(\n",
-    "    initializer=optimizers.initializers.q_batch_nonnegative(),\n",
-    "    solver=optimizers.solvers.scipy(method='bfgs'),\n",
+    "sampler = samplers.halton_uniform(\n",
+    "    distributions.uniform.uniform(bounds[:, 0], bounds[:, 1])\n",
     ")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 19,
-   "id": "e8bff142-5b36-4bc3-a80a-5285f423ea71",
+   "id": "4dd5042f-daed-415d-907a-a4ccb4f9086e",
    "metadata": {},
    "outputs": [],
    "source": [
-    "batch_size = 1\n",
-    "num_samples = 500\n",
-    "num_restarts = 100"
+    "def optimizer_fn(acqf):\n",
+    "    return optimizers.batch(\n",
+    "        optimizers.initializers.q_batch_nonnegative(\n",
+    "            acqf, sampler, batch_size, num_results, num_restarts,\n",
+    "        ),\n",
+    "        optimizers.solvers.scipy(\n",
+    "            acqf, bounds,\n",
+    "        ),\n",
+    "    )"
    ]
   },
   {
@@ -474,19 +477,21 @@
     "            distributions.mvn_to_norm,\n",
     "        )\n",
     "\n",
-    "        # Optimizing\n",
+    "        # Selecting\n",
     "        match p:\n",
     "            case 'ei':\n",
-    "                acqf = acquisition_fn(\n",
-    "                    obj_model,\n",
+    "                acqf = models.outcome_transformed(\n",
+    "                    vmap(obj_model),\n",
     "                    acquisitions.expected_improvement(\n",
     "                        best=jnp.max(_y_train),\n",
     "                    ),\n",
     "                )\n",
     "            case 'cei':\n",
     "                feasible = _y_train[_c_train <= 0]\n",
-    "                acqf = acquisition_fn(\n",
-    "                    models.joined(obj_model, fsb_model),\n",
+    "                acqf = models.outcome_transformed(\n",
+    "                    models.joined(\n",
+    "                        vmap(obj_model), vmap(fsb_model)\n",
+    "                    ),\n",
     "                    acquisitions.constrained(\n",
     "                        acquisitions.expected_improvement(\n",
     "                            best=jnp.array(-2) if not jnp.any(feasible) else jnp.max(feasible),\n",
@@ -497,13 +502,8 @@
     "                    )\n",
     "                )\n",
     "    \n",
-    "        next_x, _ = bfgs(\n",
-    "            random.fold_in(optimizer_key, i),\n",
-    "            acqf,\n",
-    "            bounds,\n",
-    "            batch_size,\n",
-    "            num_samples,\n",
-    "            num_restarts,\n",
+    "        next_x, _ = optimizer_fn(acqf)(\n",
+    "            random.fold_in(optimizer_key, i)\n",
     "        )\n",
     "    \n",
     "        # Evaluating\n",
@@ -527,7 +527,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/AAAAIWCAYAAADwNWYlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAChuElEQVR4nOzdeXwU9f0/8NfeR+47IeQgARIiiICoeHGoKCAe9US8aPWrtZ7YaqtVe2lbba1U8ar3gUqp+lNRPFDxqIpoxSMhkISQEMhF7uy9O78/YjZZkuxussnOZyav5+PBo7Oz89l5Z/JymvfOpZEkSQIRERERERERCU0rdwFEREREREREFBobeCIiIiIiIiIFYANPREREREREpABs4ImIiIiIiIgUgA08ERERERERkQKwgSciIiIiIiJSADbwRERERERERArABp6IiIiIiIhIAfRyFyAan88Hl8sFnU4HjUYjdzlERERERESkcpIkwev1wmg0Qqsd+jg7G/iDuFwufPrpp3KXQUREREREROPMMcccA7PZPOT7bOAPotPpAADTp0/3T4tqx44dKC4ulrsMoiExoyQ6ZpREx4ySEjCnJDolZNTr9eL7778P2YOygT9I72nzOp1O+AY+MzNT+BppfGNGSXTMKImOGSUlYE5JdErKaKjLuHkTOwWTJEnuEoiCYkZJdMwoiY4ZJSVgTkl0asooG3gFa2xslLsEoqCYURIdM0qiY0ZJCZhTEp2aMsoGnoiIiIiIiEgB2MAr2JQpU+QugSgoZpREx4yS6JhRUgLmlESnpoyygVewvXv3yl0CUVDMKImOGSXRMaOkBMwpiU5NGWUDr2B2u13uEoiCYkZJdMwoiY4ZJSVgTkl0asqo6hv4hx56CIceeiji4+MRHx+PefPm4a233pK7rFFhNpvlLoEoKGaURMeMkuiYUVIC5pREp6aMqr6BnzhxIv7yl7/gq6++wrZt27Bo0SKcfvrp+OGHH+QuLWI5OTlyl0AUFDNKomNGSXTMKCkBc0qiU1NGVd/AL1++HEuXLsWUKVMwdepU3HnnnYiNjcXnn38ud2kR27Vrl9wlEAXFjJLomFESHTNKSsCckujUlFG93AVEk9frxb///W90d3dj3rx5QZft7OyEVtv3/YbJZILJZBrrEomIiIiIiIgGNS4a+O+++w7z5s2Dw+FAbGwsXnnlFZSUlAQdM336dNhsNv/rVatW4ZprrkFWVhYqKysBABkZGZAkCY2NjQB6Hk+wd+9e2O12mM1m5OTk+L/tSU9Ph1arRX19PQCgsLAQ9fX16O7uhslkQn5+PsrLywEAqampMBqN2LdvHwBg0qRJaGpqQldXFwwGAyZPnoyysjK4XC7U19fDYrGgrq4OAJCfn4+WlhZ0dHRAp9OhqKgIZWVlkCQJiYmJiIuLQ21tLQAgNzcXHR0daGtrg0ajwbRp01BeXg6v14v4+HgkJSVhz549AHouRbDZbGhpaQEAlJSUYOfOnfB4PIiLi0Nqaip2794NAJgwYQKcTicOHDgAACguLkZVVRVcLhdiYmKQkZGBqqoqAEBWVhY8Hg+ampoAAFOnTkVNTQ0cDgcsFguys7NRUVHh394A0NDQAACYPHky6urq/Ns7NzcXO3fuBACkpaVBr9dj//79AICCggI0NDSgu7sbRqMRBQUF2LFjBwAgJSUFJpMpYHs3Nzejs7MTer0eU6dORWlpKQAgOTkZVqvVfyfLvLw8tLa2Drm94+PjUVNTA6Dn1J3Ozs4ht3dycjKqq6sBANnZ2bDb7f7tPW3aNFRUVMDtdiM2NhZpaWkB29vlcqG5uRkAUFRUhOrqajidTsTExCAzM9Of2czMTPh8voDM1tbW+rf3xIkTAzKr0Wj827uwsBD79++HzWaDyWRCXl5e0O3d2NgIl8uFioqKAdvbbDYPmtmDt3dSUhJiY2MDMtve3o729nZotVoUFxdjx44d8Pl8SEhIQEJCQsD27urqQmtr64DMDra9HQ7HoJmNjY1Fenp60Mzu2bMHTqcTVqtVqH1Eb2a5jxh6H+FyuVBaWsp9hEz7iK6urkG3N/cRffuI3oxyH8G/I0TeR7hcLuzevZv7CP4dIew+ondfKvI+QpIkhEMjhbukgrlcLtTU1KC9vR0bNmzAY489hi1btgzaxHs8HmzZsgUFBQXCH4FvaWlBcnKy3GUQDYkZJdExoyQ6ZpSUgDkl0Skho16vF9u3b8f8+fOh1w99nF3118ADgNFoxOTJkzFnzhz8+c9/xsyZM7FmzZqgY+Li4vx3ro+PjxeueQfg/4aNSFTMKImOGSXRMaOkBMwpiU5NGR0XDfzBfD4fnE6n3GUQERERERERhU3118D/5je/wZIlS5Cbm4vOzk6sW7cOH374Id5++225S4tYYWGh3CUQBcWMkuiYURIdM0pKwJyS6NSUUdUfgW9sbMTFF1+MoqIinHDCCfjyyy/x9ttv46STTpK7tIip6VQQUidmlETHjJLomFFSAuaURKemjKr+CPzjjz8udwljpru7W+4SiIJiRkl0zCiJjhklJWBO6WDdVbUou/VeuJpbgy531JuPQWvoa0mr//US9q1/K+Tnx88sxvS//Tpg3jdX3AZbVe2AZVMXHQXP6ceHWbn4VN/Aq5mIN9Yj6o8ZJdExoyQ6ZpSUgDml/hz7m/DlOdfCUdcQeuGDHojmrG9Gx3c7Qw7Tx8UOmNddUYPOH3YNmB8zJR+xKsooG3gFy8/Pl7sEoqCYURIdM0qiY0ZJCZhT6q+7Yg9cB3qOvGt0OkAX/lXbGp0OGqMh9HIG3cB5+sHHanQ6VWV0XDwHfjh6nwM/c+ZM6HQDgyGS0tLSQZ9lTyQKZpREx4yS6JhRUgLmlA7WuvVblP32Psx+9m6YM1LlLkcRGQ33OfA8Ak9ERERERESjJumIQzHv7ceh0WjkLkV1VH8XejVLTZX/2yyiYJhREh0zSqJjRkkJmNPxTZIkNH3w+YD5IjXvasooG3gFMxqNcpdAFBQzSqJjRkl0zCgpAXM6vu288yF8tWI1yv+4FqJena2mjLKBV7B9+/bJXQJRUMwoiY4ZJdExo6QEzOn4Vf3Ii9j9wHMAgN0PrkP7/8pkrmhwasooG3giIiIiIiIaln0bNmHHHf/0vy75841InC32jeLUgA28gk2aNEnuEoiCYkZJdMwoiY4ZJSVgTsefps2f4bvr7/S/Lrzxp8i99CcyVhScmjLKBl7Bmpqa5C6BKChmlETHjJLomFFSAuZ0fGn7+gd8c9mtkDxeAEDOxWdi8i9/JnNVwakpo2zgFayrq0vuEoiCYkZJdMwoiY4ZJSVgTsePrp3V+GrljfDaHQCAjFMXouTPq4W64/xg1JRRNvAKZjAY5C6BKChmlETHjJLomFFSAuZ0fLDXNWDbihvgbu0AACQfMxsz194BjU4nc2WhqSmjbOAVbPLkyXKXQBQUM0qiY0ZJdMwoKQFzOj54bXbgx8fExc+YitlP/RVakzIez6amjLKBV7CyMjEf00DUixkl0TGjJDpmlJSAOR0fYqfk48jXH0HaScdgzrp7oY+LkbuksKkpo3q5CyAiIiIiIiLxWbIzMOfZe+QuY1zjEXgFS05OlrsEoqCYURIdM0qiY0ZJCZhTdZJ8PtQ+/xp8bo/cpURMTRllA69gFotF7hKIgmJGSXTMKImOGSUlYE7VqfyPD+KHG/+C/116MzzddrnLiYiaMspT6BWsrq4OCQkJcpdBNCRmlETHjJLomFFSAuZ09LiaW9FdVRt0mZiCHBhTk/yvPV3d6CytDOvzE2aXQKvvawHtdQ1w1DUMWK55y1ZUP7QOAND0wRdo++p7pB4/N6x1iEhNGWUDT0REREREJLPOHVX4YvkV8HR2B11u5sO/R9YZJ/lfd+3cgy9OuzKsdZywYxO0ifH+13UvvYmKu/8VdMwhd/9K0c272vAUegXLz8+XuwSioJhREh0zSqJjRkkJmNPI+VxufHfNH0I279E25ebLkXPh6XKXETE1ZZRH4BWspaUFVqtV7jKIhsSMkuiYURIdM0pKwJyOjpQFR6LjhwpY87ORvvjYIZeLKcwNeG1KT0b+lSvCWofWGPjc9sQ5hww5Nn7GVGT9ZHFYnys6NWWUDbyCdXR0yF0CUVDMKImOGSXRMaOkBMxp5LRGA4pu/TnSTzoGOqsZ8dOnhj3WMjETxb+7ZkTrTZ1/BFLnHzGisUqipoyygVcwnU4ndwlEQTGjJDpmlETHjJISMKejJ+mIQ+UuQZXUlFFeA69gRUVFcpdAFBQzSqJjRkl0zCgpAXM6cs6mFrlLGBfUlFE28ApWVlYmdwlEQTGjJDpmlETHjJISMKcj0/zRl9gy9yfY/eA6SF6v3OWompoyygZewSRJkrsEoqCYURIdM0qiY0ZJCZjT4XO3d+L76++Ez+FC+R8eQP3rH8hdkqqpKaNs4BUsMTFR7hKIgmJGSXTMKImOGSUlYE6Hr+zWe+HY1wgASDnucGSetkjmitRNTRllA69gcXFxcpdAFBQzSqJjRkl0zCgpAXM6PPWvv499G94GAOjjYzH9vluh0bItG0tqyiiTomC1tbVyl0AUFDNKomNGSXTMKCkBcxo+Z+MB/HDzPf7X0+68AZbsDBkrGh/UlFE28ERERERERGNMkiR8f+Nf4G5pBwBkLFuACWefInNVpDRs4BUsNzdX7hKIgmJGSXTMKImOGSUlYE7DU/fCG2h691MAgDEtGYf89VfQaDQyVzU+qCmjbOAVrKOjQ+4SiIJiRkl0zCiJjhklJWBOQ7Pt2Yey29b4X0//+69hTE2SsaLxRU0ZZQOvYG1tbXKXQBQUM0qiY0ZJdMwoKQFzGpqnswum9GQAwMQLliN98bEyVzS+qCmjerkLoJHjKTckOmaURMeMkuiYUVIC5jS0+OlTcfR7T6Pq/mdQ8IsL5S5n3FFTRjWSmp5qPwo8Hg+2bNmCmTNnQqfTyV0OERERERERqZzX68X27dsxf/586PVDH2fnKfQKVl5eLncJREExoyQ6ZpREx4ySEjCng/O5PZC8XrnLIKgro2zgFczLHQIJjhkl0TGjJDpmlJSAOR1c5b1P4IszroKteq/cpYx7asooG3gFi4+Pl7sEoqCYURIdM0qiY0ZJCZjTgdq++h6Va55B25ff4b8n/wzuji65SxrX1JRRNvAKlpTER0+Q2JhREh0zSqJjRkkJmNNAXpsD317zR8DnAwBMuvJ8GOJjZa5qfFNTRtnAK9iePXvkLoEoKGaURMeMkuiYUVIC5jRQ+Z8ehK2qFgCQMKsEk665SOaKSE0ZZQNPREREREQ0Cpq3bEXNExsAAFqLCTPuvw3aIHcUJxouNvAKNnHiRLlLIAqKGSXRMaMkOmaUlIA57eFu68B319/pf130218gdnKejBVRLzVllF8HKZjNZlPVDRlIfZhREh0zSqJjRkkJIslp6xfbsf+Vd1Hyl18GzK99/jW0fva/kOMTDitB3mXnBMwr++0/4G7rCDk2e8VypBwz2//asb8JO+98MKy6i/9wPYzJCf7XTe9/jl1/fRTO/U0AgJTj5yJ31U/C+iwae2ral7KBV7CWlhZkZmbKXQbRkJhREh0zSqJjRkkJRprTls+/wbbzr4dGrx/QwLd/9QP2bXg75Gd4Ha4BDXzDm1vg2NcYcmzSvFkBDbynoyusdQLAlN9cCaCvge/eVY2O7TsAAPr4WEz/xy3QaHmysyjUtC9lqoiIiIiIKKo6vivH1xf9Cj6HC/D64OnslrukUaHR6TD977+GJTtD7lJIpTSSJElyFyESj8eDLVu2YObMmdDpdHKXQ0RERESkKt2VNfjitCvhOtAGAEg6aibmrv8ntEaDfxlnUws8XbaQn6WzmmHOSA2YZ6vZD8nrDTnWmJIY8Hg3n8sNe11DWD+DZWImtIa+k5nd7Z1wtbTDmBQPQ6I6TtWm6PJ6vdi+fTvmz58PfZAbH/IUegXbuXMnpk6dKncZRENiRkl0zCiJjhklJRhOTh37GvHludf5m/fEuTNw+Lp/BDTvAGBKS4YpLXlE9Vhzs0Y0Tms0IGbSyG52ZkiIgyEhbkRjaeypaV/KU+gVzOPxyF0CUVDMKImOGSXRMaOkBOHm1HWgDV+edx0cPx7ljiuZjDnP3gOd1TyW5RGpal/KBl7B4uL4LR+JjRkl0TGjJDpmlJQgnJx6urqx7YLV6N61BwBgzc/G4S/+g6ebU1SoaV/KBl7BUlNTQy9EJCNmlETHjJLomFFSglA59Tqc+PqSm/13aTdlpuLwl9bAlJ4SjfKIVLUvZQOvYLt375a7BKKgmFESHTNKomNGSQlC5VSj0/mbdUNSPA5/8T5Y8yZEozQiAOral/ImdkRERERENGa0Bj0OXXsHTOkpyDz9BMQVF8hdEpFisYFXsAkT+M0liY0ZJdExoyQ6ZpSUIJycarRaFP/+2ihUQzSQmvalPIVewZxOp9wlEAXFjJLomFESHTNKSjBYTqv/9RK6dlVHvxiiQahpX8oGXsEOHDggdwlEQTGjJDpmlETHjJISHJzTPY/9GztuW4MvTr8K7d+Wy1QVUR817UvZwBMRERER0ajY95+3UfbbfwAA3C1taNv6rcwVEakLG3gFKy4ulrsEoqCYURIdM0qiY0ZJCXpz2vjOJ/ju2j/55xdcfwnyLjtHrrKI/NS0L2UDr2BVVVVyl0AUFDNKomNGSXTMKClBVVUVWj77H775v99C8noBADmXnIkpN/+fzJUR9VDTvpQNvIK5XC65SyAKihkl0TGjJDpmlJSg6/td+Prim+Bz9OQ168yTUPLnG6HRaGSujKiHmvalfIycgsXExMhdAlFQzCiJjhkl0TGjJLquij1o+s0aeDu7AQCpi+Zhxj9vg0bL44QkDjXtS9nAK1hGRobcJRAFxYyS6JhREpWruRUd3++Eu7wK1drPAQCJc6YjcfYh/mV8ThdqnnklrM/LOv1EmNJT/K+7K2vQ9P5nIcdpDQbkXvqTgHnNW7aia+fukGNjJuUg7cSjA+bVPv8avDZ7yLEpx81FXHGB/7XrQBv2vfx2yHEAMPGC5dDHWP2v278tR+sX34QcZ0xKwISzTwmY1/DmFtjr6kOOjZ8+FcnzZvlfS5KEPY+tD6vejFOOhyUny//aVrMfjW9/FNbY/MvPC3jd8t//oeOHnSHHWSZmImPJ/IB5devfgru9I+TY5KMOQ/yMIgCAu6ML2867Ht7WnnGJRxyKWY/dCa2BLQaJRU3/f8//uhSsqqoKJSUlcpdBNCRmlETHjJIIHPVN6Ph2Jzq+K//x30446hoGLDf5V5cFNPBehxM7blsT1joS50wPaOA7vtsZ1lh9fOyABn7/K++i7sWNIcdmnnbCgAa+4p7H4KxvDjl2+r23BDTwzsYDYf+smacuCmjgWz79CuW/fyDkuNiiSQMa+JqnX8aBLV+GHJt/xfkBDTwkKex6Y6fkBzTw3buqwxur0Qxo4Bve2oI9/wr9xUHqwqMGNPBV9z+D7l17Qo4t/sN1/gbeEB+LvJ+ejfI/rkXcIVMw59l7oLOaQ9dOFGVq+v97NvBEREREUSBJEtwt7TCmJAbM/+qCG9FZWiFPUUQRmvSLlWj2uXDo+afDkBAndzlEqscGXsGysrJCLxSEx+3B3vLaoMuYcicEjmlph7erO+Rnay1mGNKSA+a59jVA8nhDjtUnxUMXF+t/Lbk9cO1vDDkOAIxZ6dD0O23L29UNT0t7yHEavQ7GCYGn1ribWuCzO0KO1cVaoU9ODJjnrN0PSFLIsfrUJOisFv9rn8MJd+OBkOMAwDgxM+D6Mk9bB7wdXSHHaU1GGDJSA+a56psgudwhx+oS4qDv93/OktcL1yBHiXp5Oz2o/r7nNEtDZiq0RmPfe102eFraQq5To9XCODEzYJ77QCt83aFPwdTGWGBISQqY56qrh+T1hRyrT06ELrbvCI7P5YI7jKNGAGCckAGNXud/7e3ogqct9GmJGqMBxsy0gHnuhmb4nKFvvKKLi4E+KcH/WpIkuGr3h1WvIT0FWrOpr16bHZ7m1rDGKn0fAY3Vn9FguI8Ym31Ef2rcR0hOJ+zlVbD/sAv2H3bCXlYBXVICpr3zbOBnFU8G+jXw2hgrLIdMgaVkCqScLMQkJfbMnzoJ+zucfet0a5B7720hawWAzpQ02PuNdRVPDWusxqALWCcAWM5aitwjZocca8xKGzA26/br/Dc6C8ZzaHHAWG9cYtg/6wGtEdr+6z1qblhjdXExA+pNvOwCxJx+yhAj+hgKcgLGSpIUdr22iYFj3bm5YY89uF7j0hOQO60odL3pyQPGpv/qSni7bCHHSiWTB4yNO2MZWkwxwEHziUShjUuFT5KgVcGNFTWSFMZfEOOIx+PBli1bMHPmTOh0utADZNTU1IS0tLTQCw6iq60Tm45ZidgDQzckXq0Wa/5wf8C8hW+sx6zPt4T8/Kqi6Xj1op8HzFv1j98h6UBTyLHvLzsH38xb4H+d0NKMn917R8hxAPDE9XegLTXd//rQLz7Cia+/FHJcW1IKnrjxDwHzTnv+EUwu+zbk2O1HHIvNp60ImHft766H3hP6j93/d8H/obJkpv919u5dOO/x+0KOA4B/3vEPeAx9f+we8+5rOHJL6OsD9+YVYv3lqwPmXfDgX5G5rybk2E9OXI6tC/r+kDHbunDVXTeHVe/zP78JDdl5/tfF32zF0g1Phxxnt8TgoVvvDph3yoanUfLN1pBjyw49HG+duypg3hV//jViujtDjn3rrItRNutI/+u0/bW4aO1fQo4DgId//WfYYuP9r+d+9A6Oe+f/hRzXmJmN566+JWDe2Y+vQe7u0NczfnncSfj45DP8r7VeL66/49qw6v33qmtRW9j3B9+kHd/hzOceDjmO+wjuIw7GfUR4+4gHfvs3uMx9X8wUlm1HVs1uNE7IQeOEHLQlpQK8ARgR0ah649KZMOrF3bd6vV5s374d8+fPh14/9HF2HoFXsEga+P9t/Cxo805ERESjqysuAY0TcmCy2wIa+MppM1E5bWaQkURERD3YwI9Tjva+0yjb8vLhmzhh4EJaLRYWBp5amHzYNLToQh81MuXlDRjrPvJwtLSHPp09d0YBkvqN1adq0XLM0UFG9JlblAFPYqL/dZytAC0tocd64uIH1GueMwMtybFDjOiTeEjJgLHtR8+DxusJObaoJBe5/caazNlh/6zHT06B1O/bucTGIrR4Qh810mVNGFAvjpiNlqaJIcdOOHRywFit3Ry0XrfLDYPRAAA4rCgLroy+sTGefLTsD/2z+szmAfXGzT4ELTGhd1+xU6cOGGubdwSc9tCn1hZOz0dmv7GGeE/Yv5t5U9Lgs/adWhvfNhkt3WH8rCmpA+rVzz0MLRNShxjRJ21mUeBYrzfseqdPm4jJOX1jLdrc8MaqYB+hbcoIayz3EWOzj+hPlfsIjRaOrCzYJ+XDlp8PT1LP8oeH/PQ+He3tiE9ICL0gkYyYUxJdR3s7VHD2PACeQj+Akk6h93g8QU+vCOaN9R+h/MlXYHA5MXXVmTjlwpNGuTqiyDJKFA3MKImOGSUlYE5JdErIaLin0It7EQCFVFMT+nrEoTiKi/D2WRfjjRWXw3r0cI4FEIUvkowSRQMzSqJjRkkJmFMSnZoyygZewRyO0Hc/Hord03eHXbOBMaCxEUlGiaKBGSXRMaOkBMwpiU5NGWXnpmAWiyX0QkOwu/saeItB7EsFSLkiyShRNDCjJDpmlJSAOSXRqSmjbOAVLDs7e8RjHf0aeCuPwNMYiSSjRNHAjJLomFFSAuaURKemjLJzU7CKiooRj025/0H8/M6bcNk9v4W2IfRzl4lGIpKMEkUDM0qiY0ZJCZhTEp2aMir2rfhozGg6u2Cxd/f8izHJXQ4RERERERGFwCPwCpaRkTHywf2ecRubEDMK1RANFFFGiaKAGSXRMaOkBMwpiU5NGVV9A//nP/8Zc+fORVxcHNLT03HGGWegvLxc7rJkp+l3J0ZLnFXGSoiIiIiIiCgcqm/gt2zZgl/84hf4/PPP8e6778LtdmPx4sXo7u6Wu7SINTQ0jHis9scG3m00QafjXehpbESSUaJoYEZJdMwoKQFzSqJTU0ZVfw38pk2bAl4/9dRTSE9Px1dffYXjjz9epqrkp3M4AQAeE69/JyIiIiIiUgLVN/AHa29vBwAkJycHXa6zsxNabd8JCiaTCSbBmt3JkyePeKze2XME3mM2j1Y5RANEklGiaGBGSXTMKCkBc0qiU1NGx1UD7/P5cP311+OYY47B9OnTgy47ffp02Gw2/+tVq1bhmmuuQVZWFiorKwH03AxBkiQ0NjYCAKZMmYK9e/fCbrfDbDYjJycHu3btAgCkp6dDq9Wivr4eAFBYWIj6+np0d3fDZDIhPz/ff21+amoqjEYj9u3bBwCYNGkSmpqa0NXVBYPBgMmTJ6OsrAw2mw0TJ06ExWJBXV0dACA/Px8tLS3o6OiATqdDUVERysrKIEkSEhMTERcXh9raWuidPx6BN5pQWloKjUaDadOmoby8HF6vF/Hx8UhKSsKePXsAABMnToTNZkNLSwsAoKSkBDt37oTH40FcXBxSU1Oxe/duAMCECRPgdDpx4MABAEBxcTGqqqrgcrkQExODjIwMVFVVAQCysrLg8XjQ1NTzKLupU6eipqYGDocDFosF2dnZ/sc+9N58ovcUmMmTJ6Ours6/vXNzc7Fz504AQFpaGvR6Pfbv3w8AKCgoQENDA7q7u2E0GlFQUIAdO3YAAFJSUmAymQK2d3NzMzo7O6HX6zF16lSUlpYC6Pnix2q1Yu/evQCAvLw8tLa2Drm94+PjUVNTAwDIyclBZ2cn2traBt3eycnJqK6uBtDzrEq73e7f3tOmTUNFRQXcbjdiY2ORlpYWsL1dLheam5sBAEVFRaiurobT6URMTAwyMzP9mc3MzITP5wvIbG1trX97T5w4MSCzGo3Gv70LCwuxf/9+2Gw2mEwm5OXlBd3ejY2NaGxsRGJi4oDtbTabB83swds7KSkJsbGxqK2tBQDk5uaivb0d7e3t0Gq1KC4uxo4dO+Dz+ZCQkICEhISA7d3V1YXW1tYBmR1sezscjkEzGxsbi/T09KCZ3bNnD5xOJ6xWq1D7iN7MjmQf0bu9Ozo6hsysGvYRlZWVsFqt3EfItI/o6uoadHtzH9G3j/j2229htVq5j+DfEULvI1paWpCSksJ9BP+OEHYfsWfPHlitVqH3EZIkIRwaKdwlVeDnP/853nrrLXzyySeYOHHioMt4PB5s2bIFBQUFwh+BLy0tRUlJybDHOe1OfDBpIQCgdfJUrPjkqVGujKjHSDNKFC3MKImOGSUlYE5JdErIqNfrxfbt2zF//nzo9UMfZx83R+CvvvpqvPHGG/joo4+GbN77i4uLE/7mbuYRnv5u6+w7s0Cy8BR6GjsjzShRtDCjJDpmlJSAOSXRqSmjqm/gJUnCNddcg1deeQUffvghJk2aJHdJoyY3N3dE45w6PV674HIYnE4UTs4a5aqI+ow0o0TRwoyS6JhRUgLmlESnpoyq/jFyv/jFL/Dcc89h3bp1iIuLQ319Perr62G32+UuLWK912AMl1OrR0XJYSibdSTccw4b3aKI+hlpRomihRkl0TGjpATMKYlOTRlVfQP/0EMPob29HQsWLEBWVpb/30svvSR3abKxu33+abNe7MsEiIiIiIiIqMe4OIVerdLS0kY0zu7pa+AtBtV/h0MyGmlGiaKFGSXRMaOkBMwpiU5NGVV9A69mwe5OGIytoQXZ1RVwmUywFsWOclVEfUaaUaJoYUZJdMwoKQFzSqJTU0Z5+FXBep87OFy2z77CeY/9Axet/QusH348ylUR9RlpRomihRkl0TGjpATMKYlOTRllAz8Oubr6HiNnjLXKWAkRERERERGFiw28ghUUFIxonKd/Ax/HBp7GzkgzShQtzCiJjhklJWBOSXRqyigbeAVraGgY0ThPd98j9HgEnsbSSDNKFC3MKImOGSUlYE5JdGrK6Khfzb93717U19fDZrNh7ty5sFgso70K+lF3d/eIxnm7+47Am+NjRqscogFGmlGiaGFGSXTMKCkBc0qiU1NGR+UIfGdnJ2677Tbk5OQgLy8PRx55JBYuXIjdu3cHLPfiiy/i3HPPxeWXXz4aqx33jEbjiMb5bA7/tIWn0NMYGmlGiaKFGSXRMaOkBMwpiU5NGY34CPyuXbuwdOlSVFVVBTxzXaPRDFj2qKOOwoUXXghJknDJJZfg2GOPjXT149pIr+Xw2fpOobfE8zFyNHbUdL0RqRMzSqJjRkkJmFMSnZoyGtEReIfDgWXLlqGyshJWqxU33XQT3njjjSGXz8/Px8KFCwEAr732WiSrJgA7duwY2UB7XwNvTeAp9DR2RpxRoihhRkl0zCgpAXNKolNTRiM6Av/QQw+hoqICMTEx+Pjjj3HYYYeFHLNkyRJs3rwZn332WSSrpkjY+06hj01kA09ERERERKQEER2Bf/nll6HRaHDdddeF1bwDwMyZMwH0nHpPkUlJSRnROI2d18BTdIw0o0TRwoyS6JhRUgLmlESnpoxGdAS+rKwMALB48eKwx/RuvLa2tkhWTQBMJtOIxr2/+tfY39COWJ8bp+h0o1wVUZ+RZpQoWphREh0zSkrAnJLo1JTRiI7Ad3V1AQBiY8O/EZrT6QQAGAyGSFZNAPbt2zeicQ4v4DJbIKUkj3JFRIFGmlGiaGFGSXTMKCkBc0qiU1NGI2rge4+mV1dXhz3mhx9+AABkZmZGsmqKgM3tBQBYDDz6TkREREREpBQRNfCzZ88GAHz00Udhj3nmmWeg0Wgwb968SFZNACZNmjSicQ63DwBgMUT06ycKaaQZJYoWZpREx4ySEjCnJDo1ZTSiDu7ss8+GJEl49NFHUVNTE3L5++67z9/sr1ixIpJVE4Dm5uZhj3E6XZj35n9w1PsbMenrrWNQFVGfkWSUKJqYURIdM0pKwJyS6NSU0Yga+IsuugiHHnooHA4HFixYgLfeeguSJPnf12g0kCQJX375JVauXIkbb7wRGo0Gxx13HJYsWRJx8eNdZ2fnsMfY2rtx+CebcfT7b2Li5/8dg6qI+owko0TRxIyS6JhRUgLmlESnpoxGdBd6rVaL1157Dcceeyyqq6tx6qmnwmq1QqPRAAAWLFiAzs5O/43rJElCYWEh1q9fH3nlBL1++L++7rbuvhcW8yhWQzTQSDJKFE3MKImOGSUlYE5JdGrKaMQXQefm5uKbb77BihUroNVq0d3dDUmSIEkSmpqa4HA4/Eflzz33XGzduhXp6ekRF07A1KlThz3G1tHln9ZYLaNZDtEAI8koUTQxoyQ6ZpSUgDkl0akpo6PyVURycjKef/553HXXXdi4cSO2bduGxsZGeL1epKSkYNasWVi+fLmqNpwISktLUVJSMqwx9va+I/BaCxt4GlsjyShRNDGjJDpmlJSAOSXRqSmjo3ouQV5eHq666qrR/EgaZY5Om39aG8MGnoiIiIiISCn4HDEFS05OHvYYZ78GXh/LBp7G1kgyShRNzCiJjhklJWBOSXRqymhEDfxRRx2FBx54AE1NTaNVDw2D1Wod9hhnZ98p9LqY4Y8nGo6RZJQomphREh0zSkrAnJLo1JTRiBr4rVu34rrrrkN2djaWLFmC5557Dt3d3aEH0qjYu3fvsMe4u/qOwBvZwNMYG0lGiaKJGSXRMaOkBMwpiU5NGY2ogZ8yZQokSYLH48E777yDSy65BBkZGbjggguwceNGeL3e0aqTRklAAx/HBp6IiIiIiEgpImrgy8vL8eWXX+KGG25AVlYWJEmCzWbDSy+9hNNOOw1ZWVm4+uqr8d///ne06qV+8vLyhj3GmZiI2klTUJ+dC1O6eq4FITGNJKNE0cSMkuiYUVIC5pREp6aMRnwTuzlz5uDvf/87amtr8d577+GnP/0pEhISIEkSmpub8dBDD+G4445DQUEBbrvtNpSVlY1G3QSgtbV12GPajjkG//7Z9Vj385sRf+Rho18UUT8jyShRNDGjJDpmlJSAOSXRqSmjo3YXeo1Gg0WLFuGxxx5DfX09/vOf/+Css86CyWSCJEmorq7GXXfdhenTp2P27Nm49957R2vV41ZHR8ewxzjcPv+0Rc+HENDYGklGiaKJGSXRMaOkBMwpiU5NGR2TDs5oNOLMM8/Ev//9bzQ0NODxxx/HCSecAK1WC0mS8M033+BXv/rVWKx6XNHpdMMeY/f03ZfAYmADT2NrJBkliiZmlETHjJISMKckOjVldMw7uLi4OKxatQrvvPMOnnrqKSQmJo71KseNoqKiYY+x9T8Cb1BPkElMI8koUTQxoyQ6ZpSUgDkl0akpo2PewH/99df45S9/iZycHFx88cVob28f61WOGyO5n0D+ww9j5dq/4Nx/3Qujxz0GVRH14T0vSHTMKImOGSUlYE5JdGrKqH4sPrSqqgrPP/881q1bh507dwIAJEkCAMTExOCMM87AypUrx2LV40rvNh0O8759SNpfCwCwxJhGuySiACPJKFE0MaMkOmaUlIA5JdGpKaOj1sA3NTXhxRdfxLp167B161YAfRtKr9dj8eLFWLlyJU4//XRYrXz++GgYyeUIWocDAOA2mlR1LQiJiZfMkOiYURIdM0pKwJyS6NSU0Yga+O7ubrz88st4/vnn8f7778Pr7blBWm/jPm/ePKxcuRLnnnsuUlNTI6+WAsTHxw97jM7hBAC4TTz6TmNvJBkliiZmlETHjJISMKckOjVlNKIGPj09HY4fj+j2Nu3FxcVYuXIlLrjgAkyaNCnyCmlINTU1KCkpGdYYvbPn9+VlA09RMJKMEkUTM0qiY0ZJCZhTEp2aMhpRA2+32wEAEyZMwPnnn4+VK1di1qxZo1IYjQ2Ds+cIvM9slrkSIiIiIiIiGo6IGvhVq1Zh5cqVWLhwITQazWjVRGHKyckZ1vJOuxM6rwcAG3iKjuFmlCjamFESHTNKSsCckujUlNGIHiP3+OOPY9GiRWzeZdLZ2Tms5W2dNv+0xAaeomC4GSWKNmaURMeMkhIwpyQ6NWV0zJ8DT2Onra1tWMt3t3X3vbBaRrcYokEMN6NE0caMkuiYUVIC5pREp6aMsoFXsOGe+WBv7+oba2EDT2OPZ+eQ6JhREh0zSkrAnJLo1JTRsK6BX7RoEYCeH3zz5s0D5o/EwZ9Fwzdt2rRhLe+Ki8MHS8+C0elEwdzhjSUaieFmlCjamFESHTNKSsCckujUlNGwGvgPP/wQwMBvLj788ENoNBr/I+TC0bu8mr4FkUt5eTmKiorCXt4VH4//Hd3zpcukmRljVRaR33AzShRtzCiJjhklJWBOSXRqymhYDfzxxx8/aMM91HyKDq/XO6zlHW6ff9pi4NUTNPaGm1GiaGNGSXTMKCkBc0qiU1NGh3UEPtz5FB3x8fHDWt7u7guuWc8GnsbecDNKFG3MKImOGSUlYE5JdGrKaETPgSd5JScnD2t5e2sHYtta4TaZYOGJExQFw80oUbQxoyQ6ZpSUgDkl0akpozwMq2DV1dXDWt7z+jv4v7/9Fr+481cw/PfzsSmKqJ/hZpQo2phREh0zSkrAnJLo1JTRiBp4rVYLvV6P0tLSsMdUVlb6x1F0ebrt/mlTHB8jR0REREREpCQRH4Efzh3oR2Mc9cnOzh7W8t5um3/aHBcz2uUQDTDcjBJFGzNKomNGSQmYUxKdmjIq2yn0vHt95Ox2e+iF+vHZHP5pSzwbeBp7w80oUbQxoyQ6ZpSUgDkl0akpo1Fv4JubmwEAMTFsICPV0tIyrOV9tr7gsoGnaBhuRomijRkl0TGjpATMKYlOTRkdlQY+3KPp3d3duP/++wEAhYWFo7FqGo5+3zxZEmJlLISIiIiIiIiGa1h3kisoKBh0/uLFi2EwGIKOdTqdaGxshM/ng0ajwfLly4ezahrEtGnThjeg3xH4GB6BpygYdkaJoowZJdExo6QEzCmJTk0ZHVYDP9jt9yVJQl1d3bBWetRRR+Gmm24a1hgaqKKiAlOmTAl7eY2j7xr4mATrWJREFGC4GSWKNmaURMeMkhIwpyQ6NWV0WA38JZdcEvD66aefhkajwWmnnYbExMQhx2k0GpjNZmRlZeHoo4/GokWLeBO7UeB2u4e1vPbHBt5tNEGn041FSUQBhptRomhjRkl0zCgpAXNKolNTRofVwD/55JMBr59++mkAwJ133omSkpLRq4rCEhs7vOvYdQ4nAMBtMo1FOUQDDDejRNHGjJLomFFSAuaURKemjA6rgT/YHXfcAQBIT08flWJoeNLS0oa1/Bs/Xw1XRzcyLFr8ZIxqIupvuBklijZmlETHjJISMKckOjVlNKK70N9xxx244447kJqaOlr10DDs3r17WMs3xSahOTMb9kmD34yQaLQNN6NE0caMkuiYUVIC5pREp6aMRv058CQPt9cHt08CAFgN/LUTEREREREpTUSd3H//+1/odDpYLJaw7kRfV1cHs9kMvV6Pr776KpJVE4AJEyaEvazD4/NPm9nAU5QMJ6NEcmBGSXTMKCkBc0qiU1NGI+rkXnzxRUiShFNPPRXZ2dkhl8/Ozsby5cvh8/mwbt26SFZNAFwuV9jLdrV04LDPP0TJ158htbpqDKsi6jOcjBLJgRkl0TGjpATMKYlOTRmNqIH/5JNPoNFosGTJkrDHLFu2DADw0UcfRbJqAtDc3Bz2sl11TVj0xr9xysvPIeuDD8awKqI+w8kokRyYURIdM0pKwJyS6NSU0Yga+MrKSgAY1iPkiouLAQAVFRWRrJqGyd7Z7Z/WxFhkrISIiIiIiIhGIqIG3uFwAADMZnPYY0w/PoO8u7s7xJIUSlFRUdjL2jv6treODTxFyXAySiQHZpREx4ySEjCnJDo1ZTSiBj45ORkAUFNTE/aYvXv3AgASExMjWTUBqK6uDntZZ4fNP61nA09RMpyMEsmBGSXRMaOkBMwpiU5NGY2oge89df61114Le8yrr74KQF3fgsjF6XSGv2xn/yPw1rEoh2iA4WSUSA7MKImOGSUlYE5JdGrKaEQN/NKlSyFJEp555hl8/PHHIZf/6KOP8Oyzz0Kj0eDUU0+NZNUEICYmJuxl3d19R+CNbOApSoaTUSI5MKMkOmaUlIA5JdGpKaMRNfBXXHEFUlNT4fV6sXTpUjzwwAP+6+L7czgc+Oc//4lly5bB4/EgKSkJP//5zyNZNQHIzMwMe1l3l90/bYhjA0/RMZyMEsmBGSXRMaOkBMwpiU5NGY2ogY+NjcW6deug0+lgs9lw3XXXIT09HQsXLsQFF1yACy64AAsXLkRaWhpuuOEGdHd3Q6/X44UXXkB8fPxo/QzjVu9TAMLh6XcE3sQGnqJkOBklkgMzSqJjRkkJmFMSnZoyqo/0A0488US8/fbbuOiii7Bv3z50dXUNeMa7JEkAgOzsbDz77LNYsGBBpKulYfJ29TXw5nj1nEJCREREREQ0XkTcwAPAwoULUVlZiWeeeQZvvPEG/ve//6G5uRkAkJqaitmzZ2P58uW48MIL/Y+Ro8gN51QQl8mCtqQUGF1OWOJjx7Aqoj5qOl2J1IkZJdExo6QEzCmJTk0ZHZUGHuh5vvvll1+Oyy+/fLQ+kkLw+XxhL1t35hnYPGM+AODJGVPGqiSiAMPJKJEcmFESHTNKSsCckujUlNGIroEneTU2Noa9rN3dF1qzQTcW5RANMJyMEsmBGSXRMaOkBMwpiU5NGWUDP070b+Atev7aiYiIiIiIlGbUOrnNmzfjoosuwuTJkxEbGwu9Xo/S0tKAZT766CM8+OCDeO6550ZrtePalCnhnwpvd3v902YDG3iKjuFklEgOzCiJjhklJWBOSXRqymjEnZzNZsM555yDxYsXY926daiqqoLNZvPfeb4/nU6Hq6++Gpdccgl27doV6arD9tFHH2H58uWYMGECNBoNXn311aiteyzV1taGvey0p5/Aac8/ilP+3zpoNZoxrIqoz3AySiQHZpREx4ySEjCnJDo1ZTTiBv7cc8/Fyy+/DEmSMHfuXPzyl78cctljjjkG06dPBwD85z//iXTVYevu7sbMmTOxdu3aqK0zGhwOR9jLppWVYnLZduTv+G4MKyIKNJyMEsmBGSXRMaOkBMwpiU5NGY2ogf/Pf/6DN998EwDw6KOP4vPPP8fdd98ddMxPfvITSJKELVu2RLLqYVmyZAn+9Kc/4cwzz4zaOqPBYrGEvaze2RNaLx/jR1E0nIwSyYEZJdExo6QEzCmJTk0Zjegxck8//TQA4MILL8Rll10W1pg5c+YAAMrKyiJZ9Zjr7OyEVtv3/YbJZBLuGfYTJ04Me1m9ywkA8JnNY1UO0QDDySiRHJhREh0zSkrAnJLo1JTRiBr4bdu2QaPR4Lzzzgt7TFZWFgCgqakpklWPuenTp8Nms/lfr1q1Ctdccw2ysrJQWVkJAMjIyIAkSf7HEkyZMgV79+6F3W6H2WxGTk6O/1r/9PR0aLVa1NfXAwAKCwtRX1+P7u5umEwm5Ofno7y8HACQmpoKo9GIffv2AQAmTZqEpqYmdHV1wWAwYPLkySgrK0NnZyfy8vJgsVhQV1cHAMjPz0dLSws6Ojqg0+lQVFSE77Z/C73HA6DnCHzvzQVzc3PR0dGBtrY2aDQaTJs2DeXl5fB6vYiPj0dSUhL27NkDoCf0NpsNLS0tAICSkhLs3LkTHo8HcXFxSE1Nxe7duwEAEyZMgNPpxIEDBwAAxcXFqKqqgsvlQkxMDDIyMlBVVQWgJw8ej8efh6lTp6KmpgYOhwMWiwXZ2dmoqKjwb28AaGhoAABMnjwZdXV1/u2dm5uLnTt3AgDS0tKg1+uxf/9+AEBBQQEaGhrQ3d0No9GIgoIC7NixAwCQkpICk8kUsL2bm5vR2dkJvV6PqVOn+rdZcnIyrFYr9u7dCwDIy8tDa2trwPYuKyuDJElITExEfHw8ampqAAA5OTno7OwccnsnJyejuroaAJCdnQ273e7f3tOmTUNFRQXcbjdiY2ORlpYWsL1dLheam5sBAEVFRaiurobT6URMTAwyMzP9mc3MzITP5wvIbG1trX97T5w4MSCzGo3Gv70LCwuxf/9+2Gw2mEwm5OXlBd3ejY2N2L9/P1JSUgZsb7PZPGhmD97eSUlJiI2N9V+3lJubi/b2drS3t0Or1aK4uBg7duyAz+dDQkICEhISArZ3V1cXWltbB2R2sO3tcDgGzWxsbCzS09ODZnbPnj1wOp2wWq1C7SN6MxtqH9E/s3FxcQHbW+37iJ07dyIuLo77CJn2EV1dXYNub+4j+vYRX3/9NeLi4riP4N8RQu8jmpqakJ6ezn0E/44Qdh9RVVWFuLg4ofcRg91DbjAaKdwlB2EymeDxePDVV1/hsMMO88/XarXQaDT47rvvUFJSEjDm66+/xuGHHw6TyQS73T7SVY+YRqPBK6+8gjPOOGPQ9z0eD7Zs2YKCggLhj8CXlpYO2L6DaW1sxReHLgMAtMw4FBe8+/BYl0YEIPyMEsmFGSXRMaOkBMwpiU4JGfV6vdi+fTvmz58PvX7o4+wRXQOfkJAAAP5vE8LR+81JampqJKsec3FxcYiPj/f/E615B3q+aQtHd1t33wsLT6Gn6Ak3o0RyYUZJdMwoKQFzSqJTU0YjauCnTp0KANi+fXvYY3of4TZr1qxIVk3oOZsgHPaOvgZeo6IbOJD4ws0okVyYURIdM0pKwJyS6NSU0Yga+GXLlkGSJNx///1h3Zr/448/xosvvgiNRoPly5dHsuph6erqwjfffINvvvkGQM9ZAN98843/Ghel6r0+IxR7Z18Dr7XyCDxFT7gZJZILM0qiY0ZJCZhTEp2aMhpRA/+LX/wCycnJaGhowNlnn+2/6cDBPB4P/vWvf+HUU0+Fz+dDTk4OLr300khWPSzbtm3DrFmz/Ef9V69ejVmzZuH222+PWg1y6n8EXhtjlbESIiIiIiIiGqmI7kIfHx+Pl156CUuXLsVbb72FnJwczJ8/3//+TTfdBJfLhW3btqG9vR2SJMFsNmP9+vUwGAwRFx+uBQsWhH1XPyUpLCwMazl3YhK+OnohjE4n8ounjHFVRH3CzSiRXJhREh0zSkrAnJLo1JTRiI7AA8AJJ5yA999/H7m5ubDb7di0aZP/GoO33noLmzdvRltbGyRJQk5ODj744AMcccQRERdO8D+2IBRX9gRsWXo23j1zJfQLjh7jqoj6hJtRIrkwoyQ6ZpSUgDkl0akpoxEdge91zDHHYNeuXXjxxRfx2muvYdu2bWhsbITX60VKSgpmzZqF0047DZdccgmMRuNorJKAgOfUB2P3+PzTFoNurMohGiDcjBLJhRkl0TGjpATMKYlOTRkdlQYeAPR6PS688EJceOGFo/WRFEK4j7azu/s38BGfdEEUNhEfv0jUHzNKomNGSQmYUxKdmjLKbk7B8vLywlrO4XABP94DgA08RVO4GSWSCzNKomNGSQmYUxKdmjLKbk7Bdu7cGdZyxieeww23X4Nf/PFGaEvDG0M0GsLNKJFcmFESHTNKSsCckujUlFE28OOAz2aHRpJgcjpgsfIeBEREREREREoU1jXwBQUFAACNRoPKysoB80e8cr0eCQkJmDp1KpYsWYIVK1ZAp+NN1sKVlpYW1nK+brt/2hIfM1blEA0QbkaJ5MKMkuiYUVIC5pREp6aMhtXAV1dXA4D/8XAHz4/U119/jRdffBH//Oc/8d577yE+Pn5UPlft9Pow70Fo79fAJ8SOUTVEA4WdUSKZMKMkOmaUlIA5JdGpKaNh/SSXXHLJsOaHy+fzoaOjA9u3b0d1dTW++uor/OlPf8Ldd98d0eeOF/v370dSUlLoBe0O/2QMj8BTFIWdUSKZMKMkOmaUlIA5JdGpKaNhNfBPPvnksOaPxKpVq/D000/jlVdeYQM/yjSOvgbeGmeRsRIiIiIiIiIaKWFuYnfppZcCAGpra+UtREHCvQeB9scG3m0wQm9Qz+kjJL5I75NBNNaYURIdM0pKwJyS6NSUUWEa+Ly8PBx//PGYN2+e3KUoRmNjY1jL6XobeLN5LMshGiDcjBLJhRkl0TGjpATMKYlOTRkd9cOxkiShqqoKLS0tAIDk5GQUFBQMuAHewfLz8/Hhhx+Odjmq1tXVFdZy+h8beK/JNJblEA0QbkaJ5MKMkuiYUVIC5pREp6aMjloDv2nTJjz44IP48MMP0d3dHfCe1WrFggULcNVVV2HJkiWjtcpxz2gM75nuepcTAODjEXiKsnAzSiQXZpREx4ySEjCnJDo1ZTTiBt5ms+Giiy7Cq6++CqDnCPzBuru78eabb+LNN9/Eaaedhueeew4xMbwbeqTCuZbD45Pw8sW/gMHlRE4aHyFH0aWm641InZhREh0zSkrAnJLo1JTRiBp4n8+HpUuX4uOPP4YkSTAYDFi8eDGOOOIIZGRkAAAaGhrw5Zdf4p133oHL5cJrr72GpUuX4sMPPwx5Wj0Ft2PHDpSUlARdxu72Ym/BVABAysS4aJRF5BdORonkxIyS6JhRUgLmlESnpoxG1MA/8sgj+Oijj6DRaHDyySfjscceQ3Z29qDL1tXV4fLLL8emTZvwySef4OGHH8bPf/7zSFZPYbC7ff5pi0EnYyVEREREREQUiYjuQv/0008DAObOnYuNGzcO2bwDQHZ2Nl5//XUcccQRkCTJP5ZGLiUlJeQyjv4NvF6Yhw7QOBFORonkxIyS6JhRUgLmlESnpoxG1NGVlZVBo9HghhtugFYb+qN0Oh1Wr17tH0uRMYdxU7qu5lYU7PgOOVU7Edd6IApVEfUJJ6NEcmJGSXTMKCkBc0qiU1NGI2rge69hnzp1athjpkyZEjCWRq6uri7kMp2lFTjjuYdxzhNrkPze5ihURdQnnIwSyYkZJdExo6QEzCmJTk0ZjaiBLywsBAA0NjaGPaZ32d6xNLacnTb/tM5qkbESIiIiIiIiikREDfyKFSsgSRKeeeaZsMc888wz0Gg0OO+88yJZNQHIz88PuUz/Bl4fywaeoiucjBLJiRkl0TGjpATMKYlOTRmNqIG/9tprMXv2bLz44ou4++67Qy5/zz334IUXXsCsWbNw/fXXR7JqAtDS0hJyGXd3XwNvjLWOZTlEA4STUSI5MaMkOmaUlIA5JdGpKaNhPUaupqZmyPcee+wxXHHFFfjNb36DF154AZdccgnmzp2L9PR0aDQa/3Pgn332WXzzzTeYO3cuHn30UdTX1yM3N3fUfpDxqKOjI+Qy7i47TD9OG+JixrYgooOEk1EiOTGjJDpmlJSAOSXRqSmjYTXwkyZNCuvDvv32W9x4441Bl9m2bRtmz54NjUYDj8cT1ufS4PT60L8+T1e3v4E3xfEUeoqucDJKJCdmlETHjJISMKckOjVlNKyfRJKksa6DRiCcu/97u+3+aVMsj8BTdA3nCRVEcmBGSXTMKCkBc0qiU1NGw2rgn3zyybGug0agtLQUJSUlQZfx2foaeEs8G3iKrnAySiQnZpREx4ySEjCnJDo1ZTSsBv6SSy4Z6zpojPRv4K3xvIkdERERERGRUkV0F3qSV1JSUshlfB4ffBoNAMCSGDfWJREFCCejRHJiRkl0zCgpAXNKolNTRtVzNf84FBsbG3KZ7y6/Ev9begH0bjf+nZkShaqI+oSTUSI5MaMkOmaUlIA5JdGpKaOj2sA3NDTgww8/xPfff+9/1l5ycjKmT5+OBQsWICMjYzRXN+7V1taGvJbD4fECGg08RiPMRn5fQ9EVTkaJ5MSMkuiYUVIC5pREp6aMjkpHt3//fqxevRovv/zykI+G0+v1OOuss/D3v/8dWVlZo7FaCoPN7QMAmPRa6LQamashIiIiIiKikYr4Gvjt27fj0EMPxfr16+F2uyFJ0qD/3G43XnrpJcycORPffffdaNQ+7uXm5oZcxvFjA2/R83YHFH3hZJRITswoiY4ZJSVgTkl0aspoRF1dd3c3li1bhgMHDkCSJJx44ol46aWXUF1dDYfDAYfDgerqaqxfvx6LFy+GJElobm7GsmXLYLPZRutnGLfa29tDLjNzwzos2LgBh326OQoVEQUKJ6NEcmJGSXTMKCkBc0qiU1NGI2rgH3jgAezbtw9arRb/+te/8M477+Ccc85Bbm4ujEYjjEYjcnNzcfbZZ2PTpk147LHHoNFoUFdXh7Vr147WzzBuhRPE4s8/xuzPPkDhl59FoSKiQGraWZI6MaMkOmaUlIA5JdGpKaMRNfD/7//9P2g0Glx66aX42c9+FnL5n/70p1i1ahUkScIrr7wSyaoJgFYb/Nfncrqh//GeBD6zORolEQUIlVEiuTGjJDpmlJSAOSXRqSmjEf0kO3fuBACcf/75YY9ZsWJFwFgaueLi4qDv29q7/NMSG3iSQaiMEsmNGSXRMaOkBMwpiU5NGY2oge/q6mkQk5OTwx6TlJQEoOf6eYrMjh07gr7f1dZvG1vYwFP0hcookdyYURIdM0pKwJyS6NSU0Yga+LS0NABAWVlZ2GN6N15qamokqyYAPp8v6Pv2jr4GXmO1jHU5RAOEyiiR3JhREh0zSkrAnJLo1JTRiBr4o446CpIk4d577x3y+e/9eTwe3HvvvdBoNDjqqKMiWTUBSEhICPq+vbOvgdeygScZhMookdyYURIdM0pKwJyS6NSU0Yga+IsvvhgA8M0332DZsmXYt2/fkMvu27cPy5cvx9dffw0AuPTSSyNZNSGMBr7fEXhtjHWsyyEaQE07S1InZpREx4ySEjCnJDo1ZVQfyeDly5fjjDPOwKuvvor33nsPBQUFWLx4MY488kikp6dDo9GgoaEBX3zxBd599124XC4AwJlnnolly5aNyg8wntXU1KCkpGTI950dNv+0jkfgSQahMkokN2aURMeMkhIwpyQ6NWU0ogYeAF544QVcfPHF+Pe//w2Xy4WNGzdi48aNA5aTJAkAcM455+CZZ56JdLUUBme/U+j1sWzgiYiIiIiIlCziB+KZTCa89NJLeP3117FkyRJYLBZIkhTwz2KxYMmSJXjjjTfw0ksvwWQyjUbt415OTk7Q911JidhVMhPVhcUwTJwQpaqI+oTKKJHcmFESHTNKSsCckujUlNGIj8D3WrZsGZYtWwav14uqqiq0tLQA6HnEXEFBAXQ63Witin7U1dWFuLi4Id+3T5+O1y/oeWzfrQvyo1QVUZ9QGSWSGzNKomNGSQmYUxKdmjIa8RH4g+l0OkyZMgVHHnkkjjzySEyZMoXN+xhpbW0N+r7D7fVPWwyj/qsmCilURonkxoyS6JhRUgLmlESnpoyyq1Mxu7vveYdmPb9EISIiIiIiUjI28AoW6k6Kdk9fA88j8CQHtdztk9SLGSXRMaOkBMwpiU5NGWVXp2A7d+4M+n7yI4/hZ3+7HRfffyd0TU1RqoqoT6iMEsmNGSXRMaOkBMwpiU5NGR21m9hR9Hk8nqDvaw+0IKHtAADAYjZGoySiAKEySiQ3ZpREx4ySEjCnJDo1ZZRH4BUsPj4++AJ2h38yJj5mjKshGihkRolkxoyS6JhRUgLmlESnpoyygVew5OTkoO9rHH0NvDXeOtblEA0QKqNEcmNGSXTMKCkBc0qiU1NG2cArWHV1ddD3tT828G6DEXoDr5ag6AuVUSK5MaMkOmaUlIA5JdGpKaNs4FVM9+Mp9B6TSeZKiIiIiIiIKFJs4BUsOzs76Pt6Z28Db45GOUQDhMookdyYURIdM0pKwJyS6NSUUTbwCubod437YPQuJwDAZ+YReJJHqIwSyY0ZJdExo6QEzCmJTk0ZHbUGfteuXbjttttw4oknYvr06SgsLERFRUXAMt9//z3efPNNbNmyZbRWO64dOHBgyPdcTjf0Pz4uwWfmEXiSR7CMEomAGSXRMaOkBMwpiU5NGY34zmY+nw833XQT1qxZA5/PB0mSAAAajQYulytg2ZqaGpx66qnQ6/XYvXu3qk5lEI2tvcs/LVksMlZCREREREREoyHiI/BXXHEF/vGPf8Dr9WLChAk4++yzh1x26dKlmDRpErxeLzZs2BDpqse94uLiId9z6vTY9JOLsPnUc3HghBOiWBVRn2AZJRIBM0qiY0ZJCZhTEp2aMhpRA79582Y8/vjjAIBbbrkF1dXVWL9+fdAx55xzDiRJwvvvvx/JqglAVVXVkO85tQaUzj4K24+aD8fRR0axKqI+wTJKJAJmlETHjJISMKckOjVlNKJT6B999FEAPUfW//SnP4U15ogjjgAA/PDDD5GsmoABlyj0Z/d4/dNWA+9VSPIIllEiETCjJDpmlJSAOSXRqSmjEXV2n332GTQaDX72s5+FPWbixIkAgPr6+khWTQBiY2OHfM/u9vmnzQZdNMohGiBYRolEwIyS6JhRUgLmlESnpoxGdAS+sbERAJCfnx/2GIPBAADw/HiHdBq59PT0Id/rbm5Han0dXEYTLO7E6BVF1E+wjBKJgBkl0TGjpATMKYlOTRmN6Ah8TEwMAKCpqSnsMXv37gUAJCcnR7JqQvBrOez//RIXP3AXLrv3DsS+x/sNkDzUdL0RqRMzSqJjRkkJmFMSnZoyGlEDX1BQAAAoLS0Ne8xbb70FADjkkEMiWTWF4Oqy+acNsTEyVkJERERERESjIaIGfvHixZAkCWvXroXP5wu5fGlpKZ566iloNBosXbo0klUTgKysrCHfc3f2NfDGWGs0yiEaIFhGiUTAjJLomFFSAuaURKemjEbUwF977bWIiYlBZWUlrrzyyqDXtb/77rtYvHgxHA4HkpOTcfnll0eyakLw+wh4uvsaeFO8JRrlEA3Ae12Q6JhREh0zSkrAnJLo1JTRiBr4jIwMPPzwwwCAxx9/HIWFhbjqqqv8769Zswb/93//h0MOOQSnnHIK9u3bB61Wi6eeekpVdwKUS7B7D3i77f5pE0+hJ5kM5/4YRHJgRkl0zCgpAXNKolNTRiO6Cz0ArFy5EgaDAVdccQVqa2vxyCOPQKPRAAAee+wxAIAkSQB6bt//9NNPY9myZZGulkLw2foaeEs8G3giIiIiIiKli+gIfK9zzz0XFRUV+P3vf485c+ZAp9NBkiT/v0MOOQS/+c1vUFFRgTPPPHM0VkkApk6dOuR7Uv8GPoENPMkjWEaJRMCMkuiYUVIC5pREp6aMjkoDDwApKSm47bbbsHXrVjgcDjQ2NmL//v1wOp347rvvcOedd6rq+Xsi2LNnz5Dv9W/grfG8XIHkESyjRCJgRkl0zCgpAXNKolNTRiM+hX4wWq0WqampY/HR1I/T6Rz6TbvDP2nlEXiSSdCMEgmAGSXRMaOkBMwpiU5NGR21I/AUfVbr0I+H0zj6GviYON6FnuQRLKNEImBGSXTMKCkBc0qiU1NGx+QIPEVHsOcZfnzV9aivb0esz4VTDPw1kzzU9MxNUidmlETHjJISMKckOjVlNKzObtGiRaO+Yo1Gg82bN4/6544nlZWVKCkpGfS9Tr0ZXYla6M1s3kk+wTJKJAJmlETHjJISMKckOjVlNKzu7sMPP4RGo/E/Di4SvZ/T+6i5aFm7di3uuece1NfXY+bMmbj//vtxxBFHRLWGaLK7vQAAi4FXSRAREREREalBWA388ccfH/WGezS99NJLWL16NR5++GEceeSRuO+++3DyySejvLxc0XfGz8jIGPI9u8cHgA08yStYRolEwIyS6JhRUgLmlESnpoyGfQReye69915cfvnlWLVqFQDg4YcfxsaNG/HEE0/g17/+tczVjdxQZ0S4XG7Mfu9NuI0mxBdOBH4yLcqVEfUYjbN2iMYSM0qiY0ZJCZhTEp2aMqr6w7MulwtfffUVTjzxRP88rVaLE088EZ999tmQ4zo7O9HR0eH/J+KjBxobGwedb2vrwjGb38CCt/6D/A/fj3JVRH2GyiiRKJhREh0zSkrAnJLo1JRR1d/hrLm5GV6vd8BpExkZGdixY8eQ46ZPnw6bzeZ/vWrVKlxzzTXIyspCZWWl/zMkSfIHYsqUKdi7dy/sdjvMZjNycnKwa9cuAEB6ejq0Wi3q6+sBAIWFhaivr0d3dzdMJhPy8/NRXl4OAEhNTYXRaMS+ffsAAJMmTUJTUxO6urpgMBgwefJklJWVobOzE/X19bBYLKirqwMA5OfnY3fF7r4fxGJGWVkZJElCYmIi4uLiUFtbCwDIzc1FR0cH2traoNFoMG3aNJSXl8Pr9SI+Ph5JSUnYs2cPAGDixImw2WxoaWkBAJSUlGDnzp3weDyIi4tDamoqdu/uWe+ECRPgdDpx4MABAEBxcTGqqqrgcrkQExODjIwMVFVVAei5I6TH40FTUxMAYOrUqaipqYHD4YDFYkF2djYqKir82xsAGhoaAACTJ09GXV2df3vn5uZi586dAIC0tDTo9Xrs378fAFBQUICGhgZ0d3fDaDSioKDA//tPSUmByWQK2N7Nzc3o7OyEXq/H1KlTUVpaCgBITk6G1WrF3r17AQB5eXlobW1FR0cHdDodioqKArZ3fHw8ampqAAA5OTno7OwccnsnJyejuroaAJCdnQ273e7f3tOmTUNFRQXcbjdiY2ORlpYWsL1dLheam5sBAEVFRaiurobT6URMTAwyMzP9mc3MzITP5wvIbG1trX97T5w4MSCzGo3Gv70LCwuxf/9+2Gw2mEwm5OXlBd3ejY2N6OzsREVFxYDtbTabAzLb0tKCjo6OAds7KSkJsbGxAZltb29He3s7tFotiouLsWPHDvh8PiQkJCAhISFge3d1daG1tXVAZgfb3g6HY9DMxsbGIj09PWhm9+zZA6fTCavVKtQ+ojezB+8jerf3YJkdb/uIzs5OlJaWch8h0z6iq6tr0O3NfUTfPqI3o9xH8O8IkfcRnZ2d2L17N/cR/DtC2H1E775U5H1EuGcJaKQxOJ/A4/H4w56UlAS9Xr7vCfbt24fs7Gz897//xbx58/zzb7rpJmzZsgVffPFFwPIejwdbtmxBQUEBtNq+ExRMJhNMJlPU6g6H2+2GwWAYMH/X1ztRufRSAEDr/OOx4qW/RLkyoh5DZZRIFMwoiY4ZJSVgTkl0Ssio1+vF9u3bMX/+/KD986idQl9aWoprr70WJSUlMJvNyMzMRGZmJsxmM6ZNm4ZrrrkG33///WitLmypqanQ6XT+b1J6NTQ0IDMzc8hxcXFxiI+P9/8TrXkH4P/29mD2ji7/tMZqiVY5RAMMlVEiUTCjJDpmlJSAOSXRqSmjETfwPp8PN954I2bOnIm1a9f6T0ORJAmSJMHn86G8vBwPPvggZs2ahRtuuAE+n280ag+L0WjEnDlzAp457/P5sHnz5oAj8kpkt9sHn9/e7Z/WsYEnGQ2VUSJRMKMkOmaUlIA5JdGpKaMRn9t+wQUX4N///rf/nP1DDjkERxxxRMA1BF9++SW+//57eL1e/POf/8S+ffvw0ksvRbrqsK1evRqXXHIJDj/8cBxxxBG477770N3d7b8rvVKZzeZB5zs7+67d18VYo1UO0QBDZZRIFMwoiY4ZJSVgTkl0aspoRA38iy++iPXr10Oj0WDmzJl49NFHMXfu3EGX/fLLL3HllVfif//7HzZs2IAXX3wR559/fiSrD9t5552HpqYm3H777aivr8dhhx2GTZs2Kf55gDk5OYPOd3X1NfD6WB6BJ/kMlVEiUTCjJDpmlJSAOSXRqSmjEZ1C/+ijjwLouXviJ598MmTzDgBz587FRx99hKKiIkiShEceeSSSVQ/b1Vdf7b/D4xdffIEjjzwyqusfC713nTxY/wbeEMsj8CSfoTJKJApmlETHjJISMKckOjVlNKIGfvv27dBoNLj55psRExMTcvmYmBjcfPPN/rE0Ntxdfdd4GGND/16IiIiIiIhIfBGdQu9yuQAAhx56aNhjepd1u92RrJrQ87zHwbisVrRNyIHB5URuSkKUqyLqM1RGiUTBjJLomFFSAuaURKemjEbUwOfl5aGsrAzt7e1hj+no6PCPpcj0f059f23z5+PfydMAAH87eko0SyIKMFRGiUTBjJLomFFSAuaURKemjEb0k5x11lmQJAn/+c9/wh6zYcMGaDQanHnmmZGsmgDU19cPOt/u6XtMn8WgnrCS8gyVUSJRMKMkOmaUlIA5JdGpKaMRdXerV69GQUEBHnnkEaxfvz7k8hs2bMAjjzyCSZMm4Ze//GUkq6YgHG6vf5oNPBERERERkTpE1N0lJCTgvffew+zZs7FixQqcccYZePXVV1FXVwe32w2Px4O6ujq8+uqrOPPMM3Heeedh9uzZ2Lx5MxISeG12pAoLCwedb3P3OwKv10WrHKIBhsookSiYURIdM0pKwJyS6NSU0bCugdfpQjeBkiTh9ddfx+uvvx50mW3btqGgoAAajQYejyf8SmmA+vr6Qe8lMPHJp3BO1R64TSYYz70PgCHqtREBQ2eUSBTMKImOGSUlYE5JdGrKaFgNvCRJYX1YOMuF+1kUWnd396DzY6p3I6m6EgBgsRijWRJRgKEySiQKZpREx4ySEjCnJDo1ZTSsBv6OO+4Y6zpoBEwm06DztQ4HAMBtMEJviOhBA0QRGSqjRKJgRkl0zCgpAXNKolNTRtnAK1h+fv6g83U/NvAeFQWVlGmojBKJghkl0TGjpATMKYlOTRnlLcoVrLy8fND5OqcTAOAxmaNZDtEAQ2WUSBTMKImOGSUlYE5JdGrKKBt4FTL82MD7zDwCT0REREREpBZs4BUsNTV1wDyXyw29xw0A8Jkt0S6JKMBgGSUSCTNKomNGSQmYUxKdmjIa1jXwzzzzjH/64osvHnT+SPT/LBo+o3HgHeZtbV3+acnCU+hJXoNllEgkzCiJjhklJWBOSXRqymhYDfyll14KjUYDjUYT0HT3zh+Jgz+Lhm/fvn1ITEwMmNfd0e8RCWzgSWaDZZRIJMwoiY4ZJSVgTkl0aspo2M8YG+r57Xyuu1hsHTb/tMbKU+iJiIiIiIjUIqwGfvfu3cOaT9ExadKkAfPcFiv+u2gZDC4ncuaUyFAVUZ/BMkokEmaURMeMkhIwpyQ6NWU0rAY+Ly9vWPMpOpqampCbmxswz5WQgM8XLQUAnHdouhxlEfkNllEikTCjJDpmlJSAOSXRqSmjvAu9gnV1dQ2YZ3f7/NNmgy6a5RANMFhGiUTCjJLomFFSAuaURKemjLKBVzCDwTBgnsPj9U9bDfz1krwGyyiRSJhREh0zSkrAnJLo1JTRiDs8m80Gm8025Pv3338/jjvuOEybNg1Lly7F66+/Hukq6UeTJ08eMM/eaYfRYYfG6+UReJLdYBklEgkzSqJjRkkJmFMSnZoyGlED//rrryMuLg5ZWVno7Owc8P5Pf/pTXH/99fjvf/+L8vJyvP322zjjjDPw5z//OZLV0o/KysoGzHO/+hau/tMvccMd18K45VMZqiLqM1hGiUTCjJLomFFSAuaURKemjEbUwL/99tuQJAmnnXYa4uLiAt775JNP8NRTTwEArFYrZs2aBbPZDEmScPvtt+P777+PZNU0BE9X39kQplg+Ro6IiIiIiEgtImrgP//8c2g0GixcuHDAe48++igAYMKECSgrK8NXX32FHTt2ICcnBz6fD4888kgkqyYAycnJA+Z5bXb/tCnOGs1yiAYYLKNEImFGSXTMKCkBc0qiU1NGI2rgGxsbAQBFRUUD3tu0aRM0Gg2uueYaTJw4EQCQk5ODa665BpIkYcuWLZGsmgBYLAOPsPv6NfCWhNholkM0wGAZJRIJM0qiY0ZJCZhTEp2aMhpRA9/U1AQAA06f/+GHH9Dc3AwAOP300wPeO/zwwwEAe/bsiWTVBKCurm7APKm77xR6SzyPwJO8BssokUiYURIdM0pKwJyS6NSU0YgaeJ2u5y7nLS0tAfM/+eQTAEBaWtqAo/NJSUkAAIfDEcmqaQiSvW+7WuN5BJ6IiIiIiEgtImrgs7OzAQDffPNNwPyNGzdCo9HguOOOGzCmvb0dAJCamhrJqglAfn7+wJn9G/iEmOgVQzSIQTNKJBBmlETHjJISMKckOjVlNKIG/rjjjoMkSXjggQf8p8x/+eWX2LRpEwDg5JNPHjCm9xb+mZmZkayaMPDMBwDQ2PuugY/hKfQks8EySiQSZpREx4ySEjCnJDo1ZTSiBv6qq66CVqvF7t27UVBQgMMPPxzz58+Hx+NBUlISzjvvvAFj3n//fWg0GpSUlESyagLQ0dExYJ7W4QQAuA0G6A36aJdEFGCwjBKJhBkl0TGjpATMKYlOTRmNqIGfPXs27rnnHmg0GnR1deHrr7+Gw+GAwWDAv/71rwE3t2tvb8fGjRsBAAsWLIhk1YS+exD0p3X2nELvMZmjXQ7RAINllEgkzCiJjhklJWBOSXRqymjEh2hvuOEGnHjiidiwYQPq6+uRlZWFFStWDPpouQ8//BBz584FAJx66qmRrnrcG2wbv3vpVXC1dyLNrMOZMtRE1N9gGSUSCTNKomNGSQmYUxKdmjI6KudYz5gxAzNmzAi53Omnnz7gsXI0cmVlZZg2bVrAvH2pmXAlZcCczCPwJL/BMkokEmaURMeMkhIwpyQ6NWU0olPoSV6SJAW89vokuLw988x69ZwmQsp1cEaJRMOMkuiYUVIC5pREp6aMsoFXsMTExIDXdrfXP20x8FdL8js4o0SiYUZJdMwoKQFzSqJTU0ZH/Tbl1dXVaG5uht1uD/lNx/HHHz/aqx9XDr5JYFd7F6b97wu4jSYkayYBmCxPYUQ/OjijRKJhRkl0zCgpAXNKolNTRkelgS8vL8ddd92F1157Lexb9Gs0Gng8ntFY/bhVW1sb8Di+zr2NWPKfZwAArfOPB362QKbKiHocnFEi0TCjJDpmlJSAOSXRqSmjETfwr776KlauXAmHw6GqawuUyNHR7Z/WWC0yVkJERERERESjLaIGvra2FhdeeCHsdjuys7Pxq1/9ClarFf/3f/8HjUaD9957Dy0tLdi2bRueffZZ7Nu3D8ceeyx+97vfqepZfHLJzc0NeG3r18DrrLwLPcnv4IwSiYYZJdExo6QEzCmJTk0ZjehOZ//85z9hs9kQFxeHL774Atdeey3mzZvnf3/hwoU466yz8Oc//xm7du3C+eefj08//RSPP/445s+fH3Hx493Blys4O23+aV2MNdrlEA0Q7iU1RHJhRkl0zCgpAXNKolNTRiNq4N977z1oNBpcddVVmDBhQtBlLRYLnnvuOcyaNQsvvvgi/vOf/0SyagLQ1tYW8NrV1XcEXh/DU+hJfgdnlEg0zCiJjhklJWBOSXRqymhEDXx1dTUA4Oijj/bP02g0/umDb1Kn1Wpx7bXXQpIkPPHEE5GsmhC4rQHA1Wn3TxviYqJdDtEAB2eUSDTMKImOGSUlYE5JdGrKaEQNfHd3zxHfnJwc/zyrte/U7fb29gFjDjnkEADA9u3bI1k1AZg2bVrAa3dX3yn0Rp5CTwI4OKNEomFGSXTMKCkBc0qiU1NGI2rgExISAAAOh8M/LyUlxT9dWVk5YExvU9/c3BzJqgk9j+/rz9Pd18Cb4tjAk/wOziiRaJhREh0zSkrAnJLo1JTRiBr4oqIiAEBVVZV/XlxcHPLy8gAA77zzzoAx7777LgAgMTExklUTAK/XG/i6u+8UejbwJIKDM0okGmaURMeMkhIwpyQ6NWU0oga+947zn3/+ecD8U089FZIk4Z577sEHH3zgn79+/XqsWbMGGo0GxxxzTCSrJgDx8fEBr90aLezWGHj0epjjeQ08ye/gjBKJhhkl0TGjpATMKYlOTRnVSJIkjXTwBx98gBNOOAETJkzAnj17/M92r6mpQUlJCez2niPCycnJcDgcsNlskCQJOp0OH3/8MY466qjR+SlGkcfjwZYtWzBz5kzhn1Xf3d2NmJi+Rv3uD6vxXkUrAODxs4uRk8g70ZO8Ds4okWiYURIdM0pKwJyS6JSQUa/Xi+3bt2P+/PnQ6/VDLhfREfgFCxbgjjvuwKpVq1BXV+efn5ubi3//+99ISEiAJEk4cOAAuru7IUkSTCYT/vWvfwnZvCvNnj17Al7b3T7/tNUw9C+dKFoOziiRaJhREh0zSkrAnJLo1JTRiLo8jUaDO+64Y9D3lixZgl27dmHDhg344Ycf4PF4MGXKFJx77rnIzs6OZLU0BLunr4E3GyL6boaIiIiIiIgEM6aHaVNSUnDFFVeM5SrGtYkTJwa8trv7bs5g1rOBJ/kdnFEi0TCjJDpmlJSAOSXRqSmj7PIUzGazBbye/MI6nLzhGSza9DJ0Wo1MVRH1OTijRKJhRkl0zCgpAXNKolNTRke9gd+7dy+2bduGjz76yH8TOxobLS0tAa+ztn+NQ775AkX/2ypTRUSBDs4okWiYURIdM0pKwJyS6NSU0VFp4Ds7O3HbbbchJycHeXl5OPLII7Fw4ULs3r07YLkXX3wR5557Li6//PLRWC0dROd0AgC8ZpPMlRAREREREdFoi/ga+F27dmHp0qWoqqpC/yfSaTQDT+E+6qijcOGFF0KSJFxyySU49thjI139uFZSUhLw2uBv4M1ylEM0wMEZJRINM0qiY0ZJCZhTEp2aMhrREXiHw4Fly5ahsrISVqsVN910E954440hl8/Pz8fChQsBAK+99lokqyYAO3fu9E+7XW7oPW4AgI8NPAmif0aJRMSMkuiYUVIC5pREp6aMRnQE/qGHHkJFRQViYmLw8ccf47DDDgs5ZsmSJdi8eTM+++yzSFZNADwej3+6u63LPy2xgSdB9M8okYiYURIdM0pKwJyS6NSU0YiOwL/88svQaDS47rrrwmreAWDmzJkAek69p8jExcX5p7s7uvvesFpkqIZooP4ZJRIRM0qiY0ZJCZhTEp2aMhpRA19WVgYAWLx4cdhjUlJSAABtbW2RrJoApKam+qdtHX2PRtBYeASexNA/o0QiYkZJdMwoKQFzSqJTU0YjauC7unpO246NjQ17jPPHG60ZDIZIVk1AwF3+Hf2OwGt4BJ4EcfCTKIhEw4yS6JhRUgLmlESnpoxG1MD3Hk2vrq4Oe8wPP/wAAMjMzIxk1XQQe78GXhfDBp6IiIiIiEhtImrgZ8+eDQD46KOPwh7zzDPPQKPRYN68eZGsmgBMmDDBP+2Ki8f3s+ehfPosYHKBjFUR9emfUSIRMaMkOmaUlIA5JdGpKaMRNfBnn302JEnCo48+ipqampDL33ffff5mf8WKFZGsmtB3OQIAuHMn4p2fXIiN518G3UnzZayKqE//jBKJiBkl0TGjpATMKYlOTRmNqIG/6KKLcOihh8LhcGDBggV46623IEmS/32NRgNJkvDll19i5cqVuPHGG6HRaHDcccdhyZIlERc/3h04cMA/bXf7/NMWfUS/VqJR0z+jRCJiRkl0zCgpAXNKolNTRiN6DrxWq8Vrr72GY489FtXV1Tj11FNhtVqh0WgAAAsWLEBnZ6f/Gw9JklBYWIj169dHXjkFCGjgDToZKyEiIiIiIqKxEPGh2tzcXHzzzTdYsWIFtFoturu7IUkSJElCU1MTHA6H/6j8ueeei61btyI9PT3iwgkoLi72T9s9/Rt4HoEnMfTPKJGImFESHTNKSsCckujUlNGIjsD3Sk5OxvPPP4+77roLGzduxLZt29DY2Aiv14uUlBTMmjULy5cvx9SpU0djdfSjqqoqTJ48GQBgfOJZXPX6JriNJujyfg/kzpW5OqLAjBKJiBkl0TGjpATMKYlOTRkNq4F/7bXXAAAnnHACYmJihlwuLy8PV1111ehURiG5XC7/tK+jC2aHHWaHHSajQcaqiPr0zyiRiJhREh0zSkrAnJLo1JTRsM61PuOMM/CTn/wEe/bsCZj/05/+FD/72c+wf//+MSmOguv/ZYpks/unLXFWOcohGiDYF35EImBGSXTMKCkBc0qiU1NGw75Yuv/d5Xs99dRTeOqpp9Da2jqqRVF4MjIy/NOS3eGfjkmIlaMcogH6Z5RIRMwoiY4ZJSVgTkl0aspoWA28yWQCAHR1dY1pMTQ8VVVVfS/6NfDWBPV8w0TKFpBRIgExoyQ6ZpSUgDkl0akpo2E18NnZ2QCAjz/+eEyLoZHTBByBZwNPRERERESkNmHdxO6EE07Av/71L9xyyy3YunUrpk6dCoOh70ZpDz744IgeDXf77bcPewz1ycrK8k9rHT0NvNtggN4wKg8XIIpY/4wSiYgZJdExo6QEzCmJTk0ZDavT++1vf4uXX34ZBw4cwIYNGwLekyQJDz300IhWzgY+Mh6Pxz+tc/TcxM5jNMtVDtEA/TNKJCJmlETHjJISMKckOjVlNKxT6HNycvD111/jsssuQ35+PgwGAyRJgkajAdDTxI/kH0WmqanJP61zOgEAXrNJrnKIBuifUSIRMaMkOmaUlIA5JdGpKaNh34U+JycHjz76KCorK+FwOODz+fxN/Pfffw+fzzfsf9Fw55134uijj4bVakViYmJU1ikHQ28Db+IReCIiIiIiIjVS/cXSLpcL55xzDubNm4fHH39c7nJG1dSpUwEAXp+EjeetgsHlQmZqnMxVEfXpzSiRqJhREh0zSkrAnJLo1JTRiBr4J598EgAwceLEUSlmLPz+978H0PPMerWpqalBQUEB7G4vKqfNBADEZLOBJ3H0ZpRIVMwoiY4ZJSVgTkl0aspoRA38JZdcMlp1CKezsxNabd8VBiaTCSaTWNeXO36887zd03c5gsUQ9lURRGOuN6NEomJGSXTMKCkBc0qiU1NGVX8K/UhNnz4dNpvN/3rVqlW45pprkJWVhcrKSgBARkYGJElCY2MjAGDKlCnYu3cv7HY7zGYzcnJysGvXLgBAeno6tFot6uvrAQCFhYWor69Hd3c3TCYT8vPzUV5eDgBITU2F0WjEvn37AACTJk1CU1MTurq6YDAYMHnyZJSVlcFms6G+vh4tbp2/ToNGwt69e9HR0QGdToeioiKUlZVBkiQkJiYiLi4OtbW1AIDc3Fx0dHSgra0NGo0G06ZNQ3l5ObxeL+Lj45GUlIQ9e/YA6DnLwmazoaWlBQBQUlKCnTt3wuPxIC4uDqmpqdi9ezcAYMKECXA6nThw4AAAoLi4GFVVVXC5XIiJiUFGRgaqqqoA9DzSwePx+G8sMXXqVNTU1MDhcMBisSA7OxsVFRX+7Q0ADQ0NAIDJkyejrq7Ov71zc3Oxc+dOAEBaWhr0ej32798PACgoKEBDQwO6u7thNBpRUFCAHTt2AABSUlJgMpkCtndzczM6Ozuh1+sxdepUlJaWAgCSk5NhtVqxd+9eAEBeXh5aW1uH3N7x8fGoqakB0HMfic7OziG3d3JyMqqrqwEA2dnZsNvt/u09bdo0VFRUwO12IzY2FmlpaQHb2+Vyobm5GQBQVFSE6upqOJ1OxMTEIDMz05/ZzMxM+Hy+gMzW1tb6t/fEiRMDMqvRaPzbu7CwEPv374fNZoPJZEJeXl7Q7d3Y2AibzYaKiooB29tsNqOurg4AkJ+fj5aWFnR0dAzY3klJSYiNjQ3IbHt7O9rb26HValFcXIwdO3bA5/MhISEBCQkJAdu7q6sLra2tAzI72PZ2OByDZjY2Nhbp6elBM7tnzx44nU5YrVah9hG9mbVYLINub+4j9LDZbCgtLeU+QqZ9RFdX16Dbm/uIvn1Eb0a5j+DfESLvI2w2G3bv3s19BP+OEHYf0bsvFXkfEe5N3jWSAm8H/+tf/xp//etfgy5TVlaG4uJi/+unnnoK119/Pdra2oKO83g82LJlCwoKCoQ/Au9yuWA0GlFW2Yh7nvgQLqMJx8wtwM+XzpC7NCIAfRklEhUzSqJjRkkJ+ufU7XbDbrfD5XIBgP+pVURy8vl8Ab3daOltpXU6HUwmE8xm84jX4/V6sX37dsyfPx96/dDH2RV5BP7GG2/EpZdeGnSZSK9xiIuLg06nC72gjCoqKlBSUoLOH3bi3MfvAwB0Np4OsIEnQfRmlEhUzCiJjhklJejNqc1mg91uR1xcHOLi4ti8kzDsdjssFsuYfb7X6/WfhZGYmAiDwTBm61JkA5+Wloa0tDS5yxCGs7PvVH9djFXGSoiIiIhoPOo98p6cnMzGncYdnU6HmJgYmM1mtLS0IDk5ecwOBiuygR+OmpoatLS0oKamBl6vF9988w2AnmsaYmNj5S0uQr3XaTi7uv3z9LFs4EkcvRklEhUzSqJjRkkJMjIy/Efe2byTiMbyiHh/Op0OsbGxsNlsiIsbm6eDqb6Bv/322/H000/7X8+aNQsA8MEHH2DBggUyVTW6XJ029F5pYYgZu1NDiIiIiIgG43K5xqxhIVISs9mMAwcOjNl/D6p/5thTTz0FSZIG/FND8957h0R3l90/zxAbI1c5RAP0ZpRIVMwoiY4ZJSXozSmPvpOo3G531NY11v8dqL6BHw883X3XwJvjeAo9EREREUUXm3eiPmP53wMbeAWbPHkyAMDX3XcE3sQGngTSm1EiUTGjJDpmlJSAOSXRifY48EiwgVewuro6AIDX1tfAm+N5Cj2JozejRKJiRkl0zCgpAXNKoovmKfRjjQ28gtntPY27r18Db2EDTwLpzSiRqJhREh0zSkrAnJLofD6f3CWMGjbwCmY2m3smbA7/PGsCG3gShz+jRIJiRkl0zCgpAXNKotNq1dP2qv4xcmqWm5sLAPj257/At3taYHQ68FxWqsxVEfXpzSiRqJhREh0zSkqQm5uL9vZ2ucsYN9atW4err74aAPDNN99wPxEGo9EodwmjRj1fRYxDO3fuBAA4vIDHaIItLgEWk0Hmqoj69GaUSFTMKImOGSUlYE5ppGpqapCcnDysfxdeeOGAz/nLX/7if/+TTz4Z8L7D4RgwT6nYwKuA3e0FAJh0Gui0fIQHERERERGRGvEUegVLS0sDANjcPTdlMBt0cpZDNEBvRolExYyS6JhRUgLmlEbD0qVLccstt4RcLj4+ftifrderp+1Vz08yDvUGseSNV1Hg9UGbngpcOEPmqoj6qGlnSerEjJLomFFSAr1er6q7fJM8EhISUFJSMiafrdGo5yxlnkKvYPv37wcAzPjwHcz74C0Uf/KBzBURBerNKJGomFESHTNKSsCckuj4HHgShtvlht7TE0gfH+FBREREREQAmpubceedd2L+/PnIz89HVlYWZs6ciSuvvBKff/55WJ/x+eef4+KLL0ZxcTGysrIwa9YsrF69GlVVVQCA5cuXIzk5GcuXLx/LH4X64XlZClZQUABbh83/WmIDT4IpKCiQuwSioJhREh0zSkpQUFCA7u5uucugft5//32sWrUKnZ2dAfNra2tRW1uL9evX47LLLsNf/vKXIZ+RvmbNGvzhD3+AJEn+eXv27MFTTz2FDRs24KmnnhrLH2FUmUwmuUsYNTwCr2ANDQ3oau/qm2G1yFcM0SAaGhrkLoEoKGaURMeMkhIwp2L57rvvcMEFF6CzsxMGgwE///nP8dprr+G9997Dvffei7y8PADAY489hj/84Q+DfsYrr7yC3//+95AkCUlJSfjd736HTZs2YdOmTfjd734HvV6Pyy67TDG/ezWdQs8j8ArW3d0Ng6Pfr5ANPAmG38aT6JhREh0zSkrQ3d2N2NhYucugH91www1wuVzQ6XR44YUXsGjRIv97s2fPxhlnnIElS5agvLwcDzzwAM477zxMmzbNv4zT6cRvfvMbAEBKSgrefvvtgLOBjjjiCCxduhQnn3wyKioqRq3u9vZ2lJaWhlwuLy8PMTExw/psNd1kkQ28ghmNRtgb+06L0VrYwJNYjEaj3CUQBcWMkuiYUVIC5lQcX331Fb7++msAwMUXXxzQvPdKTEzEP/7xDyxduhQ+nw9PPPEE7rnnHv/7GzduRGNjIwDg5ptvHvRSnsmTJ+Omm27yN/qj4c0338Sbb74ZcrnXXnsNxx577LA+W013oWcDr2AFBQXYVvpf/2ttDK+BJ7Hw2k0SHTNKomNGSQkKCgrQ2toa1rK/eHUHWm2eMa5IXklWPdaeUSzLurds2eKfvvDCC4dc7qijjsLUqVOxc+fOgDH9P0Or1eKcc84Z8jPOPfdc3HLLLQHXyItKTdfAs4FXsB07dsDZ2XcTO32MVcZqiAbasWPHmD3Pk2g0MKMkOmaUlGDHjh3IyMgIa9lWmwfNNvVcjyyasrIyAD1nRcyYMSPosnPmzMHOnTtRWVkJl8vlP5Oi9zPy8/ORkJAw5PikpCTk5+dj9+7do1L7ihUrsHbt2lH5rIM5HA5YVHK2Mht4hXN29W/g1RFKIiIiIlKnJKv62w85f8beMyGSkpKg1wevo/dLF0mS0NbWhvT0dAA916IDPde/h5KSkjJqDTyFR/3/BalYSkoKqmNisLdgKgwuJzIz0+UuiShAODt+IjkxoyQ6ZpSUYDg5levU8vFGTdd8j4ZQX2YoiXp+knHIZDLBPnMmNvw0FQDwm4X58hZEdBA1XW9E6sSMkuiYUVICk8kEr9crdxmEniPvANDS0gKPxxO0ce19BJxGo0FiYqJ/fu9p8wcOHAi5vnCWEYGavtDgc+AVbN++fbB7+h6JYDHw10li2bdvn9wlEAXFjJLomFFSAuZUHL2Pg3O5XPjuu++CLtt7t/rCwsKAJwkUF/ecJVFdXY22trYhx7e2tqK6ujqygqNETc+BZ8encA5337edVjbwRERERETj1vz58/3Tzz///JDLbd26FeXl5QPG9H/t8/mwYcOGIT9j/fr1irgDvdqw41OwSZMmwe7uOwJvNuhkrIZooEmTJsldAlFQzCiJjhklJWBOxTFnzhzMmjULAPDMM88MeEQcAHR0dGD16tUAeh4Vt2rVqoD3ly1bhrS0NADAX//610FvUldZWYm77757tMsfM2q6HInXwCtYc3MzEp56Fhd9+T+4jCboj/0zkMpHyZE4mpubkZOTI3cZRENiRkl0zCgpQXNzM6xW/g0qivvuuw8nnXQSXC4XzjvvPFx++eU45ZRTYLVa8e2332LNmjX+U9+vvvrqAY+qNJvNuOuuu3D55ZfjwIEDOOmkk3Dttddi3rx5AIDPPvsMa9asgc/nQ2FhISorK0flGvP29naUlpaGXE6n06GoqGhYn+3xeAIuE1AyNvAK1tnZCf3+BiTX1wEAzDyFngTT2dkpdwlEQTGjJDpmlJSgs7OTDbxAZsyYgXXr1mHVqlXo7OzE2rVrB32++mWXXYbbb7990M8466yzUF1djbvuugstLS343e9+F/C+1WrFk08+ifvuuw+VlZWjcoT7zTffxJtvvhlyufj4+GFfe6+mmyyy41MwvV4P2O3+1zGJcTJWQzSQmh7ZQerEjJLomFFSAuZUPIsWLcJXX32F1atXY8aMGYiLi4PJZMLEiRNxzjnn4M0338Tdd98NrXbodvDGG2/EG2+84T+l3mQyIScnBytXrsTmzZtx0kkn+b9kjI+Pj9aPNiJqugu9RuKdBwJ4PB5s2bIFM2fOhE4n/jXlLxx9MZKqKgAAJ9Z+BL2BO1AiIiIiiq4DBw4M63nwpHxutxv5+fmw2+248cYbceutt8pdkjBG8t+D1+vF9u3bMX/+/KBfivEIvIKVlpZC63AAANwGA5t3Ek441zERyYkZJdExo6QEzOn4tHHjRth/PBt47ty5MlcTnL3fWctKxwZe4XTOngbeYzTLXAkREREREalFVVXVkO/V1NTgt7/9LQAgPT0dixYtilZZ4x4P2SpYcnIydA4nAMBrVs+jEUg9kpOT5S6BKChmlETHjJISMKfqdOSRR+Kkk07C4sWLUVxcjJiYGDQ1NeGTTz7Bk08+ifb2dgDAH/7wB+HvgyB6fcOhnp9kHLJardC7eo7Ae808Ak/i4R1pSXTMKImOGSUlsFqtcLvdcpdBo8zr9WLTpk3YtGnToO9rtVrceuutOPfcc6Nc2fAFu1mf0qjnJxmHqqv3wPDjztLHBp4EtHfvXrlLIAqKGSXRMaOkBMypOr3wwgv42c9+hhkzZiAzMxNGoxGxsbGYMmUKVq1ahY8++gg33HCD3GWGxeVyyV3CqOEReAVzdjv80xIbeCIiIiIiGiUnn3wyTj75ZLnLoIOwgVewtOwcbFh2NgxOJ7KmTJS7HKIB8vLy5C6BKChmlETHjJIS5OXlweFwhF6QSCZGo1HuEkYNG3gFa7I58L95CwEAJ07hzUNIPK2trYiJiZG7DKIhMaMkOmaUlKC1tRUWi0XuMoiG5PV6odPp5C5jVPAaeAU70NHln7bo+ask8XR0dMhdAlFQzCiJjhklJWBOSXRer1fuEkYNuz4F80h9vz6Lgb9KEo9avukk9WJGSXTMKCkBc0qi02g0cpcwatj1KVhKfBLiWw/AbOuCWSPJXQ7RAEVFRXKXQBQUM0qiY0ZJCZhTEp1ZRTf8ZgOvYHtffQuX/f12XHXXzYh/c/DnMxLJqaysTO4SiIJiRkl0zCgpAXNKorPb7XKXMGrYwCuYx+b0TxtirDJWQjQ4SeKZISQ2ZpREx4ySEjCnRNHDBl7BtJ6+mzEY4tjAk3gSExPlLoEoKGaURMeMkhIwpyQ6Nd2ngQ28gmmcbv+0KZYNPIknPj5e7hKIgmJGSXTMKCkBc0qiYwNPQrC39j2ywxTPZ8SSeGpqauQugSgoZpREx4ySEjCnJDqXyyV3CaOGDbySOfpuxmBhA09ERERERKRqbOAVTO/uuwbewmvgSUA5OTlyl0AUFDNKomNGSQmYUxKd0WiUu4RRwwZewXy2viPw1gQegSfxdHZ2yl0CUVDMKImOGSUlYE5JdF6vN/RCCsEGXsn6NfAxCbEyFkI0uLa2NrlLIAqKGSXRMaOkBMxpdK1btw7JyclITk7m/QfCpKYGXi93ATRyWmffc+Ct8TyFnsSj0WjkLoEoKGaURMeMkhIwpzRafD4f3n77bbzzzjvYunUrmpqa0NbWhtjYWGRlZWHWrFlYsmQJFi9eDIPBMGD88uXL8emnnw5rnbt370ZCQoL/dU1NDQ477DAAwIoVK7B27dqIfqbRxgZewbZedhUa61th8bpxinFggInkNm3aNLlLIAqKGSXRMaOkBNOmTcOBAwfkLoMU7tNPP8XNN9+M0tLSAe+1tbWhra0NZWVlWLduHbKysnDrrbfiggsukKFSebGBV7BGkxUHMixIMPPXSGIqLy9HUVGR3GUQDYkZJdExo6QE5eXlSE1NlbsMUrDnn38eq1evhtvtBgAcfvjhOPXUUzFjxgwkJyejo6MDu3fvxnvvvYd33nkH+/fvxy233BK0gf/kk0/80y6Xa8gb2cXFxY3uDzPG2PkpmMMjAQAsBt7KgMSkpuuNSJ2YURIdM0pKwJxSJLZs2YLrrrsOPp8PMTExeOCBB3D66acPWO64447DxRdfjJqaGvz+97/H5s2bg35uSUmJf9put8NisYx67XJgA69gLl/P/1r0bOBJTPHx8XKXQBQUM0qiY0ZJCZhTGimbzYYrr7wSPp8PWq0WL774Io455pigY3Jzc/H444/jxRdfDHs9Op0u0lKFwc5PoTxuD6Z/8gGmb/svsivK5S6HaFDJyclyl0AUFDNKomNGSQmYUzE1NzfjzjvvxPz585Gfn4+srCzMnDkTV155JT7//POwPuPzzz/HxRdfjOLiYv9N5FavXo2qqioAPTeNS05OxvLly0dU4/PPP4+GhgYAwGWXXRayee/v/PPPD3tZNvAku+4OGxa+uQGLX30ek995U+5yiAZVXV0tdwlEQTGjJDpmlJSAORXP+++/jzlz5uDvf/87vvvuO3R0dMDpdKK2thbr16/H0qVLcdNNN8Hn8w35GWvWrMGyZcvwxhtvoLGxEU6nE3v27MFTTz2FBQsW4P3334+4znXr1gHoeZLBFVdcEfHnDcXlco3ZZ0cbT6FXqO62zr4XVnVcz0FERERERJH57rvvcMEFF8DlcsFgMOCyyy7DkiVLYLVa8e2332LNmjXYs2cPHnvsMVitVvzud78b8BmvvPIKfv/73wMAkpKScN111+Goo44C0HNU/r777sNll10W0c0LOzo68N133wEAJk+ejEmTJo34s8YTNvAKZWvv7nthMctXCFEQ2dnZcpdAFBQzSqJjRkkJsrOz4fF45C6DfnTDDTfA5XJBp9PhhRdewKJFi/zvzZ49G2eccQaWLFmC8vJyPPDAAzjvvPMCHlnpdDrxm9/8BgCQkpKCt99+GwUFBf73jzjiCCxduhQnn3wyKioqRlxnaWmp/wyA3ueuj6b+j6Pzer2DnkafmJiICRMmjPq6xxIbeIWyd9j801oegSdB2e12JCQkyF0G0ZCYURIdM0pKYLfbYTAY5C6DAHz11Vf4+uuvAQAXX3xxQPPeKzExEf/4xz+wdOlS+Hw+PPHEE7jnnnv872/cuBGNjY0AgJtvvjmgee81efJk3HTTTf5GfyRaW1v902PxGMJjjz025DIrVqzA2rVrR33dY4kNvELZO/uOwOvYwJOgWlpakJmZKXcZRENiRkl0zCgpQUtLCzIyMsJefvfDL6D6kdB3EI+fUYQ5z9wdMO+ri29Cx3ehb+Ccf8X5mHTlCv9rT1c3Pj5u6GeG9zf7qb8iYWax/3XjO5/ih5vvDjKihz7GguM+Cf/O6GNhy5Yt/ukLL7xwyOWOOuooTJ06FTt37gwY0/8ztFotzjnnnCE/49xzz8Utt9wCSZJGVGtXV5d/2mq1jugzxiM28Arl7Ohr4LUxbOCJiIiISBk8nd1w7m8KuZxrQvrAeQdawxrr6XewCwAgIaxxAOBzuwNfO5zhrTNW/ia0rKwMAGA0GjFjxoygy86ZMwc7d+5EZWUlXC4XjEZjwGfk5+cHPQMoKSkJ+fn52L1794hqjY2N9U/bbLYgS45MS0uLf5rPgSfZObv6Qq6PkX9nQTSY/tdTEYmIGSXRMaOkBNOmTQtolkLRx8XAlJUWcjljStKg88IZq4+LCZyhQVjjAEB70OUAWrMpvHUKcFCt97T0pKQk6PXBW73esyYkSUJbWxvS03u+MGlvbwfQc/17KCkpKSNu4JOS+n6/TU3hfbkyUmazeu4ZxgZeodydNmh+nDYK8G0f0WAqKiowZcoUucsgGhIzSqJjRkkJKioqhvUs+ElXrgg4vX04Dj6lPlz62Bgs/N//G9HY9MXHIH3xyMbKRaPRhF5IZiUlJdBqtfD5fNi+ffuYrsvpdKqmiedz4BXK1e8IvCGODTyJyX3QKWhEomFGSXTMKCkBcyqO3qPaLS0tIZ8M0NDQAKCn2U9MTPTP7z1t/sCBAyHXF84yQ4mPj/ef5l9RUYGqqqoRf1YoI71OX0Rs4BXKZTCiLTkV3bFxMCXGyV0O0aD6X9tEJCJmlETHjJISMKfi6L3sxuVy+Z+xPpTeu9UXFhb6r38HgOLinhv4VVdXo62tbcjxra2tqK6ujqjeCy7oubGgJEl49NFHI/qsYLRa9bS96vlJxpn2k07AE6t/j0d+/RfEH3eE3OUQDSotLbxrzYjkwoyS6JhRUgLmVBzz58/3Tz///PNDLrd161aUl5cPGNP/tc/nw4YNG4b8jPXr10d8ZHvlypX+a/Efe+wxfPrpp2GPfemll8JeVk2POWQDr1B2t88/bTHoZKyEaGgjvakJUbQwoyQ6ZpSUgDkVx5w5czBr1iwAwDPPPDPgEXEA0NHRgdWrVwPoOTK9atWqgPeXLVvm/1Lmr3/966C/38rKStx998juR9Cf1WrFww8/7L8W/vzzz8drr70WdMzevXtx2WWX4eabbw57PU6nM9JShcGb2CmU3dO/gef3MEREREREBNx333046aST4HK5cN555+Hyyy/HKaecAqvVim+//RZr1qzxn/p+9dVXo6SkJGC82WzGXXfdhcsvvxwHDhzASSedhGuvvRbz5s0DAHz22WdYs2YNfD4fCgsLUVlZGdFN8+bPn481a9Zg9erV6O7uxqWXXorDDz8cp512GmbMmIGkpCR0dHSguroamzdvxqZNm+B0OhEfHx/0c0tLS/3TTqcTJpNp0OVycnIQF6ecS5LZwCuUw+31T7OBJ1FNmDBB7hKIgmJGSXTMKCnBhAkT4PV6Qy9IUTFjxgysW7cOq1atQmdnJ9auXYu1a9cOWO6yyy7D7bffPuhnnHXWWaiursZdd92FlpYW/O53vwt432q14sknn8R9992HysrKIZvjcK1cuRL5+fm46aabUFZWhm3btmHbtm1DLj9x4kTccccdQT/z2GOPDWvdzz77LJYtWzaseuXEBl6hMp9fh9N318JtNMF47l8AGEOOIYo2l8sldwlEQTGjJDpmlJTA5XJBp+MlnSJZtGgRvvrqKzz88MN49913UV1dDZfLhbS0NMybNw+rVq3CUUcdFfQzbrzxRhx99NF48MEHsXXrVnR0dCA9PR3HH388rr76ahQVFeGPf/wjAIQ8Gh6OY445Bh9//DHefvttvP3229i6dSuamprQ3t6OmJgYTJgwAbNnz8bSpUtx0kknhXzOvVppJDXdU38UeDwebNmyBTNnzhR6R/TC0RcjqaoCAHBCzRYYjOq5MQOpR2lp6YDTsohEwoyS6JhRUoLS0lJkZGQgJSVF7lIoitxuN/Lz82G323HjjTfi1ltvlbukIdntdlgslqit78CBA8P+78Hr9WL79u2YP39+0C8neO61QmkdDgCAR29g805ERERERFG1ceNG2O12AMDcuXNlrmb8YAOvUDpnTwPvjvB6E6KxVFRUJHcJREExoyQ6ZpSUgDlVp6qqqiHfq6mpwW9/+1sAQHp6OhYtWhStskbEbDbLXcKoGZ8XDqiAztHzKASvisJI6lNdXY3CwkK5yyAaEjNKomNGSQmqq6uRmJgodxk0yo488kicdNJJWLx4MYqLixETE4OmpiZ88sknePLJJ9He3g4A+MMf/iD89ehOp1M1TbzYW5qGZPjxCLzXpI4gkjqp6ZmbpE7MKImOGSUlYE7Vyev1YtOmTdi0adOg72u1Wtx6660499xzo1zZ8Knptm9s4BXI7XJD73EDAHwq+SaJ1CkmJkbuEoiCYkZJdMwoKQFzqk4vvPAC3nvvPf/d4FtaWmA0GpGVlYVjjz0WP/vZzxRzk02tVj1Xjqu6ga+ursYf//hHvP/++6ivr8eECRNw4YUX4tZbb4XRqNzHrtk6bP5pycIGnsSVmZkpdwlEQTGjJDpmlJQgMzMTXV1dcpdBo+zkk0/GySefLHcZo8JgUM9Nv9XzVcQgduzYAZ/Ph0ceeQQ//PAD/vGPf+Dhhx/GLbfcIndpEelu77eDZANPAqusrJS7BKKgmFESHTNKSsCckujUdJmHqo/An3LKKTjllFP8rwsKClBeXo6HHnoIf/vb32SsLDK29m7/tCaKzzMkIiIiIiIi+ai6gR9Me3s7kpOTQy7X2dkZcK2EyWSCSZBHtrksFmw75gQYXE5kHDpd7nKIhsRTP0l0zCiJjhklJcjMzFTVTcJIfdR0Cv24auArKipw//33h3X0ffr06bDZ+q41X7VqFa655hpkZWX5TxPKyMiAJElobGwEAEyZMgV79+6F3W6H2WxGTk4Odu3aBaDn+YharRb19fUAgMLCQtTX16O7uxsmkwn5+fkoLy8HAKSmpsJoNGLfvn0AgEmTJqGpqQldXV0wGAzwJWVh66k/gdMLLC2MQXt7O+rq6gAA+fn5aGlpQUdHB3Q6HYqKilBWVgZJkpCYmIi4uDjU1tYCAHJzc9HR0YG2tjZoNBpMmzYN5eXl8Hq9iI+PR1JSEvbs2QMAmDhxImw2G1paWgAAJSUl2LlzJzweD+Li4pCamordu3cDACZMmACn04kDBw4AAIqLi1FVVQWXy4WYmBhkZGT4nyuZlZUFj8eDpqYmAMDUqVNRU1MDh8MBi8WC7OxsVFRU+Lc3ADQ0NAAAJk+ejLq6Ov/2zs3Nxc6dOwEAaWlp0Ov12L9/P4Cesy8aGhrQ3d0No9GIgoIC7NixAwCQkpICk8kUsL2bm5vR2dkJvV6PqVOnorS0FACQnJwMq9WKvXv3AgDy8vLQ2to65PaOj49HTU0NACAnJwednZ1Dbu/k5GRUV1cDALKzs2G32/3be9q0aaioqIDb7UZsbCzS0tICtrfL5UJzczOAnmexVldXw+l0IiYmBpmZmf7MZmZmwufzBWS2trbWv70nTpwYkFmNRuPf3oWFhdi/fz9sNhtMJhPy8vKCbu/Gxka0tLSgpaVlwPY2m82DZvbg7Z2UlITY2NiAzLa3t6O9vR1arRbFxcX+S2USEhKQkJAQsL27urrQ2to6ILODbW+HwzFoZmNjY5Genh40s3v27IHT6YTVahViHzF58mSUlZX5M2uxWLiPGGIfUVNTg/r6eu4jZNpHdHV1Dbq9uY/o20f0ZpT7CP4dIfI+oqOjA6mpqfD5fP5TlfV6PTQaDdzunhsvm0wmuN1u+Hw+aDQamEwmOByOiJc1Go3wer3wer3QaDQwm82w2+0AAJ1OB51OB5fLNWBZALBYLMNa1uFwQJKkAcsaDAZIkgSPxwOg55njTqcTkiRBq9XCYDD4t0sky/bfLgcvq5TtPdQ2NBqN8Pl8/p/14GX1en1E29DtdsPtdsNsNsPlcg26DXub/MG2i1arhdFoDHsbSpLk/+8+3H1EuF+CaSQFfl3261//Gn/961+DLlNWVobi4mL/67q6OsyfPx8LFizAY489NuQ4j8eDLVu2oKCgQNgj8L1++OEHFE8rgU6rkbsUokGVlpYq5u6kND4xoyQ6ZpSUoLS0FOnp6UhNTZW7FKJB2e12WKJ46XFzc/Ow/3vwer3Yvn075s+fD71+6OPsijwCf+ONN+LSSy8NukxBQYF/et++fVi4cCGOPvpoPProo2GtIy4uDjqdLpIyx5xGo2HzTkRERERENE4osoFPS0tDWlpaWMvW1dVh4cKFmDNnDp588klVPQNwypQpcpdAFBQzSqJjRkl0zCgpwZQpU9DV1QWv1yv8ATAan6J5JrXb7Q56BD1S6ulmB1FXV4cFCxYgNzcXf/vb39DU1IT6+nr/9WNK13sNGpGomFESHTNKomNGSQlqa2sDrpsmEk3vterRMNan6yvyCHy43n33XVRUVKCiogITJ04MeE+Bl/4PwJ0kiY4ZJdExoyQ6ZpSUwOFwwGw248CBAzCbzTwKT8Lx+XxRWY/T6YTL5UJcXNyYrUPVR+AvvfRSSJI06D81iOaNGIhGghkl0TGjJDpmlJTAYrFAq9UiMTERLS0tsNvtqvl7m9RhrC+jdrvd6OjoQGdnJ5KTk6HRjN19ylR9BF7tDj6rgEg0zCiJjhkl0TGjpAS9OTUYDEhOTobNZvM/BnAsGxmicPU+Cm609X5RpdfrYbFYEBcXN+aZZwOvYLt27eKjZUhozCiJjhkl0TGjpAT9c6rT6RAXFzempxATDZeaHsmp6lPoiYiIiIiIiNSCDbyCpaeny10CUVDMKImOGSXRMaOkBMwpiU5NGWUDr2C8pohEx4yS6JhREh0zSkrAnJLo1JRRNvAK1tDQIHcJREExoyQ6ZpREx4ySEjCnJDo1ZZQNPBEREREREZECsIFXsMLCQrlLIAqKGSXRMaMkOmaUlIA5JdGpKaNs4BVs//79cpdAFBQzSqJjRkl0zCgpAXNKolNTRtnAK5TT6cT9998Pp9MpdylEg2JGSXTMKImOGSUlYE5JdGrLKBt4hXI6nXjyySdVE0RSH2aURMeMkuiYUVIC5pREp7aMsoEnIiIiIiIiUgA28EREREREREQKoJe7ANFIkgQA8Hq9MlcSnM/ng9Vqhc/nE75WGp+YURIdM0qiY0ZJCZhTEp1SMtpbW28/OhSNFGqJccbhcODTTz+VuwwiIiIiIiIaZ4455hiYzeYh32cDfxCfzweXywWdTgeNRiN3OURERERERKRykiTB6/XCaDRCqx36Snc28EREREREREQKwJvYERERERERESkAG3giIiIiIiIiBWADT0RERERERKQAbOAVau3atcjPz4fZbMaRRx6JrVu3yl0SjVMfffQRli9fjgkTJkCj0eDVV18NeF+SJNx+++3IysqCxWLBiSeeiF27dslTLI1Lf/7znzF37lzExcUhPT0dZ/z/9u4+KKrz7OP4dxWhiIKBIARFRKOQSKEBolK1KlqojS9RE1PHFjGkWoodkaTNtE4LaVLjNDOJZqxAjQk2jVHROgpNNIoRRQkiiCI1pFgtTQXxDaoYomHP84fDPq4u4Au4rv4+MzvD3vd9zrmWuWZnrj3nXOfpp6msrLRa09TURFJSEl5eXvTo0YPp06dz6tQpO0UsD5r09HRCQ0Nxd3fH3d2dqKgoPv74Y8u88lPuJUuWLMFkMpGcnGwZU46KvaWlpWEymaxewcHBlvn7KUdVwDugdevWkZKSQmpqKqWlpYSFhREbG0tdXZ29Q5MHUGNjI2FhYfzpT3+yOf/HP/6Rt99+m4yMDIqKinBzcyM2Npampqa7HKk8qPLz80lKSuKzzz5j+/btXLlyhZiYGBobGy1rFi5cSE5ODtnZ2eTn53Py5EmmTZtmx6jlQdK3b1+WLFlCSUkJBw4cIDo6milTplBRUQEoP+XeUVxcTGZmJqGhoVbjylG5FwwZMoSamhrLq6CgwDJ3X+WoIQ5n6NChRlJSkuV9c3Oz4efnZ7z++ut2jErEMABj06ZNlvdms9nw9fU13njjDctYfX294eLiYnz44Yd2iFDEMOrq6gzAyM/PNwzjak5269bNyM7Otqw5evSoARiFhYX2ClMecA899JDxzjvvKD/lnnHhwgVj0KBBxvbt243Ro0cbCxYsMAxD36Fyb0hNTTXCwsJszt1vOaoz8A7m8uXLlJSUMH78eMtYly5dGD9+PIWFhXaMTORGx48fp7a21ipfPTw8GDZsmPJV7KahoQEAT09PAEpKSrhy5YpVngYHB9OvXz/lqdx1zc3NrF27lsbGRqKiopSfcs9ISkriqaeesspF0Heo3Dv++c9/4ufnx4ABA5g1axbV1dXA/ZejTvYOQG7NmTNnaG5uxsfHx2rcx8eHzz//3E5RidhWW1sLYDNfW+ZE7iaz2UxycjIjRowgJCQEuJqnzs7O9OrVy2qt8lTupvLycqKiomhqaqJHjx5s2rSJxx9/nLKyMuWn2N3atWspLS2luLj4hjl9h8q9YNiwYWRlZREUFERNTQ2vvPIKo0aN4siRI/ddjqqAFxGRB0ZSUhJHjhyxui9O5F4QFBREWVkZDQ0NbNiwgdmzZ5Ofn2/vsET4z3/+w4IFC9i+fTvf+ta37B2OiE0TJkyw/B0aGsqwYcMICAhg/fr1uLq62jGyjqdL6B3Mww8/TNeuXW/omnjq1Cl8fX3tFJWIbS05qXyVe8H8+fPJzc3l008/pW/fvpZxX19fLl++TH19vdV65ancTc7Ozjz66KNERETw+uuvExYWxrJly5SfYnclJSXU1dURHh6Ok5MTTk5O5Ofn8/bbb+Pk5ISPj49yVO45vXr1YvDgwVRVVd1336Mq4B2Ms7MzERER5OXlWcbMZjN5eXlERUXZMTKRGwUGBuLr62uVr//73/8oKipSvspdYxgG8+fPZ9OmTezcuZPAwECr+YiICLp162aVp5WVlVRXVytPxW7MZjNff/218lPsbty4cZSXl1NWVmZ5RUZGMmvWLMvfylG511y8eJFjx47xyCOP3Hffo7qE3gGlpKQwe/ZsIiMjGTp0KEuXLqWxsZE5c+bYOzR5AF28eJGqqirL++PHj1NWVoanpyf9+vUjOTmZ1157jUGDBhEYGMhvf/tb/Pz8ePrpp+0XtDxQkpKSWLNmDZs3b6Znz56W+908PDxwdXXFw8ODhIQEUlJS8PT0xN3dnV/84hdERUUxfPhwO0cvD4Jf//rXTJgwgX79+nHhwgXWrFnDrl272LZtm/JT7K5nz56WniEt3Nzc8PLysowrR8XeXnrpJSZNmkRAQAAnT54kNTWVrl27MnPmzPvue1QFvAN67rnnOH36NL/73e+ora3lO9/5Dlu3br2hUZjI3XDgwAHGjh1reZ+SkgLA7NmzycrK4le/+hWNjY3MnTuX+vp6Ro4cydatW3Ufndw16enpAIwZM8Zq/L333iM+Ph6At956iy5dujB9+nS+/vprYmNjWbFixV2OVB5UdXV1xMXFUVNTg4eHB6GhoWzbto3vf//7gPJT7n3KUbG3L7/8kpkzZ3L27Fm8vb0ZOXIkn332Gd7e3sD9laMmwzAMewchIiIiIiIiIm3TPfAiIiIiIiIiDkAFvIiIiIiIiIgDUAEvIiIiIiIi4gBUwIuIiIiIiIg4ABXwIiIiIiIiIg5ABbyIiIiIiIiIA1ABLyIiIiIiIuIAVMCLiIiIw0tLS8NkMmEymewdioiISKdRAS8iItLJCgsLMZlMuLm58c0331jG6+vr6dq1KyaTierqajtGKCIiIo5ABbyIiEgn27t3LwDDhg3DycnJatxsNuPv70+/fv3sFZ6IiIg4CBXwIiIinaylgB85cqTV+J49e2yOy61LS0vDMAwMw7B3KCIiIp1GBbyIiEgn27dvH3BjoV5QUGBzXERERMQWk6GfqkVERDpNVVUVgwYNomvXrpw/f56ePXsC0NTUhIeHB5cvX+bQoUOEhobaOVIRERG51+kMvIiISCdquXw+NDTUUrwD7N+/n8uXL9OrVy9CQkI67HhfffUVixcvJiwsDDc3N7y8vBgxYgQrV67EbDaza9cuS7f2Xbt23bB9//79MZlMxMfHt3mc+Ph4TCYT/fv3b3NdbW0tixYtIjIyEk9PT1xcXPD392fGjBns2LGj1e1OnDhhiTMrKwuAv/3tb/zwhz/Ez88PJycnxowZY1l/s13om5qaWL58OePGjcPX1xdnZ2d69+7N+PHjWbVqlVWTQVt27tzJzJkzCQwMxNXVle7duxMQEMDw4cN56aWX2LlzZ5vbi4iI3Amn9peIiIjIzcjKymLOnDk25w4ePGizuGzpRH+t48ePt1sY21JbW0t0dDRHjx61jF26dIl9+/axb98+Nm7cSEpKyi3v93Z98MEHzJs3j8bGRqvxL7/8kuzsbLKzs0lISCAjI8Oqud/1DMMgLi6O999//47iOXToEFOmTOHf//631fjp06fJy8sjLy+PzMxMcnJy8PHxuWH7hQsXsnTp0hvGq6urqa6upqioiKysLM6cOXNHcYqIiLRGBbyIiMh94JtvvmHixImW4j0mJobExET8/f2prq5mxYoVbNu2jXPnzt2VeNavX89PfvITDMNgwIABzJ8/n8cffxxvb29OnDjBqlWr+Oijj1i1ahXu7u68+eabre5r6dKlHD58mFGjRpGYmMjgwYOpr6/nxIkTNx1PVVUVo0ePpqGhAXd3d5KSkhg6dCj+/v6cPXuWLVu2kJmZSXFxMVOmTGHPnj1069bNsn1ubq6leA8NDSUxMZHHHnsMDw8P6uvrqaioYMeOHezfv/92/2UiIiLtUgEvIiLSQaZOncrw4cMt72traxk7dixwtWGdl5cXAGazmYiICJqamtiwYQNDhgyx2k+fPn1u+diZmZmUlJQAMHfuXDIzMy1zERERTJ06lYSEBN59991b3vetOnPmDHPnzsUwDJ5//nkyMzOtzrCHh4czbdo0Fi1axOLFi1m2bBnz5s0jKCjI5v4OHz5MXFwcWVlZ7V4i35rZs2fT0NDAE088wSeffMLDDz9sNR8TE8PEiRN56qmnLGfSf/rTn1rm169fD0BAQAB79+6lR48eVtuPGTOGpKSku/YDiYiIPJh0D7yIiEgH8fDwIDg42PI6ffo0AIGBgYwYMcIyfuXKFZqamnB1dWXy5MlW2wQHB1ud+b1ZK1asAMDHx4e33nrL5pply5bh7e19+x/wJqWnp9PQ0ECfPn1YsWJFq5fHv/LKK/Tp0wez2cxf/vKXVvfXq1cvli9fftvF+549eyxPAli9evUNxXuLH/zgBzzzzDMAlvvuW9TW1gJXf3y4vni/lqen523FKCIicjNUwIuIiHSS1p7z3tLYLjIy8raK9evV1NTwj3/8A4AZM2bQvXt3m+t69OjBjBkz7vh47dmyZQsAEydOxMXFpdV1Tk5OREVFAVBYWNjqukmTJlk1ALzdeIKCgvj2t7/d5trvfe97ABQXF1s1tHvkkUcA2L17N8eOHbvtWERERO6ECngREZFO0tpz3lvOBo8YMaJDjlNeXm75+8knn2xz7dChQzvkmK1pbm6mrKwMuHpZf0tn+NZeGzZsAP7/DLctd/qIvQMHDgBQWVnZbjzz588H4MqVK1aXw8fFxQFw9uxZQkJC+NGPfsR7771HVVXVHcUmIiJyK1TAi4iIdIILFy5w+PBhoPUz8N/97nc75FjXFpq9e/duc62t7uod6dy5c+0+is2WS5cutTr30EMP3UlI1NXV3dZ218Y0btw4li9fjqurK01NTaxbt47nn3+eQYMG0bdvX372s59x6NChO4pTRESkPWpiJyIi0gH69+9/w+PJWlzfpK7F5MmTrd6npqaSlpZ2R3Hc7n3iHaW5udny9wsvvMCCBQtuajtnZ+dW565/zN7txhQWFsZf//rXm97u+maCSUlJPPvss6xZs4bt27ezd+9eGhoa+O9//0tmZiZ//vOf+c1vfsNrr712R/GKiIi0RgW8iIiIg7v2DPWpU6faXNvefJcuVy/OM5vNba67/tnuLa5t4mYYBiEhIW3u525o6f5/8eLFO46nd+/eJCcnk5ycjNlspqysjE2bNrF8+XLq6+v5wx/+wJNPPsmUKVM6InQREREruoReRESkA3zyySeUl5dbXhEREQAsWrTIanzq1KkAzJo1y2q8vLycn//857d17GsbsxUXF7e5tr35lmZx58+fb3PdF198YXPc2dnZcsVBy60C9vbEE08A8K9//avNe+1vVZcuXQgPD+fVV18lLy/PMt7yyDkREZGOpgJeRESkAwwePJiQkBBCQkIICgri6NGjAEybNs0yHhISQmVlJXC1Q/u14yEhIe3ev94aPz8/HnvsMQCys7P56quvbK5rbGxst7gMDAwEoLS0FMMwbK6pqKiw3N9vS8utAZ9//jnbtm1rN/7O1hKPYRgsW7asU44RHh5uuRLizJkznXIMERERFfAiIiIdrKSkhEuXLtGzZ0/CwsIs42fPnrUU9i2PK+soiYmJwNVu7i+++KLNNQsXLmy3odvo0aMBOHnyJB9++OEN8xcuXCAhIaHNfSxYsMDyrPQ5c+ZQUVHR5vq///3vbf4gcKdiYmIs3fffeOONdn/EKC8vJycnx2ps3bp1rf4wAlc73bdctdDyI4iIiEhHUwEvIiLSwXbv3g1c7TJ/bQO2goICDMPg0Ucfxc/Pr0OPmZiYaLlUPD09nQkTJrB582ZKS0vZvHkzsbGxrFy5ksjIyDb38+Mf/xh3d3cAEhIS+P3vf09RURH79+8nPT2d8PBwDh06ZDmWLT4+PqxevRqTyURNTQ2RkZEkJiayZcsWSktLKSoqYuPGjbz88ssMHDiQiRMnUl1d3XH/DBvWrFmDp6cnzc3NPPfcc0yePJkPPviA/fv3U1JSwscff8zixYuJiooiNDSU/Px8q+1ffvll/Pz8iI+P591336WgoICDBw+yY8cO0tLSiI2NBa423HvhhRc69bOIiMiDS03sREREOlhLAT9q1Cir8T179gAdf/YdwMnJidzcXKKjo6msrGTr1q1s3brVak1MTAwvvviipdi0xdvbm3feeYeZM2fS1NREamoqqamplnlXV1fef/99cnNzOXjwYKv7mTZtGps3byY+Pp5z586RkZFBRkaGzbVdunTBzc3tFj/xrRk4cCCFhYVMnz6dI0eOkJOTc8NZ9mu1/Ihxrfr6elavXs3q1attbuPi4kJGRka7P5KIiIjcLhXwIiIiHchsNluat93NAh6u3gt/8OBB3nzzTdauXcuxY8dwcXEhODiYuLg45s2bZ/lxoS3PPvssAQEBLFmyhIKCAhoaGvDx8SE6Oppf/vKXDBkyhNzc3Hb3M2nSJI4fP87KlSv56KOPqKio4Ny5czg5OeHr68uQIUOIjo7mmWeewd/fvyP+BW0aPHgwZWVlrF+/no0bN1JcXMzp06dpbm7Gy8uLoKAgRo4cydSpUwkPD7fa9tNPPyUnJ4fdu3fzxRdfUFtby/nz5+nevTsDBw5k3LhxJCYmMmDAgE7/HCIi8uAyGa11qBEREZH7zq5duxg7dixwtSgdM2aMfQMSERGRm6Z74EVEREREREQcgAp4EREREREREQegAl5ERERERETEAaiAFxEREREREXEAKuBFREREREREHIC60IuIiIiIiIg4AJ2BFxEREREREXEAKuBFREREREREHIAKeBEREREREREHoAJeRERERERExAGogBcRERERERFxACrgRURERERERByACngRERERERERB6ACXkRERERERMQBqIAXERERERERcQD/B304BOyha2OYAAAAAElFTkSuQmCC",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 1200x600 with 1 Axes>"
       ]
@@ -568,7 +568,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.7"
   }
  },
  "nbformat": 4,
diff --git a/docs/source/guides/Fitting_With_Priors.ipynb b/docs/source/guides/Fitting_With_Priors.ipynb
index 53ff3b2..34a7984 100644
--- a/docs/source/guides/Fitting_With_Priors.ipynb
+++ b/docs/source/guides/Fitting_With_Priors.ipynb
@@ -184,7 +184,7 @@
    "outputs": [],
    "source": [
     "def model_fn(params, x_train, y_train):\n",
-    "    return models.gaussian_process(\n",
+    "    return models.gaussian_process.exact(\n",
     "        models.means.zero(),\n",
     "        models.kernels.scaled(\n",
     "            models.kernels.matern_five_halves(nn.softplus(params['length_scale'])),\n",
@@ -311,42 +311,43 @@
   {
    "cell_type": "code",
    "execution_count": 13,
-   "id": "504ed32f-939e-4cb3-9806-9acc6d86791e",
+   "id": "132d51e3-4bbd-483b-9868-5576781c3ee3",
    "metadata": {},
    "outputs": [],
    "source": [
-    "def acquisition_fn(model, beta):\n",
-    "    ucb = acquisitions.upper_confidence_bound(beta)\n",
-    "\n",
-    "    def fn(x):\n",
-    "        return ucb(vmap(model)(x))\n",
-    "    \n",
-    "    return fn"
+    "batch_size = 1\n",
+    "num_results = 100\n",
+    "num_restarts = 10"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 14,
-   "id": "ccfa8401-028d-49f5-9e6d-1f1e0b779a33",
+   "id": "76a4d1be-b792-4016-97a8-bc49892c1f4e",
    "metadata": {},
    "outputs": [],
    "source": [
-    "bfgs = optimizers.batch(\n",
-    "    initializer=optimizers.initializers.q_batch(),\n",
-    "    solver=optimizers.solvers.scipy(method='bfgs'),\n",
+    "sampler = samplers.halton_uniform(\n",
+    "    distributions.uniform.uniform(bounds[:, 0], bounds[:, 1])\n",
     ")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 15,
-   "id": "132d51e3-4bbd-483b-9868-5576781c3ee3",
+   "id": "e5ccedcb-f8b9-453f-9557-6a921ad81313",
    "metadata": {},
    "outputs": [],
    "source": [
-    "batch_size = 1\n",
-    "num_samples = 100\n",
-    "num_restarts = 10"
+    "def optimizer_fn(acqf):\n",
+    "    return optimizers.batch(\n",
+    "        optimizers.initializers.q_batch(\n",
+    "            acqf, sampler, batch_size, num_results, num_restarts,\n",
+    "        ),\n",
+    "        optimizers.solvers.scipy(\n",
+    "            acqf, bounds,\n",
+    "        ),\n",
+    "    )"
    ]
   },
   {
@@ -553,16 +554,14 @@
     "        scale=jnp.sqrt(y_var),\n",
     "    )\n",
     "\n",
-    "    # Optimizing\n",
-    "    acqf = acquisition_fn(model, beta)\n",
-    "    \n",
-    "    next_x, _ = bfgs(\n",
-    "        random.fold_in(key, i),\n",
-    "        acqf,\n",
-    "        bounds,\n",
-    "        batch_size,\n",
-    "        num_samples,\n",
-    "        num_restarts,\n",
+    "    # Selecting\n",
+    "    acqf = models.outcome_transformed(\n",
+    "        vmap(model),\n",
+    "        acquisitions.upper_confidence_bound(beta)\n",
+    "    )\n",
+    "\n",
+    "    next_x, _ = optimizer_fn(acqf)(\n",
+    "        random.fold_in(key, i)\n",
     "    )\n",
     "\n",
     "    # Evaluating\n",
@@ -600,7 +599,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.7"
   }
  },
  "nbformat": 4,
diff --git a/docs/source/guides/Getting_Started.ipynb b/docs/source/guides/Getting_Started.ipynb
index bb93ddc..f29eb63 100644
--- a/docs/source/guides/Getting_Started.ipynb
+++ b/docs/source/guides/Getting_Started.ipynb
@@ -197,7 +197,7 @@
    "outputs": [],
    "source": [
     "def model_fn(params, x_train, y_train):\n",
-    "    return models.gaussian_process(\n",
+    "    return models.gaussian_process.exact(\n",
     "        models.means.constant(params['mean']),\n",
     "        models.kernels.scaled(\n",
     "            models.kernels.rbf(nn.softplus(params['length_scale'])),\n",
@@ -310,42 +310,43 @@
   {
    "cell_type": "code",
    "execution_count": 13,
-   "id": "e2b92eaf-4c9f-4122-bc1c-4e3637e69019",
+   "id": "9577c0e4-d6d8-4ccb-ad57-d3363c7a965a",
    "metadata": {},
    "outputs": [],
    "source": [
-    "def acquisition_fn(model):\n",
-    "    ucb = acquisitions.upper_confidence_bound(2.0)\n",
-    "    \n",
-    "    def fn(x):\n",
-    "        return ucb(vmap(model)(x))\n",
-    "\n",
-    "    return fn"
+    "batch_size = 1\n",
+    "num_results = 100\n",
+    "num_restarts = 10"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 14,
-   "id": "1013595d-d772-4580-b3e5-c6545a92c735",
+   "id": "43c77c11-c7d5-40d8-8e34-c609404df9f1",
    "metadata": {},
    "outputs": [],
    "source": [
-    "bfgs = optimizers.batch(\n",
-    "    initializer=optimizers.initializers.q_batch(),\n",
-    "    solver=optimizers.solvers.scipy(method='bfgs'),\n",
+    "sampler = samplers.halton_uniform(\n",
+    "    distributions.uniform.uniform(bounds[:, 0], bounds[:, 1])\n",
     ")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 15,
-   "id": "9577c0e4-d6d8-4ccb-ad57-d3363c7a965a",
+   "id": "75432c49-4c4d-4044-aed2-c0bafffa5a00",
    "metadata": {},
    "outputs": [],
    "source": [
-    "batch_size = 1\n",
-    "num_samples = 100\n",
-    "num_restarts = 10"
+    "def optimizer_fn(acqf):\n",
+    "    return optimizers.batch(\n",
+    "        optimizers.initializers.q_batch(\n",
+    "            acqf, sampler, batch_size, num_results, num_restarts,\n",
+    "        ),\n",
+    "        optimizers.solvers.scipy(\n",
+    "            acqf, bounds,  \n",
+    "        ),\n",
+    "    )"
    ]
   },
   {
@@ -465,7 +466,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+YAAAH0CAYAAACw4TeeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdeXxU1fn48c+dfTLZ9xAIAcIWUBRRVFQEdxTUCpZqbV3bKlXR1rUW97q0VtRqf2pbtf1WUXGr+wIKoqIgi0jYAyH7npnJ7HPv/f0xyUDIQpZJ5k5y3q9XXmTm3pk5uTy5uc895zxHUlVVRRAEQRAEQRAEQRCEqNBFuwGCIAiCIAiCIAiCMJSJxFwQBEEQBEEQBEEQokgk5oIgCIIgCIIgCIIQRSIxFwRBEARBEARBEIQoEom5IAiCIAiCIAiCIESRSMwFQRAEQRAEQRAEIYpEYi4IgiAIgiAIgiAIUWSIdgMGiqIo+P1+9Ho9kiRFuzmCIAiCIAiCIAjCIKeqKrIsYzKZ0Ok67xcfMom53+/nq6++inYzBEEQBEEQBEEQhCFmxowZWCyWTrcPmcRcr9cDMHny5PD3wtARCAQwGo3RboYgdErEqKB1IkYFrRMxKmidiNGhSZZlfvzxx8PmoEMmMW8dvq7X60ViPgTt2LGDwsLCaDdDEDolYlTQOhGjgtaJGBW0TsTo0Ha46dSi+JsgCIIgCIIgCIIgRJFIzIUhITMzM9pNEIQuiRgVtE7EqKB1IkYFrRMxKnRlyAxlF4Y2UYlf0DoRo4LWiRgVtE7EqKB1gyVGb7vtNpqamkhOTuaRRx6JdnMGDdFjLgwJ1dXV0W6CIHRJxKigdSJGBa0TMSpo3WCJ0ffee4/XX3+d9957L9pNGVREj/lB/H4/TqcTVVUHzR0tIcRms1FfX9/r17fGREJCAiaTKYItEwRBEARBEARhqBOJeQu/34/D4SA1NbXLhd+F2JSSktLn/1dFUWhoaCAxMVEk50LEjRkzJtpNEIQuiRgVtE7EqKB1gyVGP/jgA2RZFitdRZjIQFs4nU6RlA9igUCgz++h0+lITU3F6XRGoEWC0FZlZWW0myAIXRIxKmidiFFB6wZLjI4cOZLRo0czcuTIaDdlUBFZaAtVVUVSPogpihKR99HpdKiqGpH3EoSDud3uaDdBELokYlTQOhGjgtaJGBW6IjLRFmJO+eAWyf9fEStCfzCbzdFugiB0ScSooHUiRgWtEzEqdEXMMReGBHEiFLRODAcTtE7EqKB1IkYFrRssMbpmzRo8Xh8Go5FTTzlZdFpFiOgxF4YEr9cb7SYIQpd27twZ7SYIQpdEjApaJ2JU0LpYj1FvUKGk0cOVV/+Kn168gF/9+jd8W+rgh8pmKh0+ZEVM9+wLkZgPYmvWrCE1NRW73d6nfSJh7ty53HHHHf36GYIgCIIgCIIgRJasqOxr9LC5wkmFw8/B6beqgssvs6/Ry6YKJw3uvhdcHqrEUPYh7rjjjmPbtm0kJiZG5P3WrFnDvHnz2Lt3L0lJSeHn//3vf2MwRC/covnZgtAdGRkZ0W6CIHRJxKigdSJGBa2LxRj1BGR21rpxBw4UUr7w51fhdjUTZ4tvs69fVtlR6yY3yUxesmWgmxrzRLYyxJlMJrKysvr9c1JSUvr9M7oi5r4IWiduHglaJ2JU0DoRo4LWxVqMOrxBdtS6CbYMUdd5POi9HuZf9DNkaxx0cn1dbvcRlFVGp1kHsrkxTwxlj3E+n4/bb7+dcePGkZOTwznnnMOGDRva7PPtt99y0kknkZOTwxlnnEFRUVF4W0dD2deuXcucOXMYNmwYkydP5vbbb8flcrX5zHvuuYfJkyeTnZ3NMcccw3/+8x/279/PvHnzABg1ahSpqaksWrQIaDuU/f777+f0009v97OcfPLJPProo+HH//73v5k+fTo5OTlMnz6df/7zn70+TpFYx1wQ+tNgWdtUGLxEjApaJ2JU0LpYilGHN8ju4kpylr/M1Jt+xewzjuPsEws5Y/YxnDVjMqedfizTfnsFI5a/jNHe1O711c1+9jeKGk89EVu3bQZQwuzZ6KqrB/xzlawsnCtXdnv/u+++m3fffZenn36aESNG8NRTTzF//ny+//778D5LlizhoYceIisri/vvv59LLrmEdevWYTQa273f3r17WbBgAXfeeSdPPfUUdXV13Hbbbdx66608/fTTAFx77bWsW7eOhx9+mMmTJ1NSUkJ9fT25ubm89NJL/PKXv+S7774jISEBq7X9nbL58+fz+OOPs3fvXkaNGgXAtm3b2Lp1Ky+99BIAr7/+Og8//DCPPPIIRx55JD/88AOLFy8mLi6On/3sZz06poIgCIIgCIIgHOBwOPjggw9YuHBhu22v/uMFTtu4hVP+txyjq7nD15sb6sn86gsyv/qCwr/cx76Fv6T4imsJJCWH9yl3+LCadGTYTP31YwwqIjHvhK66Gp3G72q5XC5eeOEFnn76ac444wwAli5dyhdffMF//vMfpk6dCsCtt97KrFmzAHjmmWeYPHky7733HhdeeGG793z88ceZP38+1157LQBjxozh4Ycf5rzzzuOxxx6jrKyMt99+mzfffJNTTz0VgPz8/PDrW4esZ2RktJljfrCJEycyefJkli9fzi233ALA8uXLOeaYYxg9ejQADz/8MPfffz9z584FQstL7NixgxdffLFXiblYLk3QutbYFwStEjEqaJ2IUUHrtBKjDoeD+fPns379eqqqqli8eHFog6ry2VVXM3nTTv53xEyKfvkg+9OHo+h0WAM+8ptrmWSv5LRd35G/9XtMTY0A6H0+xrz0HMPfWc6WJQ9RM+vM8GftbfBgM+qJM+mj8JPGFpGYd0IZgHnXff3cffv2EQgEmD59evg5o9HI1KlT2blzZzgxP+6448LbU1JSKCgo6HS5hq1bt7J161aWL18efk5VVRRFoaSkhKKiIvR6PTNmzOjpj9bG/Pnz+e9//8stt9yCqqq88cYbXHfddUDohsPevXu54YYbDpwogGAw2OsidYFAQCTngqbV1NSQl5cX7WYIQqdEjApaJ2JU0DqtxOgHH3zA+vXrAbjvvvsAWHzNNXx05Y18MPFMtp/423avcQDVSRl8m1vI7z56BVVWyR6Rz+cnnsKIt5ah9/sxNzUw7eZfs2/hL9n2u7tQDQZkBXbXu5mcHY9O1HzqkkjMO9GT4eSDSXNzM5dffjm/+tWv2m0bPnw4e/fujcjnXHTRRdx7771s3rwZj8dDeXl5uAe/dT770qVLOeaYY9q8Tq/v3d02RVEOv5MgRFFzc8dDxQRBK0SMClonYlTQOq3E6MKFC6mqqgon5c8+8mcadgZYe8aidvsmmvUY9TpcfhlvMHQ9HfQ4CTibqNAZeOGSmznll79i0qP3kv3FpwDkL3sJa0Upmx5+Ctkah8uvUOnwkZskKrV3RSTmMSw/Px+TycS3337LiBEjgFDP8IYNG/jNb34T3m/dunUMHz4cgKamJvbs2cO4ceM6fM8pU6awY8eOTofaFBYWoigKX331VXgo+8Fa563Lstxl23Nzc5kxYwavv/46Xq+XU089NbyERGZmJjk5Oezbt48FCxZ0fRC6SVRlF7TOZBLzrwRtEzEqaJ2IUUHrtBSjraNS//v3f1F4+QOszRgR3pbmauDkY8czITMOqzHUKaaqKnXuAD9UNrMzNRudwYQxIZUPdtRTkmXjgj//nTH/e41JD92NLhgga/VKjln8K9Y/8Q8Ui4Uyu4/UOGP4/YT2RGIew2w2G1dccQV33303ycnJDB8+nKeeegqPx8Nll13Gjz/+CMCf//xnUlNTycjI4MEHHyQ1NZVzzz23w/e88cYbOfPMM7n11lu57LLLiIuLY8eOHXzxxRc8+uij5OXlsXDhQq6//vpw8bfS0lJqa2u58MILGTFiBJIk8fHHH3PGGWdgsViIj4/v8LPmz5/PI488gt/v58EHH2yz7bbbbuOOO+4gMTGR0047Db/fz8aNG2lqagpXeu8JMYxd0DqtzDsThM6IGBW0TsSooHVai9HzzryAL5lCZXwyAHFeF74vX+Puvz2EwxtEPmjAqSRJZNhMnFaQygmvvs3K3Y2sK3MAsLXahcMb5NK5F+Makc8xN/8aY7OT9O++Yuot1/H9X59FMRopafQyIdMWhZ80Nojl0mLc3Xffzdy5c7n22muZNWsWxcXFLF++nOTk5Db73HHHHcyePZvq6mpeeeWVTu/YTZo0iXfffZfdu3dz7rnncuqpp/LQQw+RnZ0d3uexxx5j3rx53HLLLUyfPp3FixfjdrsBGDZsGLfffjv33Xcf48eP57bbbuu07eeffz4NDQ14PB7mzJnTZtsvfvELnnjiCV5++WVOOukkzjvvPF555RVGjhzZq+Pk9YrlGgRt2759e7SbIAhdEjEqaJ2IUUHrtBSjO3ZX8PvPy3G2JOUjKvawb+mv+fbT/+Oj/z7PMbmJ5CVb0HeQLcYZ9Zw3MZ2fTcnCbAiNSi21+3hhfSVlRx3HuqdfIhgXSsAz13zOxMceAKDRE6TJI5Yw7oykqqoa7UYMhGAwyKpVq5gyZUqH85Tr6+tJS0uLQsuia8WKFVx88cVUVlZqanhNpHk8ng6XbuuNoRorQv8qKiqisLAw2s0QhE6JGBW0TsSooHVaidHiKju3v7qRJltoBaXxe7ew/vWH2VFXFd5nyZIlLF68GH9QYVedG4ev42mqVU4f/95Qhcsf2j4iycwvjskhe+N3HHftL9AFQ4n4lj88SOn8S7CZdByRHT+kppnKsszmzZuZOXMmBkPnA9ZFj/kQVlNTw4cffsiYMWMGdVIOdPlLIAhaIG72CFonYlTQOhGjgtZpIUarnT7uefnbA0l5SRFTTCV8s7OIJUuWhPe77777WLZsGSaDjsIsG1nxHecK2QlmrpyWg61lObRSu483ttRQd8x0frzrwFTVwkfvJWFHES6/QoM72I8/YewSifkQ9tOf/pRVq1bx6KOPRrsp/W4o3ZUTYpPFIiqVCtomYlTQOhGjgtZFO0abfUGeePkrqhJDBZdHVpdwnG8rV9/9RyBUEK41OZ82bVp4qqkkSYxOszIsMZScv/zsEzz/lwd4+dknAEi3mbjs6GzM+tD19vZaN6uKmyg7fwF7f3Y5APqAn6PvuAGdx0OZ3csQGbTdI6IbcQj7/PPPo92EARMIBESvuaBp5eXlJCUlRbsZgtApEaOC1okYFbQumjEakBX+74NNrNenABDvaWa+fQNnPfqnNvstXryY7Oxs5syZQ2JiYpttI1OsKCp89OYy6muqSMvM5pJf3whATqKZi4/M4v82VqECXxQ3kptoRrf4dlI3riNp+1bi9+5hwhMPU3T7vTS4g6TZjAPys8cK0WMuCIIgCIIgCIIwiH21s4Z3qw+UWb+jai1nPXpfh/suXLiwXVLeKj/Fgq6TgagF6XGcVpAafvzW1hrsqp5NDz2B3DJaYORr/yF50/eUO0Rh5kOJxFwYEgb7HHoh9uXn50e7CYLQJRGjgtaJGBW0LloxWtrk5eUVRQQMoR7qi7es4Ni7ftur95IkiVf++1+e/u+b/PHx59ptPyk/ifHpcQC4Awpvba3FOXI0O6/7Xej1qsoR992Op9kjKrQfQiTmwpAgyx1XkhQErWhoaIh2EwShSyJGBa0TMSpoXTRi1B9UeOujjeyzhIawj6ou4bLLz4A+rFY0derRXHD6yUycfGS7bZIkcf6kDBJaisHtqffwfZmTvZdcQdOk0P4Je3eT//ILVDr9vW7DYCQSc2FIEIm5oHUOhyPaTRCELokYFbROxKigddGI0XV7G/i4KZQk6xSZG01VmI+e0uf3tRr1jE7rOLm3mfRcMDkj/PjTXfU0+VW2/PEhVF0o/Sx4/m+4SyvDy6wJIjEXhghRlV3QOlGcUNA6EaOC1okYFbRuoGO00e3njY82hoewX1S0iomLr4rY+6fFGcmM77iAW0FaHFOHJQDgk1Xe216LY9xE9v/kZwAY3C4mPPUo1aLXPEwk5oPYmjVrSE1NxW63R7spEbNo0SJ+/vOf9/h10V6eQhAOZ9y4cdFugiB0ScSooHUiRgWtG+gY/XTtLn60pgOQ3VjNTy6eiWTseyX0oqIiNm/eTFFREfkpVizGjlPKM8elkmAO9dbvqvOws87NzutuJpAQKiyX++4beDf/gKyIpdNAJOaCRu3fv5/U1FS2bNnS5vmHHnqIp59+usfv5/F4ItU0QegXRUVF0W6CIHRJxKigdSJGBa0byBitcHh5d2tt+PHlzh2knTgtIu+9YMECZs2axYIFC9DrJMakdjyk3WrUc/a4tPDjj3bU40lKYffVocJzkqpS8NRj1LhErzmIxDwiHA4Hy5Yt63DbsmXLhtScJ7+/f3+xEhMTxRqlgiAIgiAIgtAJVVV55+NNVFtD18xT923h+EWX9tvnJVoMZCd0vALSpCwb+SmhkasNniDflNgpufgyPJnZAGSt+gzvl9/0W9tiiUjM+8jhcDB//nyuu+46li5d2mbb0qVLue6665g/f36/Jec+n4/bb7+dcePGkZOTwznnnMOGDRva7PPtt99y0kknkZOTwxlnnNHmbl1paSk/+9nPGDVqFMOHD+eEE07g008/DW8vKipiwYIFjBgxgvHjx/Ob3/yG+vr68Pa5c+dy6623cscdd1BQUMD8+fO55ppruPLKK9u0IRAIUFBQEL6B8dlnn3HOOeeQn5/PmDFjWLhwIXv37g3vf9RRRwEwc+ZMUlNTmTt3LtB+KPvhfv7W4fxff/01s2fPJjc3l7POOotdu3aF9/nxxx+ZN28eeXl55OXlMWvWLDZu3Njj/wtB6IuUlJRoN0EQuiRiVNA6EaOC1g1UjO5v9PBpTaiomk6RuSRTJi4rPWLv/7Of/YyrrrqKn/3sZ+Hn8pItmA3tazpJksQ549No3bJ6byNNGNj96xsOvPapx3B4gxFrX6wSiXkfffDBB6xfvx6A++67L5ycL126lPvuuw+A9evX88EHH/TL59999928++67PP3003z++eeMHj2a+fPn09jYGN5nyZIl3H///axYsYL09HQuueQSAoHQuoG33HILPp+P999/nzVr1nDPPfdgs9kAsNvtXHDBBRx55JGsWLGC119/nZqamnZJ97JlyzCZTHz44Yc89thjLFiwgI8//pjm5ubwPitXrsTj8XDuuecC4Ha7ue6661i5ciVvv/02Op2Oyy67DEVRgFDiDvDWW2+xbds2/v3vf/f65wd4+OGHw8fAYDBw/fXXh7f96le/YtiwYXz22Wd8/vnn3HjjjRgjMP9GEHoiPj4+2k0QhC6JGBW0TsSooHUDFaPvvrmGZnNoLfEzdn3LhGsuiej733XXXfz5z3/mrrvuCj+n10nkp3Q8pD07wcyxI0Lzyv2yyqe7GiibtwDXiJEApH/7Fe6vv41oG2ORSMz7aOHChSxZsiT8+L777mPUqFHhpBxCifHChQsj/tkul4sXXniBe++9lzPOOIMJEyawdOlSrFYr//nPf8L73XrrrcyaNYvCwkKeeeYZamtree+99wAoKytj+vTpFBYWkp+fz1lnncWJJ54IwPPPP88RRxzBH//4R8aNG8eRRx7JU089xZdffsnu3bvD7z969Gjuvfdexo4dy9ixY5k9ezZxcXG8//774X2WL1/O2WefTUJCqDrjvHnzmDt3LqNHj+aII47gqaeeoqioiO3btwOQnh66q5eamkpWVlaHdxi7+/ND6AbEjBkzmDBhAosXL+a7777D6/WGj8HMmTMZN24cY8aM4YILLmDy5Ml9/v8RhJ4oLS2NdhMEoUsiRgWtEzEqaN1AxOj+ikY+9YduABiCAc47vgCT1dzvnwuQGmckNa7jyvOzxqRgbSkS90NVM/ubg+y5/Dfh7Wl/e2LIF4ETiXkELF68uE1yfnAV9CVLlrB48eJ++dx9+/YRCASYPn16+Dmj0cjUqVPZuXNn+Lnjjjsu/H1KSgoFBQXh7b/61a947LHHOPvss3nooYfYunVreN8ff/yRNWvWMGLEiPDX8ccfD9Bm2PmUKW3XQjQYDFxwwQW8/vrrQCiB/vDDD1mwYEF4nz179nD11Vdz9NFHk5eXFx66Xl5eHvGfH6CwsDD8fVZWFgB1dXUAXHfdddx4441ceOGFLF26tM3PJgiCIAiCIAix4u3lq/GYQj3XZ+3fRMGFZw7o5+enWNF3kGHGGfXMHpMafvzZ7gbKzrsQT1YOAJlffEbz+k0D1EptEol5hCxevLhdUbKkpKR+S8oj5Re/+AUbNmzgpz/9Kdu2bWP27Nk899xzQCihPuuss1i1alWbr/Xr14d71QHi4uLave/8+fNZvXo1tbW1fPDBB1gsFk477bTw9ksuuYTGxkaWLl3Kp59+yieffAL0X/G4g9vYuqZ567D522+/na+//pozzjiD1atXc8IJJ4RHFAjCQMnLy4t2EwShSyJGBa0TMSpoXX/HaFlFA58YQ0XVzAEf5849Dr2u/bzv/mQ26MhN7LiH/pjcBFKtoR71fY1e9jhlin9xTXi79e/PDEgbtUok5hGydOnSduuF2+32dgXhIik/Px+TycS33x6YkxEIBNiwYQPjx48PP7du3brw901NTezZs6fNOorDhw/niiuu4N///jeLFi0Kz+eeMmUKO3bsIC8vj9GjR7f5ap2H3pnp06eTm5vLW2+9xeuvv875558fnrfd0NDArl27+P3vf8/MmTMZP358u2PXuq8sy33++Q/3PgAFBQVcd911vPnmm5x33nm8/PLLXe4vCJF26O+AIGiNiFFB60SMClrX3zH66htf4TeGqqPPqfyR0cdPOcwreuf888/nhBNO4Pzzz+9we06iucO1zfU6qU2v+YrdjZRecDH+xFDnZsb7/yNQ1v3Rs4ONSMwj4OBCb0CbnvODC8JFms1m44orruDuu+/ms88+Y/v27SxevBiPx8Nll10W3u/Pf/4zq1atoqioiEWLFpGamhouwnbHHXewYsUKSkpK2Lx5M19++WU4ab/qqqtobGzk6quvZsOGDezdu5cVK1awaNGiwya6EOo1f+GFF/jiiy/aDGNPTk4mNTWVl156ieLiYlavXt2meARARkYGVquVFStWUFNT02FV++7+/NB5Yu7xeLj11ltZs2YNpaWlrF27lo0bN7a5cSEIA0FcUApaJ2JU0DoRo4LW9WeMVlY1sMIU6i2P87mZO3c6Oql/est3797Njh072tScOphOkshPtnS4bVK2jaz40M2DcoePrc2wf0FoKTddMID0/57rlzbHApGY99GyZcvaFXrbu3dvu4Jwna1z3ld33303c+fO5dprr2XWrFkUFxezfPlykpOT2+xzxx13MHv2bKqrq3nllVcwmUK/EIqicOutt3L88cezYMECCgoK+Mtf/gJATk4OH374IYqicNFFF3HSSSdx5513kpSUhE53+NCZP38+O3bsICcnp808cJ1Oxz/+8Q82bdrEjBkz+MMf/sC9997b5rUGg4GHHnqIF198kcLCQi69tOO1F7vz83dFr9fT0NDAtddey7HHHstVV13F6aefzu23396t1wtCpHTnd0oQoknEqKB1IkYFrevPGH3z9S8J6kPDxM+r2srwY4/ot8+Kj48Pf3UmJc5IsrV9ITidJHFawYGizit2N7B3wWUohtBo2eT/+ze43ZFvdAyQVFWN2fJ35eXl3HbbbXz44Ye43W4KCgp44YUXmDZtWrt9g8Egq1atYsqUKej1+nbb6+vrSUtL63EbWtcxX79+fbtCb6096dOmTWP58uUkJib2+P0F7eltrAiCIAiCIAhCpDnqmrjstSI8JivmgI9npycybGrh4V/YzzwBmc2VzRyabaqqyj/XVVBq9wFw4aQMfvH3exj+3psANDzxFNJlHXfKxSJZltm8eTMzZ87EYOi4aj3EcI95Y2MjM2bMwGg08uGHH1JUVMRjjz3W4bJa/SkxMZHly5fzzDPPtCv0tnjxYp555hmRlGuAx+OJdhMEoUutSwUKglaJGBW0TsSooHX9FaPvvrIiXIn99NodmkjKAaxGfXjY+sEkSeL0ggNzzVcVN7L3ogOJuPnFFweieZrTecqucY888ggjRozghRdeCD83atSoqLQlMTGx03XK+2P9ckEQBp/WVQIEQatEjApaJ2JU0Lr+iFFPs5u31XQA9LLMBecdd5hXDKwRyRbqXAGCh6xRnp9qZVSqhb0NXho8Qb4aM5YpY8eTuGsHto3fEywqQinUxg2GgRKzPeb/+9//mDZtGgsWLCAzM5Ojjz6a559//rCvczqdOByO8JfP5xuA1grR1tH0BUHQkkOXWxQErRExKmidiFFB6/ojRj99dSX2uNDI3FPqdjFy6sSIf0ZfGHQSw5M7Xj5t5qgDI51X722i5KJLwo91//xXv7dNa2K2x7y4uJi///3v3Hzzzdx5552sW7eOG264AZPJxC9/+ctOXzd58mTcBxUUuOKKK7j++uuJi4sLD3c2Go2oqkowGATAbDYTCARQFAWdTofRaAwn9AaDAUmSCAQC7faVJAmz2YzX6+3zviaTCVmWkWUZSZKwWCzh9ur1evR6fXgN8IP3BbBarXi9XlRV7XBfRVHCP+uh+xoMhvDPeuhxsVgs+Hw+VFVtd1w62tfv93d4DFuXRuvouOh0OkwmU5+PoaqqGAwGgsFgh8fQYDCg0+k6PIaH7uv1emlubmb//v0AjBgxAqfTSVNTE5IkMXHiRHbs2IEsyyQmJpKamsq+ffsAyM3NxePx0NDQAMDEiRPZvXs3gUCA+Ph4MjIy2Lt3LwDDhg3D7/dTV1cHwPjx49m3bx8+nw+bzUZ2djZ79uwBIDs7G0VRqKmpAWDs2LGUlpbi9XqxWq0MHz6cXbt2AZCZmYkkSVRXVwMwZswYKisrcbvdmM1mRo4cyc6dO4FQdXyDwUBlZSUAo0ePpqamhubmZkwmE6NHjw4Py0pLS8NisVBeHlrmIj8/n4aGBhwOBwaDgXHjxlFUVARASkoK8fHxlJaWAqF1Pe12O3a7HZ1Ox4QJE9i+fTuKopCUlERSUlKb493c3ExjYyMAhYWF7Ny5k2Aw2OHx9nq91NfXAzBhwgSKi4vx+/3Ex8eTmZlJcXExECp2GAwGqa2tBWDcuHGUlJTg8/mIi4sjJycnfLyzsrJQVbXN8S4rK8Pj8WCxWBgxYkSb463T6aiqqgof76qqKlwuF2azmfz8fHbs2EEwGMRoNGIymaioqABCo4Bqa2tpbm7GaDRSUFDAtm3bAEhNTcVqtXZ4vPV6PePHj2fbtm2oqkpycjIJCQltjrfD4eg0ZlNSUigpKQFCyym63e5wzB58vBMSEkhPT28Tsz6fr8PjbbPZyMrK6vJ479+/Pxyzubm54WqvWVlZAOGYLSgooLy8PHy88/LyuozZ6upqXC5XhzFrNpvbHO+6ujqcTme7mE1NTSUuLo6ysjIARo4cSWNjY6fHOzExcdCdIxwOB16vV5wjonSOAEhPTxfniC7OEfX19djtdnGOENcRmj1HtMZopM4RBr2B1x0WSACAs4/Koampqd/PEStXrqS+vh6dTsf8+fMPe47A5YL6WmRFRZ+Rj1JfhqoEyTNZyUsysd/up9YV4JOps5loeQiD14vltVdx3HM3xdXVMX8d0d2SbjFb/M1kMjFt2jS+/vrr8HM33HAD69at45tvvmm3f2vxt9GjR7epiGg2mzGbzaKg1yDn8XiwWq0ReS8RK0J/KCoqonCIDdkSYouIUUHrRIwKWhfpGP1i+Ur+1BTqdZ5etZP77/ppxN67K5MmTaKyspKcnBy2bt3ardc0egJsr2lfbX13nZv/bAzddMiON/H0x0+R985rALiefBL/z38euYZHyaAv/paTk9MusCdOnBi+G9aZhIQEEhMTw19mc8dDKwRBEARBEARBELRIVVXeLHaFHy84IjOKrTm8FGvHy6eNSbOSmxjKx6qa/Xx43mXhbYYXXhyo5mlCzA5lnzFjRnhoV6udO3cycuTIKLVI0LLWddsFQatGjBgR7SYIQpdEjApaJ2JU0LpIxuj2L9azPXU4AGPr9nPElXMj9t6H89RTT+Hz+XrcwTky2YLd23b5NEmSmDk6mZc3hYaZfyAn89PxhSTvKMK8cQO+LVuQj+i/Ndm1JGZ7zG+66SbWrl3Ln/70J3bv3s3LL7/Mc889x6JFi6LdNEGDWufbC4JWNTc3R7sJgtAlEaOC1okYFbQukjH6ztri8Pc/ydEj6QYurZs9ezbnnHMOs2fP7tHr4kwdL582Lj2O7ITQ8xUOP5/89DfhbaaXX+5bY2NIzCbmxx57LG+99RavvPIKkydP5v7772fp0qVceungWYxeiByRmAta11qERhC0SsSooHUiRgWti1SM1m7bw6rUMQCkuOycfPHpEXnfgTA8yYxBJ7V5TpIkTs5PDj9+K/coZGMoUTe++Sa0FJMe7GI2MQc477zz2LJlC16vl23btnHNNddEu0mCIAiCIAiCIAgR43A4WLZsWfjx+x+sQ25ZCvj4+u14A/5oNa3HjHpdh8unTcy0kdIyB323PcC3cxYCoK+txbB69YC2MVpiOjEXhO6KVEV2QegvopKwoHUiRgWtEzEqaF1vYtThcDB//nyuu+46li5disfRzHumYQAYg36effHPzJ8/H4fDEenmdqqqqory8vLwEm49lR1vIs7YNg3V6yROHHlgnff/m3Fh+HvT8uW9a2iMEYm5MCS0rm0uCFrVun6mIGiViFFB60SMClrXmxj94IMPWL9+PQD33Xcf/+/B/4fDGlq4fMymz6lpbmT9+vV88MEHEW1rV0477TSOOOIITjvttF69XpIk8lPbd5odNSwhnLB/TyIlw0PD9Y3vvgvu9kutDTYiMY9hc+fO5bbbbuOOO+5g1KhRjB8/npdeegmXy8WiRYvIy8vjmGOO4dNPPw2/pqioiAULFjBixAjGjx/Pb37zG+rr68PbP/vsM8455xzy8/MZM2YMCxcuZO/eveHt+/fvJzU1lXfffZd58+aRm5vLySefzHfffTegP3tPqQeXfxQEDQoOkflTQuwSMSponYhRQet6E6MLFy5kyZIl4cebEsaEv9+25k0AlixZwsKFC/vewAGUZDGQFmds85xJr2N6XqjXXAH+Nf96AHQuF8aPPhroJg44kZjHuFdeeYW0tDQ+++wzrrnmGn7/+99zxRVXcNxxx/H5558za9Ysrr32WtxuN3a7nQsuuIAjjzySFStW8Prrr1NTU8OVV14Zfj+32811113HypUrefvtt9HpdFx22WUoitLmcx944AF++9vfsmrVKsaMGcM111yj6T+I+pZ5OIKgVYmJidFugiB0ScSooHUiRgWt622MLl68mCVLljB99JFUZucDMHrvFr6v2M2SJUtYvHhx5BrZDWeeeSbz5s3jzDPP7NP7jEyxoD8kGz1ueCLGluJwnw6bTFNc6JgNheHskjpEuhKDwSCrVq1iypQpHSZp9fX1pKWlhR8vens7je6BTzRT4gw8fcGEbu07d+5cZFkOD12RZZn8/HzOO+88/v73vwNQXV3NxIkT+fjjj1m1ahXffPMNb7zxRvg9ysvLOeKII/juu+8oKCho9xn19fWMHTuWNWvWUFhYyP79+znqqKN44oknuOyyywDYvn07J554ImvXrmXcuHF9PQT9QpbliCXnh8aKIESC2+0mLi4u2s0QhE6JGBW0TsSooHV9jdHrb1jKjiNnAjDs5QdYtXdDm5Gtsajc7mN/U9sppx9sr+Pb0tCc+Su/Ws617/wd1WDAvn07ampqNJrZJ7Iss3nzZmbOnInBYOh0P9Fj3olGd5A6d2DAv3p6M2DSpEnh7/V6PSkpKUycODH8XGZmJgB1dXX8+OOPrFmzhhEjRoS/jj/+eIDwL/WePXu4+uqrOfroo8nLy+Ooo44CQgl8Z5+bnZ0d/gyt8vtjp1qlMDTt27cv2k0QhC6JGBW0TsSooHV9idGnHnyU3ZNmAJDsbOSTLWuw2+0sXbo0Mo2LkpzE9oXgThiZROuKaq9Pn4vHaEEKBjG+804UWjhwOk/Zh7iUuOgcmp5+7qF3XSRJwmg0tnkMoCgKLpeLs846i3vuuafd+2RlZQFwySWXMHz4cJYuXUp2djaKojBjxox2iW1nnyEIgiAIgiAIQuQsXbqUnRtLkc8OXffnbPgErxwAQgXhgAEfzh4pOkliVKqVrdWu8HMpViOTs+L5oaoZp8HMu8eexcVfv4Np+XL8V1wRxdb2L5GYd6K7w8ljyZQpU3j33XfJy8vrcBhFQ0MDu3btYunSpZxwwgkArF27dqCb2S8OvpEgCFqUm5sb7SYIQpdEjApaJ2JU0LrexOiyZcu4/4EHOPP2/wIgKQp3/fE3jDomO5yU33fffWRnZ8dcAbhWiRYDmfFGapoD4edm5CfxQ1UzAP+ZfQk/WfsuhrVrkWpqUFtGBA82Yij7EHLVVVfR2NjI1VdfzYYNoTkpK1asYNGiRciyTHJyMqmpqbz00ksUFxezevVq7rrrrmg3OyKGSCkFIYaJJf0ErRMxKmidiFFB63oTo3PmzGHh7IuoTw4loyfW7yFjXH64IBzAtGnTmDNnTkTb2pXLL7+cuXPncvnll0fsPUemWDHppfDj7AQzBWmhJdWqEtNZecRMJFXFOIDLwg00kZgPITk5OXz44YcoisJFF13ESSedxJ133klSUhI6nQ6dTsc//vEPNm3axIwZM/jDH/7AvffeG+1mR4SWK8YLAtBm2UJB0CIRo4LWiRgVtK43MZqYmEjC1LPCj+eOSwl/v3jxYp555hmWL18+oKsSrFu3jq+++op169ZF7D0NutCQ9oPNGJkc/v4/My9GBUzvvRexz9QaMZQ9hr377rvtntu8eXO75xoaGsLfjxkzhn//+9+dvuepp57abvj6wa/Py8tr8xggKSmp3XOCIAiCIAiCIPTN9u372ZA+CoBhTdUc9YvZbbbH6vD1jqTGGcmwGal1hYa0j0q1kJNgotLpZ/vwcXw/5iiOWb0ayW5HTUqKcmsjT/SYC0OCxWKJdhMEoUsTJgy+uhbC4CJiVNA6EaOC1vU0Rn1BhY8+2hB+PE+qR2cyRbpZPbZhwwYqKyvZsGHD4XfuofzUA0PaJUlq12suBYMYP/kk4p+rBaLHXIMCsoKYER1ZQb8fQ4ROZN6gTKXTF5H3ihXSYbe33UPq5AWSdOC9wq9peU46+HtJavn3wGOd1PJvy/O6g54bDIqLiykoKIh2MwShUyJGBa0TMSpoXU9iVFVViqqcrDGkAWAK+Dnj3On92bxuM5vN/fbeBp1EQXoc22pcqCoUZtlI2m3A7g3y9YTp7M7KJ+/dd/EvWNBvbYgWkZhrkF9WRbGyCFMVBSUYmeXcvAGFmgZRYEYrdAcl6TqdhF6SMOgk9Dpa/pUwtv6r12HUSRj1Eia9Dr1OO0n9oUsSCkKkKKqKooKsqOHvoX1RzNYbYjpd6AaYvuX3ppWIUUHrRIwKWteTGC1p8rJl9SYcltDc8VlV20iYcHk/tUxbkiwGhiWYKXf40OskTshL4qOdofn5/z1lAUve+xu43RAXF+WWRpZIzIUhQZLErI3BSlFDiQcAcs9uaBl0EmZDKEk3G3RYjDqsBh1Wox6zYWBjJj4+fkA/T4h9sqLiDSp4gwr+oIJfVvDLKgFZJSArBBUVWVWR+3hP0qAL3ewKYmJbjQuT/sDvjEkvhX53DLpBM3pFiF3iPNo3sqKGzyNBWSWgKMhK6Bwiq6Ebe7Iauql3aP+R1DKCTS+FRsTpdS039yQJg17CIIVuihv0ofOGboieL7obo7UuP5UOP99WuaGlqNvcwvT+bJrmDE824/AFcfpkpuYm8EVxI96gwkdHn8a1H79A3MqVBM47L9rNjCiRmAtDg04f7RYIGhRUVIJ+FRftMxe9DuKMemym0Fe8SU+cqf/iKHOQrskp9J2qqngCCs1+GXdAxu1X8ARk/D28EdVbQUUlqKio5iSaPB2vcKGTwGzQtfzOhP6NN+sx6sVNUWHgiPPo4alq6Iae26/gCsihm3sBBV8wdDNvoLS7Md76ZRzcN/q6E6NOX5Dieg/VlQ3sSswGYGz1XsZeek5/N6/bli9fjtvtJi4ujvnz5/fLZ+gkiXHpcS1rmes4dngiX+5rImgw8uqMC/jVe++JxFwQYpEqB5AM0S+WIcQOWQGnT8bpk8PPGXQSCWY9SRYDSRZDRBP14uJiCgsLI/Z+QuxSVbUl9oI4Wv7ta693JCiNFegz8jvepoInoOAJKNS7DzxvMepINOtJtBhIMhswDfBIFGFoEefR9mRFxeENhs8proCsifNJVzfGdRJYDDriTPrwzT6baXDc6DtcjHoCMjtq3SgqbNmwE0yhRP5cpQbJau30dQPt7rvvprKykpycnH5LzAFMBh3jWuabT89L5OuSJmQV3jx+Lpc/dQ34/aCBYniRIhJzQRCEbgoqKo2eII0tvYZmg0SK1UhanJEEs37Q3uEX+p+sqDR6AjR4gtg9wQHtuepP3kCoN66mObT0jc2kJ8VqIC3O2K8jUARhKPMEZBrcAZo8QZx+ud2wc61TVHAHFNwBBQiEn7cYdMSb9SSY9SSYDcQZB1fPui+osL3GTUBW8QRkvtMlA2Dzuph9jjaKvkVDosXAqFQre+o9HJmTwMYKJ83WeP438RTmffUVwVmzot3EiBGJuTAkSHoR6kLk+YIqVU4/VU4/Jr1Eus1Ihs3Uq4QjJyenH1ooaJmqhm701LkCNHoCaD0X1yWk9fk9XH4Zl1+mzO4jzqgjI95Eus2IaRD0hAnRN5TPo/6gQq3LT50r0JLQDj6tNTXqWta4NugkEi2hETnJFgNWo/Zv9nUWo/6gQlGNC29LoeKirSX4WkZ6nrnnOyyLbhqwNnbHvffeGx7KPhAy4014gwonjkxiY4UTgFdO+gnnf/wJHJSYL1u2jDlz5pDYMi8/1ohsRRgaVPXwa34JQh/4ZZUKh58Kh59Es57sBDOpcYZu380PBjueuysMPv6gQnWzn5pm/4DNE48EVZYjehp1BxRKGr2UNnlJiTOSk2AiwSwuS4TeG4rn0UZPgGqnnyZvMOZ6xvsqqKg0uIM0uA+MYku2GEm2hqabaWnllVYdxagnILO9xh1OylVVZV25E0wJAJw5Ir7zdWijpD+Hr3cmL9mC3eFEX1eCnD6S6pQsVq1v4kRVBUli6dKl3HfffUybNo3ly5fHZHIublELQ4KqyIffSRAixOGT2VnnZmOFk0qnD7kbXaG1tbUD0DIhmjwBmT31obgos/tiKikHUN1N/fK+igr1rgA/VrnYWtVMkydw+BcJQgeGynlUVVVqXX42VzjZXuOm0TP0kvKO+IIq1c1+dtS6WV/mYFuNi2qnH78WJtW3ODRGnb4gW6sP9JQD7K13U9mSlE8t3kzeTwdXgbO+2Pr1CnavfC38+OXCWUg7doSTcoD169fzwQcfRKuJfSJuTXfhmxL7gH7eCSOTerR/c3Mzf3nkYT7+8APq6uuYPHky99z/IFOOOjq8z803Xs/y115t87qZp87iP6+EnvP5fNz6u5v49OOPyMjM5IGHHuHkU2aG9/1/z/yNivJy7nvwocO2x+l08ve/PcWH779HWVkpiYmJjJswgV/88grOnnMukiRx8U8uoHDSZO65/4Ee/ayCEIt8QZV9DV4q7D6GJ1nIjDcOqvlwQvd4gwrldi+1roC4eD4Mh0/GUeMm0aInL9kietAF4RD1rgD77V68g3S4eqQoKjR5gjR5gkiNkGDSkxpnJDXOOODLoXamptnP3gZPu2lMm7eUABYAzm7YjTH38gFvm1YtXLiQyqoq1pTvwpk7lp25Y7nj6pv5f0Vrw/ssWbKEhQsXRrGVvSf+4sWwW393Ezu2b2fpU0+TlZ3Fm28s55KL57Ni1RqyD5rDcuqs2fxl6RPhxyaTOfz9y//3H7b88ANvvfcBX6xcwQ3XXcuGLVuRJIn9+0t45b//x3sffXrYttjtdi46fy5Op4Pf33Y7U6YcjcFgYO03X/OnB+7jxJNOJimpZzceIklUZBeiyS+rFDd4qHT6GJliIcVqbLfPuHHjotAyoT/Jikq53Uel06f5+ePdoUsbMWCf5fDK/FjlIsNmJC/FIuagC90ymM+jzb4g+xq9bVYKEbpHVVtu+vlk9jV6iTfrSbUaSY0b+Hnp48aNI6iErgnqXe1HBzm8QX4ImECCVGcDk2ccMaDt6y6fzxf+3mw2d7Fn5N20eDHND/0/vmx57DjhAmhJzJcsWcLixYsHtD2RJP7SxSivx8OH77/HnX9cwvQTTiB/1Ghu/v2tjMwfxX9eerHNviaTiczMrPBXcnJyeNvuXTs546yzGD9+Ar+8/Erq6+toqK8H4A+33codf/gjCQkJh23Pow89SFnpft55/yMWXLyQcePHM3rMGC75+WV89NlKbDZbJH/8npPF0Egh+jyBUMXVHbUu/MG2vR0lJSVRapXQH+rdATZVOCl3DI6kHEC1Vw/4Z9a6AmyuaKam2T/gny3EnsF4HpUVlb0NHn6sdomkPEKafTL7m7xsqmjmh0onZXYvLv/AHNvtu4vZVOHsMCkH2LC3DkUKpWdzN63AeN6cAWlXT02dOpWcnBymTp0alc+/87Zfk1FfCcDO8ceSmTWSpKSkmE7KQSTmMSsoy8iy3O4ulcViYd1337Z5bu03X3P05EJOPekE7rztFhobGsLbJhZOYt233+L1eFj1xedkZmWRmpbGW28sx2w2c/accw/bFkVR+N87b3PBTy4iOzu73XabLR6DIbqDM1QxflTQkAZ3kM2VbZONg+8+C7ErICvsrHWzs9Ydc3PID0cNRic5Dioqe+o9bKtpf0NLEA422M6jdm+QzZVOqpx+MQ2mn7j8CqVNPn6obGZjuZO9DR6aPAGUCB5wVVWpdwXYXOGkqdlDoJO/DbKisqE0NI1Wp8jMSJBJSI6PWDsGk6eefIKML19nxJ5NjPjXHTRXl2C321m6dGm0m9YnYih7jIqPj+eYadN48vG/UjB2HBkZGbzz1pts+H49+aNGhfc7ddZszp5zLnl5eZTs28cjD/2JX1z6M95+7wP0ej0//dklbN9WxGkzTyYlNZVnnn0ee1MTj/35EV57423+/PBD/O+dtxg5Mp+/PP5EmyHyrRoa6rE3NVFQMHYgD0GPSJK4ByVoS2uy0egJMCYtbsCWHBH6T5MnwO76zi+6Yp1ktET185s8QX6oamZMmrXD6SCCMFjOo4qqUtrkpcIhRooMJG9QCS+BqpMgwWwg0RJaMz3epO9RlXdVVWn2yzS6g9S6DqzA0dV5dFuNC7sUOredsvVrDOedg06jdWmOPfZY6uvrSUvr+zKaPXVwoTe+foekpCTcLdtan4/VnnORmMewx596mltuWsxxRx+JXq9n8hFHcv4FF7Llhx/C+8y74MLw9xMmFjKhsJCTjz+Ob77+ipNOPgWj0cgDDz3S5n1/t/gGrrjqGn78cQsff/QhH6/4nL8//TfuvutOnv3nC+3aERO90Xrtr20pDE0N7iAuv5NRaZnRborQS6qqUtrko9wxuHrrDiVFYB3zvgrIKttr3OQmmclLju6NAkF7BsM65v6gws46txi2HmWKGhqxYPcGgdC53WLUEWfUYTboMOl1GHQS+pbEWVZVZEXFG1TwBBSa/UE6Kgbf1Xl0/d768Pfzdn6N/r5fRfRniqQXX3wxKp+7bNmyA0k5B+aUH5ys33fffWRnZ8dkATjRjRjD8vNH8fpb77B9z17Wfr+Jdz/8mEAwSN7IkZ2+ZuTIfFJT09i3d2+H27/+ag07d+zg8iuvYu3XXzHrtNOIi7Nx3rzz+eabrzt8TVpaOolJSezevSsiP1d/UINijrmgXb6gStHOXWIebQwKKirbatyDPikHUBrKo92EsHK7j+01rm4tRSgMHXv27Il2E/rE6QuNChFJuTZ5AwoN7iCVDj8ljV721HvYWedmZ52bPfUe9jV6qXL6sXs7Tsqh8/NotdPP3ubQ/3t+dQnZkwtIEiOD2pkzZw7Tpk0D2hZ6W7x4MUuWLAFg2rRpzJmjzbn5hyMS80EgLs5GVlYWTU1NrP7ic8446+xO962sqKCxsYHMrKx227xeL3fdcTsPPfoX9Ho9sqwQDAQBCAYCKHLHfyh0Oh3zzr+At998g6qqqnbbXa5mgsFgL386QRgaVJWWP+ye2BiFIuAJyGypam7pUREGWqMntP6vmHcuDAb1rgBF1a5BOxVG6Np3ZQeWaJ7/zf+oO+9C4kxitOehEhMTWb58Oc8880y74eqLFy/mmWeeYfny5SQmJkangX0kEvMYturzlXyxciX795ewetUXLJx/IWMKxnLxwp8BoYT4wfvuYcP36ykt3c+aL1dz1eW/IH/UKGaeOqvd+z35+F+ZfdppTD4itDTDtOOO46MP3mdb0VZeeuFfTDv2uE7bcuvtdzJsWC7nn3s2y197lZ07drC3uJhXX3mZc844DZfL1S/HoLsknZi1IWibLj4VgEqHn111nogWnhEiz+kL8mOVa0itJdwao1ri8sv8WN2MJyB6GAXI6qDTIRZUOn3srHMPmhUchM51dB71BhR+KHcCYPV5ONlegnHyxIFuWsxITEzsdJj6woULYzYpBzHHvEsnjIzeutvd4XA6eeRPD1BVWUlScjJzzj2PW26/E6MxNPRFr9OzraiI5a+9hsNhJysrm5Nnnsrvb7utXTX3Hdu38d677/DRZyvDz5173lzWfv0V8y+Yx+gxBTz1zN87bUtySgpvv/8hzzz1JE898TjlZWUkJSUxfuJE/vDHuzXwSyL+2gkad1AiXu8OEKxRGZ8R16NiM8LAaPQE2FXn7nSo4qCl0ZtFvqDK1moXhZk20cM0xMXiaKMyu5fSpsE/FUZo0UGMbqp04m95es6GT7GfcRbJFm0PY7/ppptobGwkJSWFxx9/PNrNGTQkNRbPYr0QDAZZtWoVU6ZMQd9BIbBoVRbsiMsvx+QfFy1Tg34kgyki71VdW0eNbI3IewlCK7l2H/qM/DbPJZr1TMi0ieRcQxrcoaR8KPZsdRSjWmLUSyI5H+KKioooLCyMdjO6rbTJS5ldJOVDyaHnUVVV+dvXZdS5Q7WQXvnr1ZS+9F8mTR2HUa/dgc2TJk2isrKSnJwctm7dGu3maJ4sy2zevJmZM2d2uYS0dv/HBUEQhjiHT2abKHClGY1DOCmPBQFZpajGJYa1CzGhzC6ScgH2NnjDSfnRxZtJyc1EN3y4ppNyof+IoezCkBCp3nJB6C+61OEdPu/0yWyvdTEhQ/ScR1OTJzDk54B2FqNaEpBVttW4mJwVj8kgLmyHmrFjx0a7Cd1S6fCJ4etD1KHn0YOLvi34+h0qzj+fZIv207MVK1Ygy3KHo5CF3hN/tYShQRbLpQnapjprO93m8MrsqnOLKS5R0uwLDvmkHLqOUS3xBUPJeXCo/4cNQWVlZdFuwmHVufzsa/RGuxlClBx8HrV7g2yvcQOQ7qjjlB3fUXn6HJKs2k/Ms7Ozyc3NJTs7O9pNGVREYi4MCSKhEbRODXTde9LoCVLc4Bmg1gitPAGZ7bVDsNBbBw4Xo1riDijsrHWL1Q2GGI9H2+dIhzfInnptt1HoXwefR78rdYRLE1/47fs0nngyanISCWbRCz1UicS8hUjcBjdJEkOABW3rznSLmuYA5XbR0zJQgorK9lq3WFe4RaxNCbJ7g+wVN7OGFIvFEu0mdMoTkNlRK0beDHWt51G/rPB9mQMAQzDAT9a+R8U580gwG9CJa9YhS/tjJQaIJEkoioJOJ+5VDEr6yCw7oSiK6DkT+oWUmNmt/fY3+bAY9aTFaXsplVinqCo7aofWOuWH090Y1ZKa5gA2k4/sBPPhdxZi3ogRI6LdhA4FFZUdtW4xvUIIn0c3VzTjCYb+vpy16XOSFD81J5/GiBiYXw6wcuVKfD4fZrOZ2bNnR7s5g4bIQlskJCTQ0NCAooiLsMFIDfr7/B6KotDQ0ECTElu9RkJsUBq6PzdyT70bl19Unu5PJY1eHF5xjA/WkxjVkn2NXhzeYLSbIQyAXbt2RbsJHdpd58YjbvIJhM6jiqqydv+Bom8/W/MG1aedjWKxkBQjifn111/PpZdeyvXXXx/tpgwqsfG/PwBMJhOJiYk0NjaiqmpUhz7bvUExtD7CVL8XydS3IW6yAk2KCVkSc3+E6JIV2Fnn5ojseAyiUnvE1TT7qXL2/WaeoA2qCrvq3ByRE49JLEEkDLByu5dGj7gxJBywu84TXiJt6p5NjK/YzXdn/xGjXsJmEteYQ5lIzA9iMplIS0uLXgMcDuIvv5x9k46h7pjpNE2egmoUvbORoPr8SHpr399I5EBCP9HZUnq0vzegsKfezfgMWz+1aGhy+WX2NYp5yR3paYxqiV9W2V3nYWJmnKg5MohlZmpruoXDG6RUrFUuHERnS2HttoN7y9/En5RM/bQTSI2R3nKAxYsX09zcTHx8fLSbMqjETgQMAYZvv8X4xReM/eILxgKB+ASqTz2D8rkXUX/sCSAuJnpPHDtB63oRow3uIJUOHzmJYv5sJMiKyq46UYG9UzF+HrV7g1Q4fOQmabdAmNA3WqoTFJCVlmUuo90SQUuq3TJ7WopS5tZXcHLRN1ScPx/VaIyJ9ctbXXPNNdFuwqCknTOYgGHTpjaPjc1Ohr/3JtN/fSmnXHQGw995HSkg1uPuDaW5IdpNEIQu9TZG9zd5xXzzCNnb4BHzQLswGM6jpXYfTp8YVjxYVVVVRbsJYXvqPfjFig7CIdaWNIW//+mat9CrCpVnzAGImfnlQv8ZNIn5ww8/jCRJLF68ONpN6TXv73+PfcMGfrz3UcrPOZ9AfEJ4W/zePRx5z62cctEZZK5eEcVWCoKgJYoamm8ui2q/fVLn8lPrEjc+BztVhd31HvH7IvSraqdfzCsX2nF4g/zQGPre5nUxb/2HoWHsx56IxajDbBg0aZnQS4MiAtatW8ezzz7LkUceGe2m9I0koeTnU3HBxWz+01JWrFjHhkeeon7qceFdbKUlTLvxaqbdcBVxpSVRbGxs0aXkRrsJgtClvsSoN6BQ0ijWN+8tf1Bhb4M4foczWM6j3oDC/ibx/z0YjRkzJtpNwBOQKWkSdSqE9tbut9M6iGL+1+9g83moPvVMVKORJLPoLRcGQWLe3NzMpZdeyvPPP09KyuEL0zidThwOR/jL59NuUQ7FZKbqzPP49p+v8vULy6k/Znp4W+aXKzn5ojPJ/++/EBOYDk911Ue7CYLQpb7GaHWznyaP6PHtjT0NHrG+cDcMpvNolVP8vgxG0R7KrqoqxfUeUadCaMcTkFlX5gDAKAdZuOZNgJgdxj59+nTy8vKYPn364XcWui22oqADixYt4txzz+X000/ngQceOOz+kydPxu12hx9fccUVXH/99eTk5LBnzx4AsrKyUFWVmpoaAMaOHUtZWRkejweLxcKIESPCa2VmZmai0+nCfwzGjBlDVVUVLpcLs9lMfn4+O3bsACA9PR2TyURFRQUAo0aNora2lubmZoxGIwUFBWzbto2AJ4BqSUQymlEctQA0TprEN489Sc7H7zHpub9jqa9DH/BT+Jf7Sfv6CzYveQifLnSRoUvKQvW5UL3NIEno00ei1O9HVRQksw3JkoBiD7VXl5iBGvChekInC31GPkp9GaoSRDLHIVmTUJoqQ/smpKPKAVR3qJqkLj0PtbESVQ4gmaxIthSUxtDPpotPQ1VlVFdT6HHaCFR7NWrQj2Q0IyWkozSUh7a1VPpVXKHxPbrUXFRnHWrAh2QwISVlodSXAiDZkpEkPUpz6AJRlzIM1dWI6vcg6Y1IKTkodftD+8YlIemNKM468LtR41JQPXZUnxtJZ0CXNhy5dl9oX2vb461Lykb1OlF9LiSdDl1aHnJdCagqkiUeyWxDsVe3HMNMVL8H1es86HiXoipy6HhbE1CaOj7euvSRqI3lqHIQyRSHZEtCaezkeKflodorUYMBJJMFyZaG0thyDONTQVUPOobDUR01Bx3vjPAaxDpbSmh0Rst80dDxrkcNeNsf77hkJL0exdnZ8R6GUldy4BgazCjOlmOYnI3qaT2GenRpIw4cQ2sCktGK4qhpF7OSpEOXnodStx9VVTo+3gEPqqcbxzshAzV46PGu6DhmE9JQZRnV3VHMWpAS0g7EbEfH21l7IGYTMzs/3im5qK760BJ+BiNSUg5K/X7wu1EM5gMxC+hSclBddlS/G0lvQErJbXu8D47Z5Gy2F9eSgA+DwcD48ePZtm0bqqqSnJxMQkICpaWh/9e8vDwcDgdNTU1IksTEiRPZsWMHsiyTmJhISkoKJSWhzxk+fDhut5uGhlD7CwsL2blzJ8FgkISEBNLT09m7dy8Aw4YNw+fzUV8fipcJEyZQXFyM3+/HZrORlZVFcXExADk5OQSDQWprQ+0fN24c+/fvx+v1YrVayc3NZffu3UDonAxQXR2KgYKCAsrLy8Pn5Ly8PHbu3AlARkYGBoOBysrQ79Ho0aOprq7G5XJhMpkYPXo027dvByAtLQ23oqe+vOX/NTlHnCO6Oke4GlAUZdCcI7bVS0wYlYff33HMxsfHk5mZ2WXMlpSU4PP5iIuL09R1BEBqaipWq5XylvjOz8+noaEBh8OBXq8flOeI1uMWyXOE2Wxuc7zr6upwOp0YDAbGjRtHUVFR+Hh7VAONreeToXiOENcRnV5HfFerD9ccOHfDp6Q3N+JPSKRmdC5q7T6sWRPYu3dvzJwjnE4nzc3NNDU1sW3btpg5R/THdUR3zhHdXQZbUvuwYLbD4eCZZ57h888/p6amhmeffZbjjjuOhoYGXnzxRebNm0dBQUFv3/6wli1bxoMPPsi6deuwWCyceuqpHHXUUSxdurTdvsFgkFWrVjF69Og2VTvNZjNms7YqGq8rdXTZe6N3uxj3zF8Z9d9/hZ/zZmSx8eEnaTxo2LtwgNJYPmiGYQqDU6RiNDPeyJi0uAi0aPDzBRU2VzpF71Y3DcbzaHaCiVGpEVhKU9CEPXv2RG04uzeo8IM4nwgdCMgKj68pxeWXkVSVNx79BSPqKyg9fwFb7nkUm0nPkTmxtezY+eefT01NDZmZmbzzzjvRbo7mybLM5s2bmTlzJgZD5/3ive4xLysrY+bMmZSWljJ27Fi2b99Oc3MzELoD8+yzz1JSUsITTzzR24/oUmlpKTfeeCOffvopFkv3lz5JSEhAr9f3S5sGihxnY9vv/0jd8Sdz5JLfY26sx1JbzfRf/5wtSx6ifO5F0W6i5khJOdFugiB0KVIxWtMcIN0WjLlhcdGwt0EMOe2JwXgerW72k24zkiDmdw4K+fn5UftscT4ROrOxojm8esqMmp2MqA/1rlad3jqMPfbyEpGM949ezzG/5ZZbcDqdbNq0iVWrVrXror/gggv47LPP+tzAznz//ffU1NQwdepUDAYDBoOBVatW8eSTT2IwGJDlwb98UO1Jp7Lm1Q+oO+5EAHTBAFOW/J6x/2+pmHd+CKV+f7SbIAhdimSMFjeIqtOHU+cSVZN7ajCeR1U1lFD1YfCgoCGtw3kHWq3LT5M4nwgdkBWVrw9aIu2qt54BIJCQSN300PW7uJEutOp1Yv7JJ59www03UFhYiCRJ7baPHj06PA+hP5x22mls2bKFTZs2hb+mTZvGpZdeyqZNm2K+V7y7fBmZrHv6JUou/nn4ubHPPsGUP96MFPBHsWWCIESLN6BQ7tBuYctoCyoq+0QVe6GFy69Q5RR/L4XeCSqqWBVD6NQPVc3hm8ATJTeTi38EoHrWmahGEzoJMWJHCOt1JHg8HjIyMjrd7nQ6e/vW3ZKQkMDkyZPbPGez2UhLS2v3/GCnGgxsvf0+3MNHMuHxPyGpKrnvv43e5WLTI0+hmLQ1hz4apLikaDdBELoU6RitdPjIsBmxGofGTcqeKG3yEpBFD2lPDebzaKndS1qcEZNYRzimpaenD/hnivOJ0BlZUVlV3Bh+fNmG98PfV7YMY48369Hr2ndwCkNTr/8CFRYWsnr16k63v/322xx99NG9fXuhpySJvZddzYa//B25pZhd9hefMvX316HT8JJwA0XSG6PdBEHoUqRjVFERvcIdcPllqptF72hvDObzqKxAiVjbPOaZTKYB/TxxPhG6sqnSGe4tL0gxc+r//g9oGcZ+/AwAkmN0GPsDDzzALbfc0q0VsYTu63VivnjxYpYtW8YjjzyC3R5ahkFRFHbv3s1ll13GN998w0033RSxhnbHF1980WFF9qGkevZZrH/in8gtBfEyv1zJMTf/Cp13aF9wtC4/JQha1R8x2uQJUu8WazUfbF+DR5Tg6KXBfh6tcwVweMU84VjWumTRQBHnE6EzQUVldXFT+PGFjj0YXaEi2dWnnoFqDN1ESozRxPyVV17hn//8J6+88kq0mzKo9Doafv7zn1NSUsJdd93FH/7wBwDOPvtsVFVFp9Pxpz/9iQsuuCBS7RR6oH76DNY99QLTbrgKg8dNxtermfr7a/n+8edQjYO3x0MQhPb2N3pJsRrQdVALZKipc/lx+AZ/YVCh9/Y1ejki29Zh7ZyhwBdUaPAEcHpl3AGZoKKiqmDUS1iNOhLNBlLjjJjFkH9xPhG6tLHCSVPLjb6xaVZmvPlWeFvlGaFh7HodxJvEdDPhgD7dpvnDH/7AZZddxhtvvMHu3btRFIUxY8bwk5/8hNGjR0eqjUIvNEw7nnVPv8ixv70Cg9tF5ldfcOTdt7D5gb+Cbuj9QdWlDL5lfoTBpb9i1BtUqHT4yE3q/rKSg5EsCjT12VA4j7r8MrWuAJnxAzskOtpcfpkyu5dGT7DDHuCgouIJKDS4g5Q0eUmyGBieZNZc0apRo0YNyOeI84nQlVBv+YG55bPz4slctQKAQHwCdcefBECi2RCzNwFff/11AoEARtHhF1F9PqPm5eUN+JB1oXsajz6WdU/9i+Ou+wV6n4/cD98hkJhE0W33QIyeCHpLddmRkjKj3QxB6FR/xmi5w0dGvAmTfujdlGtV4fDhFwWa+mSonEdLm0KF4IZCQaZQRXEPNc3dn/KiqqFpMk2eIGk2I6NSLBg1cm6pra0lLy+v3z+n0inOJ0Ln1pc5wqMpxqXHcdTWbzE2h4piV804OTyMPZaXSSssLIx2EwalXp9JjzvuOB5//HHKysoi2R4hwhqnHsfGR/6G0rJ8XP6r/6bguaei3KqBp/rd0W6CIHSpP2NUVkLJxlDlDypUOkURzL4aKudRv6wOiXixe4NsrnD2KCk/VL0rwObKZho92qhl0dzc3O+f4ZcVKsRylEInPAG5TSX2WWNSyP70g/Djihknhb+P5cRc6B+9Tsz1ej2/+93vyM/P56STTuJvf/sbVVVVkWybECE1M09ny92PhB+P+3+Pk/vuG1Fs0cCT9OLkJ2hbf8dorSuAyz8050OW2r3ISrRbEfuG0nk0NMJi8AZNpcPHthpXRHp9A7LKjlo35fbo3/wbiGG1pU3ifCJ0bvXeJtyBUIAcmR3PcAtkrfoMCA1jrz92OhCq2xAn5pcLh+h1Yv7NN9+wb98+HnroIXw+HzfccAMjRoxg9uzZPPfcc9TVDe7qrbGmfO5FbLvpzvDjI+67g5Tvv41iiwaWlJIb7SYIQpf6O0ZVFfYPwV5zd8ucYaHvhtJ5VFagrGlw9orubfCwr9Eb0WriofOLj+J6D2oUy5QXFBT06/t7AuJ8InSuwR3g2/2hlaoMOonTClJJX7smPIy9eubpqFmhOgix3lu+adMmvvvuOzZt2hTtpgwqfZoUlJeXxy233MK6devYvXs39913H42NjfzmN79h2LBhnH322ZFqpxABey+7mpIFPwdAFwxwzO9+Q9z+fdFt1ABR6kqi3QRB6NJAxGiTJzjkloPa3xTZBGQoG2rn0RqXH09gcI0y2VPvpsrZf+tuVzf72R3F5Hzbtm39+v7ifCJ05bPdDbQOQjkhL4lkq6HNMPaqM+aEz6OxnphfeumlnH322Vx66aXRbsqgErFqHaNHj+aOO+5gw4YNPPvss1itVj799NNIvb0QCZJE0a13U3vCyQCY7E1Mu+FKDA57lBsmCMJAKRlCveYOb5BGz9C6ESFEjqpCqX3w9JrvbehZkbfeqnMFKG7w9PvnDDSnL0iDW5xPhI7tb/KytdoFgM2k56RRyej8PrK+COVCgfgE6lquvyH2E3Ohf0QsKtauXctrr73G66+/TkVFBfHx8VxyySWRenshQlSDgY2P/I0TLp9PQvEu4kv2ctQdN7L+yX+CfvDOdZGsidFugiB0aaBitNkn0+AOkBo3+Jc4GcoF7/rDUDyP1rsCuBJlbDE+F7TM7u3XnvJD1TQHMOq85KUM7DKNqamp/fbepYN0aoPQd4qq8tGO+vDjWaNTsBh0pH35VXgYe83M01BMZiRrIhajDrNBGysZ9NYvfvELHA4HiYlD7+9Cf+pTYv7999/z6quv8tprr1FaWorVauW8887jpz/9KXPmzMFsNkeqnUIEBRMSWf/Uvzjx0vMxNzWQ+fUqxv2/x9m56PfRblq/kYwiFgVtG8gYLbN7SbHG7vqp3dHoCYSXqxEiY6ieR/c3eZmYaYt2M3qtzuWPSlJZ7vBhMeoGdE14q9XaL+9r9waxD7FpQEL3rSt1UN5SqT/DZmRqbgIAOQcNY688fQ4QOo8mD4Le8ttuuy3aTRiUen27ZsyYMRx33HE89dRTTJ06lZdffpmamhqWLVvGhRdeKJJyjfMMG87GRw8so1bwj6fJWvlxlFvVfxRHbbSbIAhdGsgYdfkV6t2Du4CR6N2KvKF6Hm3yBHH6YjMpc/qC7KmP3rDy4gbPgNa1KC8v75f3FaNvhM44vEFW7G4IP547MR29TkIK+A8MY7fFU3fCKUDoPJo4CBJzoX/0OjEvLCzkpZdeoqamhjfffJOf/vSnxMXFRbJtQj9rOPYEdtx4e/jxlD/+Dlvx7ii2SBCEgVJq90W1enJ/anAP3aXhhP4Rizd6ArLCrjo3ShR/zVUVdta58Qdjd32xRncApxh9I3Tigx11+Foqvk3NTWBkSmjURvrarzA6HUDLMPaWDktJEvPLhc71OjF/9913+fnPf05CQkIk2yMMsL0/v4qKs+YCYHC7OOb3v0HvcUe5VZGnS86OdhMEoUsDHaPegELdIO01L9PAesqD0VA+j9q9sbWigaqq7K734AtG/+ZbQFbZVe8ekBuB+fn5EX/PwVQAUIis7TUuttWErpltJj1njj1Q4yDns/bD2AFsGbkYdIN3GpnQN92+ZbN//34gtETawY8Pp3V/QaMkiS13P0x88S4Sd20nfu8eJj20hB/u+0u0WxZRqseJZBzYIjSC0BPRiNEyu4/0OOOgmmte7w7g8sdu75yWDfXzaJndR2GM9HRVOPw0aWhFAodXptzhY3hS/8ZPQ0NDREdvitE3Qme8AYX3t9eFH58zPg2rMTQ9VAr4yfr8E6BlGPuJM8P7GXzNQPqAtrU/nHPOOdTU1JCZmcmHH34Y7eYMGt3+C5Ofn48kSXg8HkwmU/jx4ciyOKFpnWyNY+Ofn2bGJfMwuF0Mf/cN6qedQPm8i6LdtIhRfS4gI9rNEIRORSNGvQGFWldgQIsz9bdy0Vveb4b6ebS111zr80NdflmTo0bK7D6SLAYSzP13/BwOR0Tfr0z0lgsdUFWV97bXhguMFqRZmZx1oEBk+rdfHxjGfsqBYewAstc1sI3tJ/v376eyshKvV3vnmljW7bPjv/71LyRJwmg0tnksDA6ukaP58a4HOerOxQBMeuiP2CcdSfOYsdFtWIRIuthe6kYY/KIVo+UOHxm2wdFr3iB6y/uVOI9qv9dcUdWozyvvjKrC7noPR2bHo++nobz6CC772ih6y4VObK5sZktVKMG2GHTMnZje5m9o9kHD2KvOODCMXa8Ds0m754+eSElJwe/3k5KSEu2mDCqSOlir/xwiGAyyatUqpkyZEtETd39YV+ogGKW/qkfcexsj3n4NAOeYcXz1n7dR+mn5EUEQtGFsupV0W+z3mv9Q2SwupIV+NynLptle8/2N3vCyTVqVk2giP0X71xVbqpppFkXfhEPUuwP8v7Vl+FsKvi04IpPJ2fHh7VIgwGmnH4vJYScYZ+OzFetRLKEpHMlWQ0wvvSj0nizLbN68mZkzZ2IwdP73o9fF36688kq+/fbbTrd/9913XHnllb19eyFKtt56D86C8QAk7NlJ4Z/vjXKLIkOuK4l2EwShS9GM0cEwXFP0bvU/cR4N0Wri2+wLUuHUZtsOVuX099vyc9u2bYvI+zR5AiIpF9qRFZU3ttSEk/Kjh8W3ScoB0r77GpPDDrQMY7ccqKuQbDFELEaFwanXifmLL77Inj17Ot2+d+9eXnrppd6+vRAlitXKhkf+RtASupud99arDPvwnSi3KgKGxsAQIZZFMUY9gdhf11yrydKgIs6jQGhd82aNrWuuqirFDZ6Y+C9SVSiu96D0Q2MjNQi0fBDcrBQi75Nd9eG/NWlxRs4Z376IW86nB1VjP2gYO0CixTBolykVIqPXifnhVFRUYBVDoGOSa3QBW++4P/x48gN/wFrWvSr8WiVZxLJ+grZFO0Zj+ULU7g2KdYYHQLRjVEu0diOowuGPqfoK7oBCZT8cw+Tk5D6/h8MbDBf1EoRW68scrN0fKuimk+CiyZmYDW3TKCkQIOuLUDX2YJyN2oOqsRv1EjaTPiIxKgxePZok9c477/DOOwd6T5977jk+++yzdvs1NTXx2Wefceyxx/a9hUJUlM+7iLR1XzP8vTcxuF1Muetmvv3HMtQu5kVomWQSN4kEbYt2jLr8Mk2eAMlWY1Tb0RsVGkuSBqtox6iWNLiDeAJyeHmkaPIFFcodsVcZudzhI91mapfc9EVCQt9vHmntposQfcX1njZLo507IZ3cJHO7/dLWfY3J3gRAzcmz2w1jh8jEqBY8/fTTOJ1OEhISWLRoUbSbM2j0KMsqKiri9ddfB0CSJL799lu+//77NvtIkoTNZuOUU07hr3/9a+RaKgy4otvuIXXjOuLKS0nd/D1j/vUMu391Q7Sb1SuKowZ9Rn60myEIndJCjJY7fDGXmDf7gppar3kw00KMakm53UdBeuTWzO6tfY0e5NjpLA+TlVDbx2dErhhWaWkphYWFvX596AalOJ8IB9S5/Lz6Q3V4pYMT8pKYNjyxw30PHsZedfo5bbYlWUMpV19jVCueeeYZKisrycnJEYl5BPXoNuUdd9yB0+nE6XSiqir//Oc/w49bvxwOB5WVlbz33nuMGzeuv9otDIBgfAKbHngcVRcKk4LnniRpy8Yot0oQhP7i8Mr9VpSpv1Q4/NFugjBE1bkD+ILRzYibPAEa3LH1O3uwBneQRo926lvE8pQeIfIc3iD/t7EKb8vv+bj0OM4cl9rhvlIgQNbnLcPYLVZqZpzaZnuSOTZHnAoDq9dRoigxeHtW6LGmo45h99W/ZexzT6KTZY76w82sWfYeclxsLfegS8qKdhMEoUtaidEKh4/xGbFxAeEJyDRo6KJ+sNNKjGqFqoZ+X0alRmeIv6Kq7GuMvSHshypp9JJkMaCT+r62eV5eXq9fK84nwsFcfpl/b6iksWUERVa8iflHZHYap2nfrz0wjP2U2W2WGraZdJhapmz0JUa15Nlnn8Xn82E2tx/SL/Reryf2bNiwgWeeeabT7c888wybNm3q7dsLGrL76t/SNHkKALbSfUz8y/2HeYX2qD5XtJsgCF3SSoy2zp2NBRUOX0xUoR4stBKjWlLr8hOI0jjyaqcfTyD2O0k8AYVqZ2RGvjgcjl6/VpxPhFYOb5AX1ldQ6wrdqEmxGrj06Owu6yFktxnG3rYae5LlwM3uvsSolpx00kmcdtppnHTSSdFuyqDS68T8D3/4Q4eF31qtXLmSu+66q7dvL2iIajSy6cGlBK2huXR5b71K1sqPo9yqnlG9zdFugiB0SUsxGgvF1PyyQp1L9G4NJC3FqFbISmhd7oEWkBXKBtGw6zK7LyI3OJqamnr1On9QnE+EkHp3gH8dlJQnmvX88picNsn1oaRgkKyVBw1jP2lWm+0Hv7a3MSoMDb1OzL///ntOPvnkTreffPLJrF+/vrdvL2iMOy+foluWhB8fcf8dmOpro9iiHorAEDlB6FcaitE6VwB/lOfOHk6V0x8uxiMMEA3FqJZUOf3IAxyMZXYfwUH0CxBU1IhUQ5d6GaPifCJAaFrFP74rDw9fT7YYuGLaMFIOUxQ1df1azE0NANSePKvNMHa9LrR+eavexqgwNPQ6MXc6nRi6WDpLp9Nht9t7+/aCBpVdcDFVs88CwNTUyBH330msjPvSp4+MdhMEoUtailFFjU4vYHfJihqxoa9C92kpRrUkqKjUuAYuHj0BmZrmwRf/oaH5fZtGM3HixB6/RlZUqpoHz+gDoedUVeXb/XZe+r4Cd8v0kEybkauOHUZq3OFXKsn57MAw9soz2g5jTzC3rZ/QmxjVopKSEoqLiykpKYl2UwaVXifmY8eO5ZNPPul0+0cffcTo0aN7+/aCFkkSP/7hQXwpaQBkrfqM3PfejHKjukep3x/tJghCl7QWo9XNA98L2F01zf5B1VsYK7QWo1pS6fChDtCN6tIm36Ds3VVUKO3j8PwdO3b0+DWhc12fPlaIYS6/zGs/1PDBjnrklt+rMalWrjo2t01Pd2ekYJCsFaHpnbLFQu0hw9iTD3mP3sSoFs2ZM4dp06YxZ86cw+8sdFuvE/OrrrqK999/n5tvvrnNfImmpiZuuukmPvroI6666qpItFHQEH9qGj/+8U/hx4WP3oulqiKKLeoeVawiIGic1mI0qKia7JVTVVXTvfmDmdZiVEt8QZV6d//PUXb6ggPyOdFS7wrQ3IclG2W5Zz3uiqpSGQM1NYTIU1SVzRVOnvmmjKKaA4UtTxyZxKVHZ2Mxdi9FSv3+2/Aw9pqTZiG31GNqdejc9J7GqDC09HpNnBtuuIFNmzaxdOlSnnzySYYNGwZARUUFiqJw2WWXcdNNN0WsoYJ2VM86k7JzL2T4+29hbHZyxL23se6Zf2t6/qFkjq3l3YShR4sxWuX0k51g0tScuAZ3MLymrDCwtBijWlLh8JNuM/XrZ+xviv3l0Q5nf5OPwqzeXZ4mJib2aP86VwC/PAiHHwhdKrN7+XBHfZsCinFGHfMKM5iY2bPzXM6nnQ9jNxsk4kz6Ns/1NEa16rzzzqOpqYnk5ORoN2VQ6XViLkkSL7zwAr/4xS944403KC4uBuD888/noosu4tRTT41UGwUNKrr1btLWfYO1poqMtWvIe/2/7L/459FuVqckS0K0myAIXdJijHqDCg2eIGndmGM3UCqconcrWrQYo1ri8svYvcEuqzf3RaMngMM7+Hvb7N4gTZ4AyYcpuNWRlJSUHu1fJc4nQ4aqqpQ0evlmv53tte422yZkxHHexHQSzD373Q1VYz9oGPvJs9ts7+hc0NMY1apHHnkk2k0YlPr812PWrFnMmjXr8DsKg0owMYktdz/McYsuB2DC43+i7oSTcY/QZnEgxV6FPiM/2s0QhE5pNUYrHT7NJOYOb5Bm3+BPTLRKqzGqJRUOX78l5qVDoLe8Vand16vEvKSkhMLCwm7t2+QJ4PKL0TeDndsvs63WxbpSB5WHTIPKsBk5Z3waY9LiOnl111I3fIe5sR6AmhmnthvG3lEM9yRGhaGnf/56CL2ytbqZVzdX0+QJdl7svIsRpX0ZbNqrkaq28STe/k/iyvcjqeD/ZBMNx5h7/GZ9GSXb3ZeqPgWpsrqHn9v5Dn14ad/+n/r1c7veo8uth3nzrjbrJAmd1PKvDvQHPdbrDtomSegl0OkkDJKEUd/6pcOolzDpdG2eM+i0M/w6ljl9Mk5fsMc9Cf2hUvRuCRrX5Ani9svthq/2Vb1raCWRzT6ZBnegWxWxe+vQJE0YHBRVpdYVYF+jh+01bvY1etoVS0w06zkpP5lpwxPR9+FaIfugYexVZ5zbZpsktS/8JgiH0+2IGTVqFDqdju3bt2M0Ghk1atRh5x1KksSePXv63MihosEdZO1+R7Sb0TOp+aGvVocMD9IW1+F3EQYNg07CatSFvgz6A98b9cSb9CRaDCSa9SSYDSSY9Rj1va6FGRG6xIyofn5XKh1+EjKie4HhDSrhtWWFzimqii+o4A0qeAIK3oCCNyjjl1WCikpAUQm2fK+qassNSglJCt1AC/0rYdBLmPUSJr0u9GWQMErJWFwB4kw6LAadpmoPaEml09frHriOqKpKqX3o9Ja3KrN7e5yYDx8+vFv7ufwyTeJ80iOqqqISWl5OUUPnGkU98Dh0E/3AjXSd7sCN99C5JbLni4Cs4PIrNHkD1LsC1LsDVDf7KbP7Oq1DMizRzAl5SUzKsvUpIQdAlsle+VHoW7OZmpPbjh5OMOs7/IzuxqgwNHX7SmvmzJlIkoROp2vzWBAEQYuCitrS2ysDh69iHGfUkWQxkBpnJC3OSHqckTRb6HurMbK9Xx1RAz7NFtdq8ATwBhUshujdvKhy+jofSTSEyIpKoydAXcuFqMMXxOGVcfiCOH1BnD65n5fSClUfNugkbCYdNpOBeJMem0lPgllPssVAitVISpyBRLOh7xe/MajOFWBEsoIpQjf76twBPIGh01veyuVXqHcFSLN1Pzl3u93dKq4lKrGHElu7N0hTyxQhl1+m2R/61+WX8QWV0Jccutnnl5U+nVv0B42Eax39pj9oZFyb71u26SSQVQgqSviGYkBRcftlfN0s2pdiNTAx08akLBu5ieaI5S6pG77D3BAaxl47YxZyXNu/38mWjuO2uzGqdZdccgl1dXWkp6fz8ssvR7s5g0a3E/MXX3yxy8dC303PS+S1SyezqaK5wzV6uzwFHeaKtS/Xad25GC549gny3lwGgHPsBNYvfQ7VYOznz+18h0O3KPVl6NK6d5eyrxf/Xf8/He61vf/wvn1u7/XhvwkVFVUFueXOu6KoB75vcze+7Z35oKoSkBUCstrypeBXDjznl1W8QRlvINRrGOjG1YQ7oOAO+Dsc3phsMZCTaCInwRz6SjRFfGi36nFAfGpE3zNSVDWUGOenWKPy+Vpduq2/OX1BKhw+Khw+Kh1+6twBGj0BTaxhHVRU7F4ZexfFyHRSqPhRhs1Ehs1IRryJTJuRdJsJcxRv8vQ3RQ2taJCXbOnze6mqSlnT0E0iS+1eUuMM3U6mGhoayM7O7nIff1AZ1EvOHUxRVZo8QWqa/dS4/NQ2h84hTZ4gTv/A1uuQW/7Wd+fvcV/Em/SMSDYzIsnC6FRrv60s0lU1doBka8fXCN2J0ViwefNmKisrycnJiXZTBhUx+UFDTHodJqsOm0nfYWKuZXW/XsT4VR+RsHc3md+vQf7v8+z+9Q3RblaYbJbQ96KQjDA4BOQDQ3vd/lAvusMXDPc2On1BHC1zqTv61Wtq6VXYVnNgqkaK1UB+ipX8FAv5KdZO/wgPFjXNfkYkWaLSA1rb7Ece5B2Gihq6+VDc4KGk0UuFw4ejh4XubEYdCRYDNpMeq0GHxajD0jKNw6TXYdRJGHShYeqtvVEqB25Gtt4oU4GgrOKTQ71k/qCKX1bwNdvxGePDPWrNfhm3X+703puiQqMnSKMnyM66ttvS44wMSzQzLNFMbpKZ7ARTxHqYtaDa6Sc30dzn35daV2BILw/oCYSS6EguQ1fl9Gvi5lakKapKnStAmd1LuT10Q6/WFehTImxqmc5iNugw60PnjjY1Xw4asq4edEO97b8HbqzLqoqihBL01hvtrc+3Pu6qxJJRL7VMUwuN0rGZdCSaDaTZWka5xRlJsnT/Rk6vHTqM/ZS21djNBglbhOtMCENDt68kV69e3asPOOWUU3r1OiG2KBYLP9z/F0745UXoZJmCfzxFzSmzcUycHO2mAYhKwkNcqECcjgRz1/u1DhOudx8YKlzr8lPl9Ldb6zaUcDjZWOEEQr3qY9PjmJARR36qtcfF57Qeo7ICNS4/OYc7iBGmqipVg7RIk8MbZEedm70NHvY2eHAfZriyUSeFp1e0TrVIthhJtIRqJfR/wcO0ds8oamhYaXPLDa9GT6DldyPUK9fgDnQ45LTOHaDOHeCHqmYgdNGdk2Air+VGV16yJaYvbIOKSq3LT3Yffl8UVaVMC3PLVRWdz4fe40bvCd2cVMxmFJMZ2WJBNfbv2u2l9tDKEN1Jtg5X7VpWVKoHyeiboBKKj70H3czr7vDueJOeZKuBZIuBZKuRRHMo0W2dlmIz6bEYdeiiMGVVaUnS5ZbEXi+FknEtTYtJ3bgOc33obmPtjFO7PYwdDh+jsWLr1q3RbsKg1O3E/NRTT21zUgwVjTn8L4ksi6Vthgr7pCnsufI6xj7/FLpgkCl//B1fvfw/FNPAXsh3pCdD2YWhS6+TSLeZSLeZGH9QLTZFVWlwB6h0+Kl0+iiz+yizezn4GqjJG2RdmYN1ZQ7MeomC9DjGZ8QxPt2GxXj4nsBYiNEqp5/s+P4ZFtiZBk9wUPUYNnoCFFW7KKpxUWbvfIiy2SAxLMEc7lUelmgm2WqIyoVyq45iVCdJxJsNxJsNZHewzLmqqjh8MrUtQ2lrmkMFmqqdvja/PypQ4fRT4fSHi6Bm2IyMTrUyNj2OkSmWmOtRr3T2LTGvbQ7gCw5Q164sYysrIWHXDhJ2bcdWshdLdSXWqgrMdTXogp0XSvMlp+LNzsGbmU3zqDE4xk/CMWESrrx80Pf95oo3oFDnDpDRjV7znTt3Mm7cuE6317j8MTcisVXrTcrd9aEbefubvF32hktAapyRDJuRzHhT6MtmIiXOoOnfJZ0kodNLaHmMY86n74e/rzy9+8PY4fAxKgxt3U7MP//88zaPfT4ft956K263m1/96leMHz8egO3bt/P8889js9l49NFHI9taQfN2X/NbMlevIGlHEQl7djL274+z48bbo90sVEVUXxV6TycdSNiPyIkHQsPjy+w+9jV6KWn0UGr3hS/4fLLK1moXW6tdGHR1TMyM46hhCYxOtXaaWMVCjHoDocro/bmE0aGqBkGRJpdfZlOFky1VzZ0u0WQ2SOSnWBmVamV0qpUMmzGqSXhHehOjkiSRZDGQZDFQkH6gUnlr3YDWOfRldl+7nsxaV4BaV4BvSx0YdBL5KRYK0uIoSLeS3s0e1GjyBpReL/mlqCrljv7rLdf5fST/sJHUDd+RsnEdKT9sxODu3col5qYGzE0NJG3fStbqFeHnA/EJ1B97AnUnnELtiafgyR3R6/aW233d+j8PdnEDIRZH3wRkhb0NXnbUudhZ6+5yekuiWc/wJAu5SWaGJ5rJSTQP6loOUSPLZK9oGcZuMrUbxt5aW6MzXcWoIPSoKvvBbr75ZkwmE2vXrsViOVDgZO7cuSxatIiZM2fy0UcfccYZZ0SutYLmqUYTP9z/GDMumYcuGGD0S89RPfMMmo46JqrtksyRW7pGECA0PH5UaiiRghR8QYU99R6217rYWecOV1EOKipbqlxsqXKRaNZzZE48U3MTSTvkYj1WYrTS6RuwxNzll3s8z1orFFWluMHD92VOdtS66GiEaWa8kcJMG2PT48hJ6Pt85P4WyRg16KTwSIBW7oBMaZOXfY1e9jeFhua2dggGFZXd9R5213tgZ2jqyITMOAozbYxItmjuJkar3v6+9Edvubm2hswvV5Dx5eekf/sVBs/hlzf1J6fgzcwmkJCIbI1DtlpRJQmd34/e78PQ3IylpgpzbTW6Q0ZIGpudZH/+CdmffwKAfeJkys/7CRVnz8Wfmt6jtnd3rnlCQgfDNlo0eoJ4Y6C6fUBW2FXnZkuVi1117k57xZMseka13MwblWrtMhkUIqfdMHZbfJvtiZauV6PoKkYFQVLV3tWgzsrK4q677uL666/vcPuTTz7Jgw8+SHV1dZ8a2JmHHnqIN998k+3bt2O1WjnxxBN55JFHwj33hwoGg6xatYopU6agj8DQqv60rtQRs0OtWo3+19+Z8FRoxIRrRD5rXn0f2Rq9xEMN+JCM0R9SLwwNshJad3hrtYstVc3tljqSgHHpcZwwMon8FAuSJMVUjB6ZEz8g839317mpdcVW9WRPQGZdmYPvy5w0edv3jAxLNFOYaWNiZlxEC1oNhIGOUW9AobjBw656N7vrOu8tjDfpmZARx8QsG6NSrJq7wdHT3xdFVdlU4YxIYm5qqCf7sw/J+eQ9Ujd8h9TJJZ83I4umyVNwjpuIY+wEmkePxZM9DMXazZUYZBlLXQ0JO7eTuGMrSdu2krLhO8xNDe12VfR6ak+aRfFlV9M49Tjo5k0Vq1HHlJz4LnvNPR4P1k7avLWqWbM3+oKKSnG9hy3VzWyvcbWraQKhm1mjUiyMy4hjTFocqdYBKHImtDPpT39k5Ov/B8DGh56k8uy5bbaPSrV0OYWlqxiNJS+//DJut5u4uDguueSSaDdH82RZZvPmzcycORODofObaL2+veZyuaisrOx0e2VlJW734e/G9taqVatYtGgRxx57LMFgkDvvvJMzzzyToqIibDZtrgU8lBT/8ldkffEpKVs2Yivdx/gnHqHo9nuj1h6lqVLzxbWEwUOvk1oqtls5a1waO2vdbKp0sqvOHa46u6POzY46N1nxJk7IS6LQUIc5a1S0m94tVU4fY9L690abX46tJY2aPAG+2W9nQ7mz3UW1zaTnqGHxTB2WEHPJ+MEG+jxqMeoozLJRmGVDVVVqXQF21bnZVe+mpNEb7k1v9susL3eyvtyJxaBjfEYck7PiGZ3W8yKM/aHC4WNsevd/X/raWy4Fg6R/s5oRby4j88uV7XqyAXwpadSeeAr1x82gYeqxoWHmfUny9Hq8WTl4s3KoPXlW6DlFIWnbj6R/s5rslZ+QtG0LADpZJmvVZ2St+oymSUdS/MtfUTX77MPOR/cEFBrcwS7XNd+7d2+HxbW0OPpGVVVK7T42VTgpqnF1uFa9zahjXIaN8RlxjE61iqHpUSYFg2Sv+BDouBo7QMphVgDqLEZjzYMPPhheLk0k5pHT68T89NNP54knnmDatGn85Cc/abPtjTfe4IknnuCss87qcwM789FHH7V5/OKLL5KZmcn3338vKsFrgV7PD/f/hZMWnove6yX/1X9TPetM6qfPiHbLBGFAGXRSOLlo9stsLHfyXak9fJFY3ezn7aJaPjPAyT47x+QmYNRwYR6AOleAvGSlX9tZHSNLGlU7/Xy5r4mt1c1t2isBBelWjslNZFx6nOZ6cWONJEnhAlYz8pPxBGR21ropqnGxu94THmXmDSpsrmxmc2UzFoOOCZmhJH1UL1ZKiJQGdwB/UMHUjaSqL3PLLRVljHj7NYa/8zrWmqp225tHjqLyjHOpmXka9sIjQdfP5xmdDvukI7FPOpI9V/8WW/Fuct9/i9z33gy3L3nrD0y99bc4xk5g+423U3fiKV3eICize7tMzDtToaFaFa01JzZUOKnrYESQxaBjYqaNI7Jt5GtwBMhQlrrhO8wN9QDUnDSrXTV2m0knbp4IfdLroezl5eXMnj2b3bt3k5OTQ0FBAQB79uyhoqKCMWPGsHLlSoYPH5gqw7t372bs2LFs2bKFyZPbL9HVOpR99OjR6A76Y2Q2mzGbtTV8dDAMZW818pUXmfRoqKfckz2ML1/7kGBC4oC3Q/U2I1niD7+jIAwAWVHZVuPim/32dpW5E8x6Ts5PZqrGE/S8ZDO5SZbD79gLiqqyodxJoJtL/0RDncvP53sa+bG6bcEsg07i6GEJnDgyaUCL5A0ErZ5H/S1zcrfVhApkdbRklNWgY0KmjclZNkalDnyyk5toJi/l8L8vNc1+9tR7uv/GqkrKhu8Y9X//JGvVZ+2Gqnszsig/9wIqzp6Hc9zEvvWKR4gUCJDz6fuMfuk5Endua7Ot7rgZbLv5TpzjO+9RHJ8R1+nvVlNTE8nJyW2e8wcVNlY4o3qjr7XmxIZyJ9tr2tecMOqk8E2kgvQ4TYz0ENqb/MAfyHvjZQA2PPo3qs44t8323CQzecld/553FKOx6N133w0Py587d+7hXzDEdXcoe68TcwCv18uzzz7Lhx9+SElJCQAjR45kzpw5XHPNNQM2h0JRFObNm0dTUxNr1qzpcJ/WxHzevHlththfccUVXH/99eTk5LBnzx4gNH9eVVVqamoAGDt2LGVlZXg8HiwWCyNGjGDXrl0AZGZmotPpqKoK3f0dM2YMVVVVuFwuzGYz+fn57NixA4D09HRMJhMVFRUAjBo1itraWpqbmzEajRQUFLBt2zaaPAFUSyKS0YziqAVAl5yN6nGi+lxIOj26tBHIdSWgqkiWBCSTFcURaq8uKQvV50L1NoMkoU8fiVK/H1VRkMw2JEsCij3UXl1iBmrAh+oJLU+jz8hHqS9DVYJI5jgkaxJKU2jKgi4hHVUOoLrtocfpeaiNlahyAMlkRbKloDSGfjZdfBqqKqM6Gzj+9t+TsWkjAPvPOpsfbl+ClJCO0lAe2teWEvp/dDWGHqfmojrrQvMZDSakpCyU+lIAJFsykqRHaQ7dsdSlDEN1NaL6PUh6I1JKDkrd/tC+cUlIeiOKsw7kALq0PFSPHdXnRtIZ0KUNR67dF9rXesjxTspG9bYebx26tLyDjnc8ktmGYq9uOYaZqH4Pqtd50PEuRVXk0PG2JqA0dXy8dekjURvLUeUgkikOyZaE0tjJ8U7LQ7VXogYDSCYLki0NpbHlGMangqoedAyHozpqUIN+JKMZKSEDpaHswPGWJJTmhoOOdz1qwNv+eMclI+n1KM7OjvcwlLqSA8fQYEZxdiNmrQlIxo5jVpJ06NLzUOr2o6pKx8c74EH1dON4J2SgBg893hUdx2xCGqoso7qbWo73CFR7dcsxtCAlpB2I2Y6Ot7P2QMwmZnZ+vFNyUV31qH4vksFIGSms2VXFjtB/c1iCEU4ZlczUpABSwIOkNyCl5LY93lE8R0hqkOGZqWSkp7N3714Ahg0bhs/no74+FC8TJkyguLgYv9+PzWYjKyuL4uJiAHJycggGg9TWhto/btw49u/fj9frRTWYsRuSBvYcAeiScw57jmisq2F1lcqmhtCUhFZxRh3T0lSOTYf4+ITBeY7wOkPt1fA5Iqio7JWT2FrtYntDAH8Htb6sBokJiSqFKRKjczLQyf5+P0eY0keQpjTh83o7vY6orKzE4ZNRk4a1OUdISTko9W1jVm2sYtjqLxj9ztskb2u7nrCq01F93PHsv2ABdSfNQnY3Hv54R+M6QlXJ3L2f8U8+TPKO7eH2KzodxT9ZwK7rbkY2m1BdbY+3XgmQlZJAbm4uu3fvBkLXbQD79+/HbDZTUFBAeXk5Ho+HAAZc1vQBO0ccfB1ht9vZWOVmY4OKvYOC8Hk2mJomUTgiA6Psi/1zxCC+jqC5kTMWXoTZbidosfDJq2+iJKW2OUeMzsvFKKlt/q6VlJTg8/mIi4sjJyeHoqIizGZz1HMNgNTUVKxWK+XloRjIz8+noaEBh8OBXq9n/PjxbNu2DVVVSU5OJiEhgdLS0P9rXl4eDoeDpqYmJEli4sSJ7NixA1mWSUxMJCUlJZwbDh8+HLfbTUNDKF4KCwvZuXMnwWCQhIQE0vvhOsJqtXZ4jmitfXbwOcJisZCXl8fOnTsByMjIwGAwhKdsjx49murqalwuFyaTidGjR7N9e+iclZaWhtlsbnO86+rqcDqdGAwGxo0bR1FREaqq4vP5+jcx14prr72WDz/8kDVr1nTaQy96zKPHUlHGyRefg9HVDMD6pc9TM/P0AW2DXLtPzDEXNK2sZC9fNlrZXtu2NkeGzciZY9MYm27VXKGfsenWfpkz/UOlE1dHGVUUuQMyq4obWVfqaNPbZTPqOGlUMtOGJ2p6beBIiLXzaEAOrZTwY7WLHbUdF9SKM4aGDU/OimdkiqVfe9LzUy3kdFEUqju95Tqvl7w3X2HUS8+1G67uycxm/4JLKZs3H19mdkTaPCAUhZxP3mf83/5MXHlp+Gl37gh+vOtB6o4/ud1LJmTEkdJBr3lRUVGb+buyEhp9M5DXVLKisrPOzfflDnbXeTj0kwdLzYmhJn3tlxx37S8AqDjzXDY98rc22016iWOGH35E6KExKgwN/V78rZXP52PDhg3U1NQwY8YM0tN7tgRGX/32t7/lvffeY/Xq1d0aNp+QkKD5quyDjXfYcLb9/o8cee9tABxx/x2sPnIqgZTUKLdMELQjJ07iZyOzqXD4+KK4kR0tCXqtK8B/N1UxOtXKWeNSu6z2OtAqnf6IX1g6vEFNJeWyorK+3MHnexrbFGeyGHScODKJ4/OSxJxCjTLqQ8PXJ2TaCMgKu+s9/FjVzM46dzhJdwcUvi938n25E5tRx8SsA0l6pJdgq3L6yY43dXiDTVVVyruYB63z+xjx1quM+eczWGrbrnZjnziZvZddTeXpc1CNMTh9Qqej8uy5VM8+k1EvPUfB839DH/ATV17Kcdf+gn0Lf8n2xXegHNSJUu7wdZiYH6rG5R+wpLzeFeD7CgebK5pp9rctNCcBBWlWpuYmMi5DDFWPRdmfvB/+/tAh7MCgm7okREefesyffPJJ7rnnnvAwhk8//ZTZs2dTV1fHhAkTePTRR7nyyisj2d4wVVW5/vrreeutt/jiiy8YO3Zsl/uL5dKiTFU5ZvHVZK1eCXR8t7F/P15BksTFs6Bdh8ZoaZOXj3bWt5mDLgFH5yZwekHqgCxX1h2TsmwkRnD93B21Lhrc7ZcZi4Y99W4+2lFPzUEFmow6iePzkpiRn4TVqI3/g4EyWM6jfllhd52bH6tDc9I7Wic63qRv6Um3kRfBJL2z+dG1Lj+769r3lksBP8PfWU7BP/6GtbrtSjjVM0/v8ZJjscC2bw+TH/gDad9/G37OMXYCGx9+CtfogvBzEzPjSD6kAraiKOFRkaqqsqmyuV/XLvfLCtuqXWyocLKvsX3RviSLganDEjhqWALJVrHOeKySAgFOO+M4TPYmgtY4Pluxvt1Sgh3FY0cOjtFY5nQ6UVUVSZLE2uzd0O895i+88AKLFy9m4cKFnHnmmW0S8PT0dGbPns2yZcv6LTFftGgRL7/8Mu+88w4JCQnheRdJSUmDYn3AQUeS+PGuh0hZcBYmexPDPnmf6llntVv/sb+ojZVIqbkD8lmC0BuHxuiIZAtXHzuMrdUuPt3VQJM3iApsKHeyrcbFGQWpHJ2bEPFevZ6qavZHLDH3BhUaPdFPyps8QT7cUdduWsGR2fGcPjaVpAjeiIglg+U8atLrKMyKpzArPlw47scqF7vqDiTpzf7QevTryhzEm/QUZoWS9BHJfUvSKx2+dom5qqqUH1IEUgoEyH3vTQqef4q4yvI226pmn8WuX98YKuY2CLnyx/Dtcy+Tt/y/TPzrg+h9PhJ3beekS+fy4x0PUD7vIiDUa35oIlRcXBwuRtzoCfZLUq6qKmV2HxsrnPxY1dyu2KBegvEZNo7JTWB0mjXq52ih79LWfY3J3gRAzSmntUvKDTqp238HD47RWHb88ceHl0vbunXr4V8gdEuvry4ee+wxzj//fF5++eXwBP2DHXPMMTz55JN9alxX/v73vwNw6qmntnn+hRde4PLLL++3zxV6z5eRydY77ufo268HYNJDf6ThmOn4MjL7/bNVOXbWQxaGpo5iVJIkJmfHMz4jjm9LHaze24gvqOIJKPxvWx0bKpycNzG9y3mr/a3BHcAXVCIynLva6SOaVU9kRWXtfjtfFDe2mY+cm2jmnPFpjDhMtd3BbjCeR016HZOy4pmUFY8v2JKkVzezq+7AEmzNfpnvSh18V+ogwaxnfHoco9OsjEqxEteDkSsej5svv1xP3oJzSDAfuPyqdwf4ZOUXTJs2jTijiWEfvkPB809hKy1p8/rqU05j128W45jYfuWZQUenY//Fl9Ew9TiOvv0GEvbsRO/1MuXu35O4Yyvbb7oThzc09eXghMjvP1BhrTLCS6Q1+4JsrmxmY4WT2g6WOUu3GZk6LIEpwxKI18iIJiEycj75IPx95VnntduebDV0+wbMwTEqCIfqdWK+e/dubrjhhk63p6amdpiwR8ogqFk3JFWedR5ZKz9m2CfvYXLYOeL+O1j/xD/6fRieZBKjKARt6ypGjXodJ+UnMyUnno93NrClKlRIsczu49m15UzPS2TWmFQsUZjrrKqhtby7sxRUV2RFpaY5eolfmd3Lu9vqqHIeuGiKN+k5Y2wqR+bEi14vBv951GzQMTk7nsnZoSR9Z20oST94nXSnT2Z9uZP15U4kIDvBxOg0K/nJVnKTzJ1OMfF43Dz80MPs2r0bd0M199xwNRC6lnniX//l5VdeR281cFt9GQn797V5bc2JM9l17WLsk4/qx59em5oLxvPV/71D4V/uDy9TNerlF0jYs5ONj/yNcquhTWJus4XWlW72BXH45A7fsye8AYXttS5+rGpmT4On3ZJrJr3EpKx4jh6WQF6yWXMFOoW+kwJ+sld+BEAwzkbtiTPb7ZPWg/nlrTEa62bMmEF9fT1paWnRbsqg0uvEPDk5mbq6uk63FxUVkZ0dQ1VBhQGz9Y77SPt+Leb6OjK/XMnwt16l7CcL+/UzpZbllgRBq7oTowlmA/OPyGRqbgLvb6ujzh1ABdbud7C1ysXZ49OYlGUb8IvD6mY/uUnmPlW0rh3AIk0H8wYUVuxpYF2pI1w9WQKOHZHIaWNSsRhjfy5gpAyl86jZoOOInHiOyInHG1TYWevix2oXu+vc4ar8KqECiJVOP18RWo4qxWpgeJKZ3CQLwxJMpNtM2Ex61q9fz66WZXteevUtbIqHWxZfzyNPPEPzP//N17VlTPC3nWNeN30GO39zE01HHTOQP7rmKBYLP971IPbCyUx66G50wQDp337FiZddwLqnX8KZVBgegdC6JFKFs/e9kq0jJ7ZUNbProP/vg41MtnB0bgKFmTZR/HGQS1/7FUZnaGm66lNPb1OEEECvo0fTm1pjNNY999xz0W7CoNTrxHzOnDk899xzXHfdde22bd26leeff77f5pcLsS2QnMKWJQ8z7cZQj0HhX+6n4ZjpuEeO6rfPVBorYmqZH2Ho6UmMjk61cu0Jw/m6pInVxU0EFBWnX+b1LTVsqrBy7sR0UrpRhCZSgopKnStAVkLvK7RX9uFCujdUVWVrtYuPdtTjPKiCcna8ibmF6QxPGtrD1jsyVM+jFoOOI3MSODInAb+sUNLopbjBQ3GDp80ICwjNa270BNlS5Qo/F2fUkW4bwwmX3cz2TesJuh3889Pv2f3Sify+sY5jHXUcfEur/pjp7Lz2JhqPmd7tNqqqiqyGlogLyCoBRe3W960O/nxJAgkJo17CpJcw6XUt3+uIM+mxGXVYjfp+XVquI6U/+RnNowqY+rtrMTfWYyst4fhfXsRL1/yWi2+/EQjN3x09bgIN7gBffrmaadOmYbXGHfa9610Bdta52VnnpqTR02EynmQxcGR2qHc8zSYqcA8VOZ8eqMZe2UE19mSLsUe/C8XFxWK5NKFTva7KXlFRwfTp01FVlblz5/Lcc8/x85//HFmWeeONN8jJyeG7774b8OXTOiOqsmvP5PtuJ++tVwFomjyFb/71er8t9RJr6+8KQ09vY7TRE+DD7fXsqDtQqMyokzh1TAon5CUN2MWz1ajjqGG9q8za6A60K7TWnxo9Ad7fVseug9aMNuokZhekMH3EwB2zWCPOo+25/DJ7GzyU2X2UO7xUOHo+8kMvy1j9HiyqjC7ehiHeRqgTVgonzJIEihqa8hEMfykEWxLsoKK2G2bdnyTAYtRhM+qJM+mJN+lJtOhJshjafNlM+ohPA7FUlnPs9VeSsGcnAE6djtcu+Tk/eXIpRUVFWLNH8dx/l7Ns2TLGFhRw+x23t0nOVVWlyRuktMlLSZOXvQ1e6t0dT6OJN+mZlGVjcnY8w5PMYkrLEKPz+zjttGMxNjsJxCew4rN17XrMx6Zbe7RsqFjHfGjqblX2Pi2XVlNTw5133smbb75JU1MTEFon/KKLLuLhhx8mM7P/i3p1l0jMtUfvdnHSz87D1jKfbtfVv2XXot/1y2epHieSVSznIGhXX2N0W42LD7bXtZlXmRVvYu7E9AErWtbd5WIOta3GRdMAVGOXFZVv9tv5Yk9jm97C8RlxzBmfLpYzOgxxHj08WVGpbvZTbvdR4/JT5wpQ5/JHZL5zLNJLkGgxkGgOJeqJFj2JFgNJ5tDc8CSLgTijrsfTbwwOO2OvuJhRxaHk3I/Emwvmc9SDf+KJV97nlWWvhve94jeLyC88mupmPxUOH/ubvDi7+P9IthgYlxFHYaatX9azF2JH5uoV4dGdZedeyA8P/LXNdr0OjslN7NHN3MbGRlJShs60ICGkXxNzn8/Hxx9/TH5+PkceeSQAtbW1KIpCRkaGJtfnE4m5NiVt3cwJl89HFwyi6nSs/ccyGo8+NuKfo7ib0MUlR/x9BSFSIhGj3qDCyt0NfHfIfOlpwxM5vaD/50snWw1MzOxZYRu3X2ZzZXM/teiAMruX/xXVUd18YOhxolnPnAnpPW7zUCXOo72gqqR/u4Zhzz+DvbaR6qRMahPTqE1MY0dSBpsyhtMUl0hSajomaxx+WcEfVOnsCkAitDRT+Esf+teokzC2DDk36iWMuq6/N+klDAddqx36ibICAUXBLx8Y+u4LKrgDMi5/678y7oCML9i36xWDTiLRHErYEy0GbEY9FqMOq1GH1aDHatRh0EnodRI6CXRS6F/V6yPxwT+SV1GOy2ylwZbEuxOm8JVkxRifgjE+GWtKJorU9TWfToIRSRbGZcQxLj2ODJtRFHETADjyrpsZ/v5bAKx78p/Unjy7zfa0OCPjMg4/VeJgtbW1ZGRkRKyN0bJo0aLwTYann3462s3RvH5dx9xkMrFgwQKeeOKJcGI+GIJMGHj2SVPY9esbGf/0Y0iKwpS7bmbNsvcJJiRG9HNUVxOIC0pBwyIRoxaDjjkT0pmSE8+72+qodPpRgXVlDrbXuDhnQhqFmf1XHK7JE8QTkLEau3/zs9IZ2SWNDuUNKqzY3b6423EjEjmtIFUUbuoBcR7tmdR13zDu738ldeN6AIYDk0p3sMUcx906PSuNEnLKCPRxiZQCCxcuZN68eajqgWHprTGrqiqSJKGX0FzSGFRUnL4gDm8QuzeI3Su3/Bv6cniDeIKdryceVFQaPEEaejNq5twb2z11cF9kR59q0ksMT7KQl2whL9nM8CSLOA8I7eh8PrK++BSAQHwCdcef1G6fnlRjbzVYEvMvvvgivI65EDm9SswlSWLs2LFdVmUXhO7ac8W1ZHy9itSN64mrKGPSI3ez+YHHo90sQYhZuUkWrjkul29L7azc3RguDvfaDzWMS49jzoS0fisOV+nwMzqte8tqBWSFug7WA46UbTUu3t9e12bYanaCiXkT08kVxd2EfpK6fi1jn32CtPVr2zy/1Wzl4fQRvKM3cNPl8/nn4sU8uvRv/P31DwFYtmwZKSnJnHzyKejb5d7aSsYPZtBJpFiNXZ5TfEEFx0HJu8MrY2957PCGljbzdpG894ZeChVsS40zkhVvCn0lmMiwmUQdCeGw0r9ehdEVGs1VPfssVGPbeeR6HWL6kxBxvY6oO++8k5tv/v/s3Xd0HOXVwOHfzPai3puL3LvBVAMxpsQ2oWNKKKGGEjokhJAPCCUEAgkQCIEAoSWhh25Mx5iOMdjYlovcLat37a62zvfHymsLS7IsSzuzq/ucoyPtzGrnanT1au/MW67h5JNPZsyYMf0ZkxhsTCa+v/1eDjn1KCxtrRS99So1B82kcs6x/XYINauk315LiIHQ3zlqUhWmD01nfK6Lt1bWs7pjcrjVdV7Wf+5j5ogMDhiAyeHqvAFK0m1YTLu+A1XdFhiQCaua20PMW1nXaUI5i6oM2M88WEg72gNNI+vrzxn5z7+RtfjrTrtah49k+XmXcMVX37F67VouPXkOV111GQDXXXUZHtXJ08+/zKiRI9lnn330iH7A2cwqOeZoUdyd9lCE1o67675gGF8wEvs6HIGIphHRtNjXqhLtnr+ufA3LlnzHCZtXclxzHemeJnKbanhlznEM++3/xfGnFMmk8N1dzMbu2L3Z2LcZPXr0HsVlFAsXLiQSiRhy+HIi63Nh/uWXX5KVlcXEiRM59NBDGTZsGA5H57skiqJw//3373GQIvm1Fxaz/IbbmHrDVQBM/OPvaZ4wGe+QYf3y+lpzNUpGYb+8lhADYaByNN1h4fSpeZTVeJm3Knr3OBjReHdNA0sr2/p9abBwBGraAru8Ix3RtJ2WmtpTwXCEzzc2s3B9U6fJ3UZlxX8JuWQk7WgXNI3sLxcy8p8PkPn9ok67PEOGseaiK9k66xgwmfjt4bNZ9u1XXHHG8Z2ed/Uvz8KVmdPrpb2Sld2sYnfv3pKLr7/+Ogufe46I38u965Zw3MRJTNm4AoDzn/8XD1utjLr6uoEIVyQxk89Lbkc39kBaOnX7T9/pOdl96MYOsGnTJkpLS/coPiPIzMzUO4Sk1OfC/MEHH4x9/cEHH3T5HCnMxe7YOuc4sj9bQPFbr2DxtLHX9ZfzxZMvEbHadv3Nu6CF4rtGshC7ayBzVFEUxue5KM1ydBpvXdUW4LGvtzKl0M3hIzJJtfdPt7zK1gAFqT0vLVTnCRLsarHgPtA0jbIaL++srqepffs4VbfVxJwxWUzIG7hx9YOJtKM70DRyPv2YkY8+QMYP33Xa1TaslPJfXk7lT49G22GSH4fDyTknHb3TS6U7LMw+/FA8gf7typ3sFi78hOeeew4AX80mbrrpJsxHHMGiy69knyXfYdM0LnjmEZ7OymboL87TOVqRSHI/fh9ze3Q5zaojjtqpG7tZVfrcjb29vX2P4xPJq8/9DyKRyC4/wuHBuTyI6LvlN9xG29DhAKSVLWPsvX/ql9dVLHte3AsxkOKRo3azys/GZnPBfoXkd9yZ0oDvt7bxt8828+HaBvz9MM4zGNZ2OXa8sqV/Jn2rbg3w1OJKnl9aHSvKVSU6udtl04uZmO+WoryfSDsKSjBA0esvccjJs9n3ivM6FeWtpaP47k9/45OX3mXrUcd3KsohOh61u4tfhalybnfXPvvsw6iRIwn7PdxwzeVcddVVOFwuhr87n+/HTgTApUU4/7H7cZev0jlakUgK578W+7qii2GVWU5Ln5fR+3HvYiF2tEfrmAMsW7aMefPmsWHDBgCGDx/OnDlzmDhxYn/E129kubTEkbJqBdN/cQKmQPTuzLd/eZjqw2bt0Wtq4SCKSbqxCuOKd46GIxpfbW5mwbqmTpMupVhNzByZwdSClD0ai+2yqkwu6HrN60ZfkJU13i739VaDN8hH6xr5obKt00JPpZkOZo/JIm83u8SKXRvM7ai5tYUhL/+XYf99Enttdad9rSPHsObCy6k6fA70MN5yYr6LFFvXhbmmaXy/ta3fJ0BLdj6fl/WLP+P80+cCEAgEsFqt1NS1oh5zFKNWLY8+L7+Qz5/6H/7cPD3DFQnA0tTI4UfuhxoK4csr4KN5n+70dz0hz9XnHmbbcjTRvfPOO/h8PhwOB7Nm7dl79MFgQNcxh+ha5hdddBHPPPMMmqbFBv9HIhEUReGMM87gscceM0zySWGeWIa8+B8m3hGdtCWYksqnz72Fr7C4z68Xrt2AKWdYP0UnBjslGMTU7sPU3o6p3Yfqb8fk37F72vaCVlMUNJOJiN1O2O4g5HASsduJWKywwxV3vXLUGwizYH0jX29u6TQRW6bDzE9KM5ic7+5zgT4u10l6F+O6y2o8NPVlaSSgyRfk0w3NfFvROd4Mh5lZo7MYm+OUO+QDZDC2o/bKCoY9+yQl/3suNkPzNg1TprH+7AupnnFEjwU5RO+Wj8t19ficqlY/6xukm+vusJtVphZu7xWzYsUKxo8fT0TT+KG8iqnn/pz0FUsBaB4zni8ff56wy61nyMLgSl76L5P++HsA1p59Iauu+l2n/Xazyl5FXV907o1tOZroJkyYEFsubfny5XqHY3gDuo45wG9/+1uefvppfvWrX3H55ZczYsQIFEWhvLycv/3tb/zjH/8gMzOT++67r6+HEIPYprmnk/XN5xS8Nw9LawtTr7+cLx9/fqdxPkL0J5OnDUdlRacPe00V1qZGrI2NWJoasDY3YfZ69vhYmqpGC3WXm2BaOgGHjWB2PsHUdALp6QRT0wmmpRFMTcefnYM/Kwd/dg5hZ89v7neX02pizphs9itO473yeso67mQ3+EK8uryWBesamT40jSkFKbu91m9la2CnwtwTCPepKK9uDfDpxiaWVbV1KsgdFpWDh6Wzf0lqr2aCF2KXIhGyv/6MIc8/Q94nH6BEtt/F1hSF6kOPZN0vLqRp6rRev2RJ2q67que6rWxp9vfb3AuDQUGqtcsLcaqiUFCQxaL7H2P62Sfi3LqFtFUr2Pu6S1l032NolsHZ80PsWuHb27uxV87euRt7tktyRwycPt8xz87O5mc/+xlPPfVUl/vPOuss3n77bcOsdS53zBOPubWFg04/BteWTQCsP+N8yn7dt6VPNG8zijOtP8MTCczc2kLK6jJS1pXjXrsa9/pyUtauxlZvjPaqJyG7I1aoB7Jy8Gdnx4r2Tp+zsvt0IWtjYzsfrWvY6c7dtjtT+xWnkdWLNyY+n5dFixZx6SlH4bJub3PX1Hl5Zf4HvZqBOhTRWFnj4duKVtY1+Drts5oUDhySxvSh6dgtUpDHQ7K3o5bmJoreeJmhL/4b16YNnfaFrVYqjjmJ9WddgGfo7s2onOk0MyandxfUKpr9bGqSu+a9YTEp7F2U0mmsb319PVlZWcD24QGmNWs48JyTsLY0A7DppNNZ9n9/1CVmYWz2qq0cNucgANqGj+CTl9/r1LMNYK+iFOy7eZF6RzvmaCJ78skn8Xg8uFwuzjnnHL3DMbwBv2MeDAY54IADut0/ffp03njjjb6+vBCEUlL5/s4HOPCcuaihIMP/8zhNE6dQOfsYvUMTCUQJBklZs5L0ZUtIX/Y96cu+x71+7W6/jqaqBNIyCKRnEExLJ+xwELZv+7ATsVpBUSF2rXP7xTU1FELt6PZuam/H5PPGPps9bVhbmjD1cqZWc7sP85ZNsQtWPQmkpe9UrMeK+ti2HALpGbGuuEMz7JwzrZCNje0sWN/I2vpoQdweivDlpha+3NRCcZqNifluJuZ1PWbW5/Ny55/uZE15Ob6Gam6+4oLYa/zr2Zd47rnnGTVyJNf/7vqdivNwRGNTUztlNR5+qGrDG+w85tZhUdm/JJX9StI6FfxC9IUSCpH9xScUv/4SuQs+wBTsPPN8e04em076OZvmnk4gK2f3X1+Bkt1YjjA/xcrWFr9cnO+FgpSeV35QFIWiNBtrh4/g23v/yX4Xn4UpGGDIy/+lZfQ4Np1yZhyjFYmg4J3tdcvW2cfuVJSn2k17VJQnEynGB0afC/NZs2bxzjvvcMkll3S5f/78+fz0pz/tc2BCADRPmMyK39zIxD/dBMDkW66jrXQkraPH7dbrRDyNmJL4To/YQSRCSvkqsr/6jKyvPydz8de96nruz8iibfgIfIXF+AqKYh/t+QX4M7MJuVN2OY50T2hbVmG3pmNpacLS3IS1pTn6uakBa30dtvo6bPW12OpqsdXXYm1u2uVrWpubsDY3kbJuTY/Pi5hMBDKyOt11H52Vw4yMTNamFfC+s5BvIikEO8bOb2n2s6XZzzur6ilIsTI8y0FppoPiNDt2s8qiRYtYU14OwJPP/w93xMu1V13B3Q89xnPPR7sJrikvZ9GiRRx08CHUeYJsbm5nY2M7q+u8+II7T4CV6TCz/5A09i5KwSpd1nWRNO2oppGyuoyit16haN6rXfaUqdvvIDaeciY1Pzl8j7o9ZzktOHfjApJJVchPiXZpF90zd5ynH6uuru50NzLHZWFri5/Gvffjh5v+xNQbrwVg/N230FY6koZ9ur/BJAafwrdfj329tYtu7DmuPR9O+eMcFWJHfe7KvmrVKk455RRGjBjBpZdeysiRIwFYs2YNf//731m/fj3PP/88OTmdrzDrtSC9dGVPYJrG5D9cR/HrLwHgKR7C5/9+jWBaeq9fYjBOWjSYmNtayf58AXkfv0/2l59ia6zv9rkRs4WWMeNonjCZ1hFjaCsdSVvpKAKZ+v6j3N0cVQN+rA31nYp1W922r+uw1ke32etqen03vidNzlTe2Hc2b+91OGsKR3b7vGx/KyX+Fqz1lTRWbEJr9xJqbyNkd+BVzYRMFoKuNLJGT8SUnU9VSKU90vVdL7MC47MdTCtwMTTDhqqa0BQleoFEJniLu4RuRzWNtBVLyX//bfI/mI9r88adnuLPzGLrnOPZdNLP8QwfsceHVBWYXODGYdm99xzBcITvtrYSlgnau1WcZqMkfeeeCF1NrFXvCbK6Ljp3xti//pHSZx4DwJ+eyef/eW2PJpYVycO9dg0/mRu9odg0cQqfP/Nqp/0mFaYVpe7RaiWQPJO/id0z4LOyqzvcOfrxxBvbXrKrCTn0WttcCvPEpvr9HHDeKbHZVWumz2DR3x6HXv4uB/MyP8nKVltD7oL3yfvoXbK++WKnLqjb+LOyqd/nQJom70XTxKm0jBlPxGa8NYMHLEc1DZPXs71471TI13XaZm2oQ+1FG70udyjvTJ3JwvEH9lik94XT72X6yq84dPlnTF/5NSnt3fd20BQFraNI1xQ1Ohl+7Gsl+vW2/1U7PpfoPlQlWuizfV/0+5VO36+ZTEQsViLWjg+LbYevt28P2x2EnC5CLhdhp5uQ00nI6SLschNyuQg5Oz67Uwg7nAl3cSHR2lHV5yPr26/I/nwB+R+9i6Nq607PCVus1PzkMCqOOYna6TP6dVKw/BQrwzP7tmbxxkYfW1u6btMGO5MKexelYu6iQOpuKaqlla14AhEIh9n38nPJ+WIhAC2jx/HFky9F/x7FoDbq739h1GMPArDiNzex4fRzO+3fk7/nHSXLcmli9wz4GPObbrpJlqQRcROx2Vh8z0McdPqx2JoayP18AaP/8VdWX/abXn2/1lqHkl4wwFGKgWZuayXvw3comvcqWV9/jtLFdcWgy03DtP2p3/8g6vY/iLbSUQlRAA1YjioKYZcbr8uNd+jwnp8biWBtaowV65ampu1d65ubY1+ntzRz5ppPOe/bN2kNaSweNolvR0ylPH84a/OG4XH0fjminOY6JmxeyaSNK5i0cQUTNq/EGg727kfTNBSdLvbuqbDFGp11Py2DQFo6wfSOzx0f2+Yz2HGiP70vKBm+HQ2HSVm7muyvPiP78wVkLv4aU2Dn4lZTVer3OZCqw2dT+dOfEUzP6PdQTCoUpfb991WQaqO6LSB3zbuQ77Z1WZQDVFRUMHz4zu1cSZqdlbVeMJn47s4HOOjM43Ft3kDq6jIm3/wbvrvrwYT4PyEGiKZROD/ajV1TVSqP/NlOT8l1908x3V2OJpq99tortlzad999p3c4SaPPhfkf/vCHfgxDiF1rLyjiuz8/yH6XnIUaDjPy8YdoHjeJ6sNn7/J7taCM10tUSihEzucLKHzrFfIWvI/Jv/Pv0pdXQPWhR1J96JE0TNs/IZfCMUSOqiqBzCwCmVm0jhrbu+/RNFS/n329Hg7weTF5PLR5Wmj2+lm6dDmr167HZLZhj4Swh0NYQgHGZqWzV7qTXG8TDr8PJRhEtQVRSwtpKs5GDQY65s7ToktVaR2f0SCiRS/IaJGOfaBoEYhEOrZrHd/X8bWmRfdr2vb9277e9j0dP8e2Y20/bvS1lVAINeDvssjrK1MwgKmuFntdba+/J5iSusNEfjk/+nr77PyB9Mxe9ybaHYbI0R2YvB7Sly0h4/tFZCz5lvSl32Fpa+3yuRGzmbr9DqLq8NnUHHrkgA9dyU+xYd2DSaKsJpVct5VKuWveiUmF/NTuCySfz9fl9gynhVSbiRZ/mFBqGovue5TpvzgBi6eNgvfm0TLmIdaef+lAhS0MLm3Z97EJVev3PRB/Tm6n/Sk2U79NNtpdjiaaQCAQ+xD9p8+FuRB6aNj3QFZdeT3j/hpd6mTK/13Dl0UltIyd0OP3KWbpNpRo7FVbKXnleYpffQFHTdVO+z3FQ6icfQxVM2fRMm5iwt/tSNgcVRQidjsBux3YXuwseP11nlv4PgCta78jxWEj4MzGmp4LtVs47bTTOPbYnSfXMTRNQwkFUQOB6Ecw0Olrk8+H2duG2ePB7PVg8rRh9nkxezyYdthubm3B2rytJ0Jjr+cAsLS2YGlt2eWqApqq4s/M2j7rfhdL6W37OuRO6fXfjl45qgSDOKq24i5fRerqlaSuLiNlzUqcWzZ22WtmG19+IbXTf0Lt9BnU7zedUEpqXOK1mJQ9ulu+TWGqjRq5a95Jntva4+SPdnv3M+CXpNtZXh0dGuMpHcmSP97LtKsvRNE0Rj/0Vxon703Dvgf2e8zC+DpN+jbnuJ3299fdcug5RxPJuHHjyMnJITs7W+9Qkkqfx5gnGhljnkQ0jSm/v5qit6OzO/ty8/n8mVfw5+Z3/y2RMIpq7N+7AMJhcj/7mJKX/0vupx933CXdzp+eSeWsn7F1zvE0Td4r4YvxHSVTji5c+An/+MfDAARb6/n1eadx1VVXcc99f+PvL86HjtndL7nkYg455Cc6RmoMans71uZGLE2NsYLd0tyEtbF++2z8O8wJ0JtVBnorbLPtdAc+kJUd7VafmkYwNY1QSirBlDQCbjehtAwi/fjGUgkGsTY2YGuoi81zYKurxbl1C84tG3FUbMZRWdGruQ/as3NonDKNxr33o/aAQ6ITuOnQRgzLtFOQ0j/DDjY0+uSueQeTClMLe16VIRQK9Th+c2WNh0ZfKPZ45CN/Y/TD9wLR+UgWPjePQPbuL4snEpcSDHLYrAOxNdYTtlr54P1vOl3Es5gU9i5K6XFpvt2xqxwVyWnAx5gLoRtF4Yeb78JZsZmMpYtx1FSxz5UX8OW/Xuh2ApdI/ebEnU14EDB52ih+/SWGPfvkTrMla6pK9U8OZ8sJp1J74E8Sspt6byRTju6zzz6MGjmSNeXlXHneGVx11eUA/PqqK/CoTp58/hVGjRzJPvvso3OkxhCx22m3F9Ce17vx2yavZ+fl8+q2zcJf27G8XvSxGgr1/Fp+f7QI3rql1/GGrVZCrhQitugkeGGbjYg1+hG22aKz5kciKBENJRLuGDIQQQlHMPm8Hb0JPJi9bX0eHhC222kdMZqWsRNomLoPjVP3wVdUovvFOodFJb8f767JXfPtdnW3HGD16tU9zng9JN1OU3sb225Jlf/yMjKWLCLni4XY6uvY63dX8PU/nkGTwmnQyPl8QWwll5oZR+zUsybXbe23ohx2naNicJOWRySkiM3Gt/c+wvSzTsC5dQtpK5cz5fdXs/juhwZkbKUYGI6tWxj63FOUvPL8TmNDffmFbD7+VDafcEqPvSGE8TgcTq7/3fUs//YrLj/j+E77rrvkXByZeeyzzz44ZCbkPgk7XXidLrwlQ3t+YiSCpaV55yX1Os3OXxct6psaen18UyCAKdD9koT9KehOwVs8BG/xEDzDSmkZPY7WUWPxlAwzZFs/JN3erxPjWk0q+W4bFS3GGt8fbyY1epFiTzmtJnJcFmraOiaYVFWW3H4vB/38aBw1VWQt+pJRj9zH6kt/vcfHEomh6I2XY19vOeakTvsUhX690CbErkhXdgOSruy95167mgPPmRsr6jaecibLr791p7smEW8TqjNdhwhFV9zlqxjxxD8oeOfNnbqo1h5wMBt/fg41Bx1qyDfeAyUZc3RygbvLCXPW1Hmp8/Ru5nURH9Eu5fWxwt3S3BQd097S0jG2vRlzYx0Wrw9LSzNmT1vH+Ho/aiCAyd/e43jvbcJWKyGnm7DL1bG0nJtAegaBrOzY+PdAZja+/AK8xUMJpqXrfhe8t9LsZsbnufr9dUMRjcUVLYP6rnlRmo0hXaxb/mO1tbXk5PTcFT0QjvD9j9aJz/juG/b/5c9j/4++eeAJag8+dE9CFgnA0tzEYUfujykYwJ+VzYfzv+jUWyLLZWF0dv9eQO5NjorkI13ZxaDQNmI03/35Qfa54nzUUIihL/wbf1YO5Rde0el5ijJ4CjwjS1v2PSMef4j8j9/rtD1stbL1qONZf8Z5tI0co1N0+kq2HM1wmLudxbYo1Ua9N8jguCycGDSLBX9ufs9zdfhaURwp3ezUUEIhTP520LTouvCqGl1TXlXQVFN0ffgkvdimKDA0Y2AmdTKrCoWpNjY3Dc675tt+/l49txdd0K0mlaJUG5t2OJ+Ne+3Lqst/w7j77gRgyo3X8Omzb9KeX9i3oEVCKHjnDUzB6HCaijnH7TSEoT8mcfyxZBlfftNNN9HU1ER6ejq33nqr3uEkjb6v5SGEQdQd+BOW3nxX7PHof9xLycv/7fScSFt8ul2KLmgaWV9/zn4XnclBZ53QqSgPpGew+uKr+ejtz/jh5rsGbVEOyZejxWndv6FxWk1kOpNzroBk1mOOKgqaxULInUIoJZWwy03Y4SRitxOx2qJveJO0KIfo+Of+Wk6pKwUpNiymxOg50N8KU63drlv+Y5WVlb16XkGqDZu582uuP+uXVP/kcACsTY1Mvf4KlF3M0SASW/HrL8W+rvhRN/Y0e/cXl/dEb3PU6F5++WX+/e9/8/LLL+/6yaLXpDAXSWHr0SdSdvUNsccT77iR/Pfm6RiRQNPI/uITDjz7JPa/6Ayyv/4stsuXm8+KX9/IR/M+pfyiKwhkynIbySTDYcZt6/muQHGqLVF6KAvRI7OqUNKLbtZ7wqQqFPVwsStZWU0K+f00w/2OVEXZuWu8qrL01nvwFhQBkLnkW0Y8/vd+P7YwBte6ctKXLwWgecx4WkeP67S/MFXGlov4S47+FEIA63/xS2z1tZQ+/ShKJMLUG65kscVCzaFHomZId7R4ylz0JaMf+iuZ333TabunZChrz7mYrUefQMQ6+N5k9iSZcrQ3RYrTaiLLaZGx5gkkmXK0Pw1Jt/f6ju6eyHNbqWoN0B4cPIPNi9PsmHbj3JaWlvb6udkuK9VtAVrat89zEkxL5/s//Y0Dzj8FNRxm1D//Rv3+B9E4VVaQSDbFO0z69uO75S6riXTHwPTq2p0cNbJXX31Vln4bAHLHXCSVlVdez5Zj5wKghkLsdd1lZH+2AM3TqHNkg0P60u/Y7+IzOeCXP+9UlLeOGM13d9zPJ/97ny0nniZFeReSJUcznb3v/lecJnfNE0my5Gh/SrGZyEuJz501VVEYkjawd+aNxGFRyXXvXnFUXV29W88fluHYqQ1qmrJ3bJ4aJRJhyg1XYW5t2a3XFQYXDlP01isARMxmts45ttPukgHsnbK7OWpUo0aNYty4cYwaNUrvUJKKFOYiuagqS2+6k4o5xwFgCgaYds2FZH35qc6BJbfUsmXsc8X5TD/7RLK/2t5lvW1YKd/96W8sfOFtKuccK2vD9kAL+PQOYY8pSvQOV285LNGli0RiSIYc7U+KAsMzHXE9ZpbLQootecfq72hoH5ae83g8u/V8l9VEXhfLYZWffykNe+0LgLOygom3/x6ZrTJ5ZH/9GfbaaIFcO31Gp+F0LquJjAGcA2V3c1QMLlKYi+RjMrH01nuoPPKo6MNAgP1uuoGchR/pHFjycZevYu9rL+Hg048hd+GHse3eohKW3HoPC198h8rZx4AqTc2uKKbEL1CznZbdniynOM1OHHoBi36QDDnanwpSBnbCt+4M1OzvRpJmN/epOLJad7/3wpB0O9YfT6xnMvH9H+8l6I6uQlD47psUvfG/3X5tYUw7/i5/3I19IO+WQ99yVAwe8m5ZJCXNbOb7P95H1cyfAtHifNo1F5L/7ps6R5YcXBvXMfV3V3LIKXPI/3B+bLsvr4AfbryDBa98QMUxJ8kd8t2gZBToHcIeURUo7sMEWDazGreuwGLPJHqO9ie7Wd2t3iH9KcVmJjuJe5rsydJzfRm/a1IVhmXs3POhvaCIH268I/Z4wp034dy0oU9xCeMwt7WS/9E7AARS06j5yWGxfam2gb1bDskzxvzrr7/m008/5euvv9Y7lKQihblIWprFwnd3PcjWnx4NdIw5/92VFL/6gs6RJS5HxWYm3/wbfnLikRTOfx2lo2tfe3YOy6+7mQWvfcTmE3+OZkneN40DJVK3Se8Q9kiu24rd3Ld/KUWpNkzy38jwEj1H+9PwzN2blKy/DU23J+3fzJ4sPbdy5co+fV+Wy0K6Y+cLyVU/PZrNx50CgNnnZeoNV6IEZcLKRJb/3jxM7e0AVM4+ptOcNwO9ugL0PUeN5txzz+XYY4/l3HPP1TuUpJKkzboQUZrFwvd33MfG2dFu7UokwuRbfsuIRx+Q8WK7wV5dyYQ//h8zjj+M4tdfQolEZwX2p2dSdvUNfPz6Ajb+/BwiNpnUbTAyq0qP65bvisWkUpQquSMSQ67bMmAzNveW1axSlJp8XdrjsfRcd0ozHV1e7Fhx3U20DR0OQPrypYz410Nxjkz0px1nY9+yQzf2TKeZVLv08hP6kgwUyc9k4off30YkK5/h//kXAGMe+iuuzRv54cY70CzSjbY71rpaRvzrIYa8/F9MgUBsezAllXW/+CUbfn4OYZdbxwiTh+JM0zuEPitMtWHZw9t3Bak2qtsC+ENywcyoEjlH+4vN3HW3Zz0UpFqp9QTwJdHyaXu69FxWVlafv9dmVhmSbmd9Q3un7WGniyW3/5UDz5mLGg4z8tEHqD14Js0TJvf5WEIfrg1rYyvGtA0fQfOEKUB0KNZO69oPkD3JUSO54IILaG1tJSUlRe9QkooU5mJQUMxWyq79P/xZOYz9211A9Kqpo7KCxff8g2Baur4BGoy1oZ7Spx5h6AvPxLp8AQRdbjacfi7rz7qAUEqqjhEmn0SdWMtmVihI3fOLW6oSvVNWXiczfxtVouZof1EUGJnl1LUL+45URWF4poMV1ckxy3N/LD1n28NeW/kpNhq8IZrbQ522N0+cytrzL2XUP/+GGg4z5cZr+PS/bxKxJ1+vhWRW8srzsa83H38q29bKK0i14bDEZyLHPc1Ro7j66qv1DiEpSVd2MShEWutAUVh37sUs/vPfCXc0jFmLvuSgM44ldeVynSM0BktzE6MfvJtDj/4JpU8/GivKQ3YHa8+5mI/f/IQ1v7pGivIBEGmt0zuEPhma4UDtp8XIc1zWQbMUVCJK1BztL4UpNsN1dU2zm3d7rW8j6q+l57Zu3brHrzEiy9HlXfvyCy6jedwkANzr1zLmgT/v8bFE/CjBQGw29ojZQsXRJwLRi8vxHErVHzkqkpcU5mLQqTryKL7657P4M6LdiZwVmznw7BMp3uFK6mBjaW5i5CP3c+jRP2Hk4w9h9nkBCNtsrD/jfD5+cwGrrvwtwfQMnSMVRpJqN5HVzzPYDhsES0GJxOO2mShON+adriHpdiw/Xu4rwRSm2HRZeq4rNrPKsMyd2yHNYuH72/8au7A//L9PkPXVZ/EOT/RR3sfvY2usB6DqsJ8SyIy+Bxye6TBMLxghpDAXg4Ka3nmZn6bJe/HZf16jaWJ0fJEpEGDyrdcz+aZfY25r1SNEXdhqaxhz35+YedTBjH74PiwdP3vEbGHDqWfx8esLKPv1/xHIytE50uT34xw1OkVhQMbaum3JcQcwGSVajvYXs6owKtvZbz1D+pvFpDLcIOPe+8JhUfvtosfw4cP75XVyXNYul6TzlI5k1RW/jT2efPNvMLe29MsxxcAqeXWHbuwnnAZAtstCRpwncuyvHBXJSQpzMShovuadtrUXFPHl48+z8ZQzY9uK33iZg0+ZQ+Y3X8QzvLhzVGxmwh03cujRhzDiqX9i9kbHKEbMZjadeBofv/4RK66/FX9uns6RDh5d5aiR5af0fUmjXdnTCaDEwEi0HO0vI7IcfV4KMF6yXBayEnBtc0WJzobeXxc96ur6b7hFaaYDu2Xn3/uG086mbr/pADiqKxl/1x/67ZhiYDi2biH7i4UAeItKqN9vOlaToksPrf7MUT0ddthhTJgwgcMOO2zXTxa9Zuz/NL3w97//nWHDhmG329l///1loXvRJc3v7XJ7xGpj+e9u4/s/3kvI6QLAWVnBAReezrg/34LJ0xbPMAece+1qJv/fNcw4biZDX/x3bKb1sNXKxpPPZMGrH7Lsxj/RXlCkc6SDT3c5akRWk0JJ2sC9obGYVIZKl3bDSaQc7S9FaTYy+3m4xkAZnmHHmmBd2vt73H5ra//1eDOpCqOznex0jVBVWfqHPxN0R2ejLn7rFfLff7vfjiv6X/GrL6B0LJG7+fhTQFUZkeXY49VE+qI/c1RP1dXVVFZWUl1drXcoSSWhC/Pnn3+ea665hptvvpnFixczZcoUZs2aRU1Njd6hCYNR1J7/8W896ngWvvA29dP2j20b/uyTzDjhCArffi2x1zwPh8n9+D32u+hMfjJ3FsVvvYIaDgMQcrpYe/aFfPzmQpbfcBu+ohKdgx28dpWjRhKPMXm5biupMhGcoSRSjvaHDIeZkjRjjivvisUULTYShcvaf13YtzGb+zdHXVZTl5PStRcUseK6m2OPJ/7x91jravv12KKfhMMUv/YiABGTiS3HnkxhqpX0OHdh36a/c1QveXl5FBQUkJcnPSv7k6JpiVtx7L///uy77748+OCDAEQiEUpKSrj88su5/vrrOz03FAqxYMECpkyZgslk7Dd732xuIRRJ2F9LYotEGPbsk4x54M+Y/P7Y5oa99qXs2v9LqHVLzS3NlLz2IkOffxpnxeZO+wJp6Wz4+blsPO0XslSc2C1ZTgujc5xxOZYvGGZpZRvSHIp4c1lVJuS5E3JSqA2NPipbAnqH0SOTCpPy3XFbompPrW/wUdX6o3Oqaez961+R/+F8AKp/chjf3vdYbAkuYQy5C95nn6t+CUD1jCNY8/C/GJ/rQpHfk4ijcDjMkiVLmDFjRo8XZxL2jnkgEODbb7/liCOOiG1TVZUjjjiCL77ofnxwa2srLS0tsQ//DsWXSF7h2g29e6KqsuGM81j4wnyqf7J93Ezmd99w0JnHMe3K80lbvnRgguwP4TDZny9g6vVXcPiR+zHur3/sVJR7Soay4jc38dG8Tym/6Aopyg2k1zmqI7OqdDlb8UBxWEwUJ9Ady2SXCDnaH6wmhbE5roQsyiE6R4PRlx0cnukYkKJ8xYoV/f6aEF0tIt3xozfTisKy39+Ov2N277xPPozdmRXGMeSFf8e+3nry6YzKdupalA9UjorkkLD9Kerq6giHwzt1ocjLy2PlypXdft/EiRPxerePkzv33HO5/PLLKSgoYO3atbHX0DQt1iV+1KhRbNmyBZ/Ph91up6SkhDVr1gCQm5uLqqpUVVUBMGLECKqqqvB4PNhsNoYNG8aqVasAyM7Oxmq1xtYwHD58OLW1tbS1tWGxWBg5ciRlZWUEfUE0eyqKxUakJdo1Sk3PR/O1ovk9KKoJNauEcN1G0DQUewqK1UGkJRqvmpaH5vegtbeBomDKHkqkfhNaJIJic6HYU4g0R+NVU3PQgn40X3RWUVPOMCL1W9AiIRSbE8WRRqSpMvrclGy0cBDNG50ASM0egtZYiRYOolgdKK4MIo3Rn011Z6FpYTRPU/RxVglaczVaKIBisaGkZBNpqIjuc0WX4Ip4GqOPM4vQWuvQgn4UsxUlLY9IfbS4VFzpKIqJSFt0yQs1oxDN04gW8KGYLCgZBUTqNkWf60xDMVmia+8GvB0/ZzOa34uimlGzimNvNBVH5/Ptyc/nm9vuJGfhB0z8x4O4KqKx5n3yIXmffEjNgQez/qijqNlnP9SMArSAD629dYfzvRktEo6eb0cKkaauz7eaPRStsQItHEKxOlFcaUQauznfWUPQmivRQkEUqx3FlUWksQI0jdTaRormv0XR26/jqN95YpGaAw9h/c864k3JAiVMpONnj57verRg+87n25mOYjIRae3ufBcSqdu4/RyabURae5GzjhQUS9c5qygqavYQInWb0LQIit2NYnMRaa7uOIe5aEEfmq8X5zslBy304/O9teucTclCC4fRvF3lrB0lJWt7zrozQdN2yNlitNba7TmbmkukYcv2/FYUIm0NHeewCM1TjxZoRzFbUNIKiNRvgoCXiKdxe85CNLc8zWgBL4rJjJJR1Pl8x7mNcFhNWEsmsnr1akKhECkpKWRnZ7N+/XoACgsL8fv91NdH82Xs2LGsW7eOQCCAy+UiLy+PdevWAVBQUEAoFKK2Nhr/6NGj2bRpE+3t7TgcDoqKiigvL0fTNGy2NNqDkYFtI4jOOt7bNkJNy0dr33a+VdSsITuc7y5yVs82orucbanZoU3O6T5nO9oIAl4ijVuTuo2w5Q3D5a2ifHUIt9tNbm5ujzm7ceNG/H4/TqfTMO8jVpaVRc+h6iJishrufUSqWaN2YwW1/dRGbDvfEL35smLFCkaOHElFRUXsfA8ZMoTVq1cDkJOTg9lsprIyGlNpaSnV1dV4PB6sViulpaWx95FZWVnYbDa2bt2KpmnY3Xl4WxpjbUQgq5glV17Nfjf/HwDj7r6VmhFD8OXnD8o2wmjvI+xLPif38wUAeAuLaZlQinfVSkpKSmhra6OxMfqzjh8/PvZ/LTU1lczMTDZs2ABAUVER7e3tXf5f60sbsS1H9a41ADIzM3E4HFR0vMcdNmwYDQ0NtLS0YDKZGDNmDGVlZWiaRnp6OikpKWzeHP29DhkyhJaWFpqamlAUhXHjxrFq1SrC4TCpqalkZGSwcWP091pcXIzX66WhoWGn8x2P9xE7thHbxsQPRBux7XzX1dXR2tqK2Wxm9OjRrFixgt52UE/Yruxbt26lqKiIzz//nAMPPDC2/brrrmPBggV89dVXnZ6/rSt7aWkpqrq9o4DNZsNmM9YdGenK3v8ibQ3Rfzp9oAb8FL/2EiP+9RCOqq2d9nkLith80s/ZOvvY+I7PDodJK/uB/A/fJe+jd3BvWLfTUwLpGVTMOY5NJ5+JZ/iI+MUm+mRPcjQeslwWRmfHpwv7j/mCYX6oaiMc0eXwooPRc3RPmVWFcblO3LaEvWfRSUt7iLIaj6GGgqTaTIzLcw3Y0nNVVVXk5+cPyGsDBEIRllW34Q91PqmTb/4Nxa+/BEDdvgfy9cP/BjVhO6UmjbF//SOlzzwGQMPvb0S59mqdIxr4HBXG1Nuu7An73yc7OxuTybTTbIDV1dU9JnxKSorhx5iL/qdY+n7xJWK1senkM9hy3NydCnRnZQVjHryHMQ/eQ/OY8VQdPpu66T+hZcwEtP6c4CMcxr1hLRnfLSL7q0/J+uYLrM1NO8dqNlNz8Ewqjp1LzcGHolms/ReDGFB7kqMDzWZWKO1iAqR4cVhMDE13sK7Bp1sMwtg5uqeSrSgHSLWbGZ7pYG29Mf5u7GaVUTkDux680zmwFw+tZpVxuS6WV3sIhrcX5yt+fSNZX3+Oo2or2d98wdDnnmLj6ecOaCyiZ6rPt33SN5sN9ZxfYIRrVAOdo/Fy77330traSkpKCldfrf8Fj2SRsP+BrFYr06ZN44MPPuD4448HopO/ffDBB1x22WX6BicMJ9JSiynHtWev0VGgbzrxNHI//YghL/6HnM8XxJbgSFu1grRVKxjz0F8JOZw0TdqLpklT8AwtpW1YKd6SYdEx3T28KVH9fqwNdbg2b8S1cT2ujetIXbmctLJlsbXGf0xTVRqm7kPV4bOpnH0MgczsPfo5hT76I0cHgqLAyCyn7uuK56VYaW4PUe8N6hrHYGbUHN1TFpPCuFwXLmvyXbTPdVvxhyJsadZ3Ph2zqjA214l1gJen2rJlC+PHjx/QYzgsJsbnulhRs704D6WksvSWu9n/ojMAGPu3u6ib/hM8w6S3ml4K330Ta0u0C3/whBPQsrJ0jigqHjkaD4899hiVlZUUFBRIYd6PErYwB7jmmms4++yz2Weffdhvv/2477778Hg8nHuuXKUUA8hkombGEdTMOAJHxWYK3nmT/A/nk77DpHBmn5fsrz8j++vPOn1rxGQimJJGMC0NzWyJjpPSItGCvKmx2+L7xwKpadTveyC1Bx1KzYwjCGQa4x+OSD4laf27zvCeKM1y4AmEaQ9Jn3bRP+xmlbG5zoSZHbwvStLthCLazrOKx4lJJenOsdNqYkKeixXVHgIdxXn9ftPZcNrZDHvuKUx+P1NuvJYvnnipf3vPiV4b+uL2Sd/8552nYyRC9F5CtxannnoqtbW13HTTTVRVVTF16lTmz58va+qJnahpAzOex1dUwrrzLmHdeZdgr6wgd+GHZH63iIzvvsFRXblzHOEwtqYGbE0Nu3ec/EKax0+maeIU6vc7kOaxE0GGZCSVgcrRPZHpNFOUFr9Z2HfFrCqMznGyvFrGm+vBiDm6J1JtJkbnOLEM8F1cIxie6SCiadS0xbfHiUmFsTkuUuI0RGDo0KFxOQ5E75xPzHdTVuPBF4w2SCuv+C05ny/AtWkD6cuWUPrkw6y9QHpxxlveyqWxFXRCU6YQnjZN54i2i2eODqQnnniCQCCA1SpDJvtTwk7+trtkHfPBLdJSi5qaE9dj2rduIWXtatwb1uHauA5HxWaszc1YWpqwNDehhMNoqhqd2dRiJZCeQTA9g0BGJt7CYjxDS/EOGUZr6SgC2fGNXcSfHjnaE6dFZWK+MddxrvcEWV3n3fUTRb8yWo7uifwUK0Mz7AM63tmI4rnG+bbu6/EqyiHaTbi4uDhuxwMIRTRW13ppbg8BkL70Ow48dy5KJELEbOGzf79K65jE77qcKHLdFib+/hrsL7wAgOf++wmcdZbOUW2nR44K/SX95G9C7A7N7wHi+4ayvbCY9sJiag85bNdPFoOeHjnaHYtJYUyucddxznJZKAnZ2Nyk77jZwcZIOdpXFlN0IsNMp0XvUHQxLMOB1aSyqamdgbwtYzNH14J3xnncfktLS1yPB9snDtzU1M7WlgBNk/di7TkXM/JfD6GGgky58Ro+//drRKzJO3miEdjMCsMzHGQ212H73/8AiGRkEDjpJJ0j60yPHBWJI/n7bwkBKLJsiTA4o+SoSYUxOU7sZmPE053iNDu57sFZXOnFKDnaV1kuC5ML3IO2KN+mMNXGmJyBm9AxzW5mUr477kU5oFuPSEVRGJrhYGyOE4tJofyiK2gZNRaA1DWrGPXw/brENRiYVChOszGlIIUMpwXb44+jhKK9F/znnAMGmwXd6L12hb6kK7sBSVd2IYQeVCValKc7EqNw0TSN1XVeGrwhvUMRBuayqgxJtydMXseLPxRhbb0v1gV7T6lK9IJZYaoVZZANEdhRIBxhQ2M7ge+WctAZx6GGgmiqyhePv0DTVOOMdU50FpNCrttKQYp1+zwRXi9pkyahNjaiWSw0f/89WkGBvoEmqTVr1hAKhTCbzYwaNUrvcAyvt13ZE/vytxC9FK7bqHcIQvRI7xxVFRiVnThFOUTvUkVjllFZ8aB3ju4up0VlZLaDSfnuhMrreLGZVcbnuRiR5cBq2rNCOt1hZnKBm6I0m65FeVlZmW7H3sZqUhmd7aTo4GlsuDS6jJQSiTDlpl9j8sncGHtCVaK5Nirbwd5FKQxJt3eavNH6/POojY0ABE44wZBFuRFytD8cf/zxHHTQQbElq0X/kHczBuS0qIQHR0eGuPEq4LTKdajd1ds01Lr5+scbNLROr7ntS02L7tv+de+PnTR0/IFVBUZnO8lIwC6+qqIwJsfJGrlzPvAS4I9SUSDdbibPbU3IfNZDrttKltNCjSdAVUug18sRKkq023pRqnGWVDRSJ9AMhwXthmvxLfwAx+JvcW3ewJj772TF9bfqHVrCUJToe+IUm5lUu5l0u7n7uU8iEewPPxx76L/kkjhFuXuMlKPCeIzRkopOJuS79Q4h6WzVsiksSNE7DLGbNE1j26iOiBYt6iM7bI9s+xyJfg5rGmFNIxLRCEeij0OR6Ec4ohGMaITCmiGHiih2ff7uzWq0sDXKG+u+UBWF0dlO1jX44r4cVKJSleib3q50d3FMrxzdFVUh+qbdYSbLacE6CJY/628mVaEgxUZBio1Wf4hGX4hWfwhfMEIoEm17VQWsZhWXxUSK3USWw4LVYHNRpKen6x1CJ4rFQuDhf2CfMQPF52PY889QM/On1O1/sN6h6U5RojllVhXMqorFpGA1KdjMKnazisNiwm5Wez0Jqfn99zGtWQNA8KCDCE+ZMpDh95nRcrSvTjrpJJqampLm5zGKxH0nJsRuSE1N1TsE0QeKorCth6WJ/useqWkawXC0UA+EIgQjGv5QhEA4gj8U/dofjsT1BqFic8XvYB0cFpUxOU4cFmPPu9EbiqIwIsuJ3exnc/PAzjhtFCY12m02+qFgMW1/g2tWFUxq9LOqgElRUDu+7u0SYT++ANbWBnanm3DHha5tF77CHRfCQhGNkNZxESwc/RzqeF5//T5UJdoF22kx4bKquG1m3FaTYVcQSEQpNnOnJc623eFLhHHjRvxfHxk5Et/NN+O8/noApt12PWvmf0iDxUlLewg9rxOrCrF2YlubYVKj/3ejnxVURUFVo22IooBK9LOigEI0L7Zlxo7bgM7bFQUVUNXo8/v7b9b+j3/Evvb/6lf9+tr9yYg52he33io9PwaCFOZiUNi0aRPjx8s6oiJKURSsZgUr4Opm5mBN0/CHNdqDYXyhCO3BCL5gBG8wTDDc/++kIs3VmHKG9fvrdifLaWFEliPpCpqiNBsuq0p5vW9Afk/xZjEpOCzb7yBtu5tkM6sDNqv2NtsujG27KLZ2a0Wf2lFN0whr7FDEby/stxX+O96l3/amX0XBpEbfwFtUBaspetEhEQrEZJJI59uo/+v9F1yAZd48LJ98gqmiguF3/IG8v/+diKbR5g/jCUQ/fKEI/lCkT22X0nEBzmyK/r2Y1ejFOouqYjbt+FjBbFKxdBThycD03XdYFiwAIFxaSnDWLJ0j6p5Rc1QYgxTmQgjRBUVRsJsV7GaV9B/tC4QjeANhvMHo521vqBLhLq1Jja5lnOu26h3KgEl3WJhcYGJ9gy9hxp2rCjitJpwWFafVhMtiwmFRO01slKgURcHc0WVViEFJVfE8+CBpBx2E0tqK7dlnCf7sZwSPOopUu3mnoUQRLTrsKthxESt60Wr7Pxi14+61Sdl+l3sw/33Z77039nX7ZZdBgi/tKAYvKczFoFBSUqJ3CCKJWE0qVodKumP7tnBEi931aA2EafOH8Id6X6mrqbkDEGln2S4LQ9PthhsXOhCsJpUxOS7qvUE2Nvp263cx0KITGplwW024bSZcHQW50e9MSjsqjM7IOaoVF+P9059wXXYZAM6rr6Zlv/3QsrN3eq66Q68u0TN15Uqsb74JQCQ/n8Bpp+kcUc+MnKNCf1KYi0GhtbWVlBSZ/E0MHJOqxO58bFugJRCO0OYP0+YP0xoI4QmECXcz4bEW8KHYnAMSW7rDTEmaDbdt8DX5WU4LGQ4z1a0BKlr8unRvt5oUUmxm3LZoMe5K0DHR0o4KozN6jgZ+/nMsb72F9e23UWtrcV5zDZ6nnup+JkaxS/a//S32dfuvfgV2u47R7JrRc7S35s6dS11dHdnZ2bz00kt6h5M0Bt+7NDEoNTU1UVhYqHcYYpCxmlQynSqZHcs2aZqGN9hRrHcU6t5gtAu81t4KKVn9dmyLSSHLaSHPbcXZzTj6wUJVFApSbeSlWKn3BqlpC9DSHh6QY5lUcFujE5K5bCZSrKak6aEg7agwOsPnqKLg/etfMX/1FWpDA9Y33yT40ksETj5Z78gSkrppE9YXXwQgkpGB/5xz9A2oFwyfo71UVlZGZWUlBQZcKz6RSWEuBgWjdxEVg4OiKLg67pjmdXRSDEc0vMEw5S1m0lOseINh2oMRArt5Z3fbGOUUm4n0jjv3vZ19e7BQFYUcl5Ucl5VAKEKjL0SzP7Tbww6irxWdHdxhiU7K5rKYcFrVpJjhvjvSjgqjS4Qc1fLy8P7lL7jPPRcAx3XXETzoILQkKNbizfbAAyjh6EVW/0UXgduYSzruKBFytDesVmvsQ/QfRRskK92HQiEWLFjAlClTMJmS942TECI5hCNax/Jt0XWEgxGNSASic1dHi0xTxyy79o6ZupPlH74eQhENXzBMIKwRDEcIR0BDiy39s+NsxzZzdHkyOd9CiL5yXnghto4uwMHDDqPtxRelS/tuUKqrSZs6FcXvR3O7aV6yBC0jQ++whOhSOBxmyZIlzJgxA7O5+/viydG/TohdWLVqld4hCNGjH+eoSVVwWEykOyxku6wUpNgoSrNRlGanKM1OQaqNXLeVDIcFh8UkReIeMqvRceBZTgv5Hee6uONcF3Z0g89yWki1m7EN0osg0o4Ko0ukHPX9+c9EOroBWz78EOuTT+obUIKx33cfit8PgP/ccxOmKE+kHBXxJ4W5GBTC4YEZTypEf5EcFUYnOSqMLpFyVEtPx3P//bHHzptuQl2/XseIEoeyZQu2J54AQHM4opO+JYhEylERf1KYi0EhNTVV7xCE6JHkqDA6yVFhdImWo6EjjohNWKZ4PLguvBCCQX2DSgCOe+5BCQQA8F94IVpens4R9V6i5aiILynMxaCQmZmpdwhC9EhyVBid5KgwukTMUe+ttxIePhwA87ff4rjjDp0jMjZ13Tqs//kPAFpKCu1XXKFzRLsnEXO0K08++SR///vfeVKGYPQrKczFoLBhwwa9QxCiR5KjwugkR4XRJWSOut14Hn0UrWNCKPv992P+6COdgzIu+113xWZib7/00oQZW75NQuZoF+6++25uvPFG7r77br1DSSpSmAshhBBCCKGT8N5747vxxthj1yWXoNTU6BiRMakrVmDtmMk+kplJ+8UX6xyREP1L1jEXg0JRUZHeIQjRI8lRYXSSo8LoEjlH/ZdeimXBAiwffohaU4PrV7+i7YUXQJV7aNs47rgDpWOV5/YrroAEHK+dyDm6o7/+9a/4fD4cDofeoSQV+WsXg4LP59M7BCF6JDkqjE5yVBhdQueoquJ56CEiublAdAk129//rnNQxmH+5BOs8+YBEMnPx3/BBTpH1DcJnaM7mDVrFscffzyzZs3SO5SkIoW5GBQaGhr0DkGIHkmOCqOTHBVGl+g5quXm4vnHP2KPHbfdhmnxYh0jMohwGMcNN8Qe+m68EZxOHQPqu0TPUTGwpDAXQgghhBDCAEIzZ8ZmGldCIVznnosyyIs56zPPYF6xAoDQ1KkETj1V54iEGBiKpnUM1khyoVCIBQsWMGXKFEwmk97hiDjTNA1FUfQOQ4huSY4Ko5McFUaXNDkaDJLys59hXrQo+nDmzOh488H4/rWlhbR99kGtq4s+nDeP8AEH6BxU3yVLjjY0NBCJRFBVNWmWgBtI4XCYJUuWMGPGDMzm7qd4kzvmYlAoLy/XOwQheiQ5KoxOclQYXdLkqMVC2xNPEMnJiT786CPsd96pc1D6cPzlL7GiPHDCCQldlEPy5OghhxzC6NGjOeSQQ/QOJalIYS4GhWAwqHcIQvRIclQYneSoMLpkylGtqAjP44+jddwld/zlL1jmz9c5qvhSV63C9vDDAGg2G74//EHfgPpBMuWo6H9SmItBwe126x2CED2SHBVGJzkqjC7ZcjR08MH4brop9th58cWo69bpGFEcRSK4rroKpaOQbb/sMiIlJToHteeSJUcPPfRQZs+ezaGHHqp3KElF1jEXg0JOR3cwIYxKclQYneSoMLpkzFH/ZZdh/vZbrK+/jtrSgusXv6D1nXfA5dI7tAFlffJJzF99BUC4tJT2a67ROaL+kSw5+ndZym9AyB1zMSisX79e7xCE6JHkqDA6yVFhdEmZo4qC54EHCI8aBYB5xQpcl18OkYjOgQ0cZetWnDt0W/feey84HPoF1I+SMkdFv5HCXAghhBBCCKNKSaHtmWfQOrpBW199NXkng9M0nNddh9LWBoD/zDMJyQRjYpCQwlwMCoWFhXqHIESPJEeF0UmOCqNL5hyNjB6N59FH0TqW2nLccw/WF17QOar+Z3nlFazz5gEQyc3Fd+utOkfUv5I5R8Wek8JcDAqBQEDvEITokeSoMDrJUWF0yZ6jwVmz8N12W+yx84orMH35pY4R9S9182acO4wl9/7pT2jp6foFNACSJUcvvPBCTjrpJC688EK9Q0kqUpiLQaGuYw1MIYxKclQYneSoMLrBkKP+Sy7Bf+65ACiBAO4zz0RdtUrnqPpBKITrl79EbWkBomuWB48/Xt+YBkCy5Ohnn33GRx99xGeffaZ3KElFCnMhhBBCCCESgaLgvfNOgh3LVKkNDaScdBLKli36xrWH7HffjfnrrwEIDxmC5957oaPbvhCDhaJpmqZ3EPEQCoVYsGABU6ZMwWQy6R2OiLNwOCy/d2FokqPC6CRHhdENqhxtaSHl2GMxL10KQHjUKFrnzUPLytI5sN1n/uwz3McdhxKJoJlMtL71FuH99tM7rAGRLDna2tqKpmkoikJKSore4RheOBxmyZIlzJgxA7O5+9XK5Y65GBQ2bNigdwhC9EhyVBid5KgwukGVo6mptL34IuERIwAwrVmD+9RTobVV58B2j7JlC64LLkDpWP6t/Xe/S9qiHJInR1NSUkhNTZWivJ9JYS4GBb/fr3cIQvRIclQYneSoMLrBlqNaTg5tL79MpKAAAPPixaTMnQsd47QNr60N9xlnoFZXAxA85BDar7xS56AG1mDLUbF7pDAXg4LL5dI7BCF6JDkqjE5yVBjdYMzRyJAhtL70EpGMDADM33xDyoknojQ16RvYroTDuC66CPMPP0QfDhuG51//giTo5t2TwZijovekMBeDQn5+vt4hCNEjyVFhdJKjwugGa45Gxo2j7bXXiHSMLzcvXoz7hBNQGhp0jqx7jltvxfr22wBEUlNpe/bZhBwfv7uSJUffeOMNXnjhBd544w29Q0kqUpiLQWHt2rV6hyBEjyRHhdFJjgqjG8w5Gp44kdbXXiOSkwOAeckSUn72M9TNm3WObGe2v/8d+wMPAKCZTHj+9S8iY8boHFV8JEuOXn/99Vx88cVcf/31eoeSVKQwF0IIIYQQIsFFxo+n9fXXieTlAWBatYqUI4/E9N13Oke2ne3xx3HeeGPsse/OOwkddpiOEQlhHN3P1y5EEkmWrkMieUmOCqOTHBVGJzkKkTFjaH3rLdynnopp7VrUmhpSjj4az6OPEjzqKF1jsz34IM6bboo99v3ud/jPP1/HiOIvWXL097//PV6vF6fTqXcoSSUh75hv2LCB888/n+HDh+NwOBgxYgQ333wzgUBA79CEQUU6luEQwqgkR4XRSY4Ko5McjYqUltL6zjsEDzwQAMXnw33mmThuuQWCQR0CiuC45ZbORfnVV9P+61/HPxadJUuOnn766VxwwQWcfvrpeoeSVBKyMF+5ciWRSIRHHnmE5cuXc++99/Lwww9zww036B2aMKiamhq9QxCiR5KjwugkR4XRSY5up2Vm0va//+GfOze2zX7//dFx5xs3xi+Q1lZc55yD/f77Y5t8N9xA+//9HyhK/OIwCMlR0ZOE7Mo+e/ZsZs+eHXtcWlrKqlWr+Mc//sE999yjY2RCCCGEEEIYgM2G95FHCE+ejOO221CCQcyLFpEyYwa+W28lcOaZoA7cPTrT8uW4zj0XU3k5AJqq4rvrrkHXfV2I3krIO+ZdaW5uJjMzc5fPa21tpaWlJfbh9/vjEJ3Q26hRo/QOQYgeSY4Ko5McFUYnOdoFRcF/2WW0vv024WHDAFBbWnBddRUpP/0ppsWL+/+YgQD2u+8m5bDDthflKSm0PffcoC/KJUdFTxLyjvmPlZeX88ADD/TqbvnEiRPxer2xx+eeey6XX345BQUFsSUM8vLy0DQt1t1k1KhRbNmyBZ/Ph91up6SkhDVr1gCQm5uLqqpUVVUBMGLECKqqqvB4PNhsNoYNG8aqVasAyM7Oxmq1snXrVgCGDx9ObW0tbW1tWCwWRo4cSVlZGQCZmZk4HA4qKioAGDZsGA0NDbS0tGAymRgzZgxlZWVomkZ6ejopKSls7lgSY8iQIbS0tNDU1ISiKIwbN45Vq1YRDodJTU0lIyODjR3dmIqLi/F6vTR0rHU5fvx4Vq9eTSgUIiUlhezsbNavXw9AYWEhfr+f+vp6AMaOHcu6desIBAK4XC7y8vJYt24dAAUFBYRCIWprawEYPXo0mzZtor29HYfDQVFREeUdjXVex+yh1dXVAIwcOZKKiorY+R4yZAirV68GICcnB7PZTGVlJRDtLVFdXY3H48FqtVJaWsrKlSsByMrKwmazsXXrVjweDxMnTqSuro7W1lbMZjOjR49mxYoVsfPtdDrZsmULAEOHDqWxsbHb852amsqmTZsAKCkpobW1tdvznZmZyYYNGwAoKirC5/PFzve4ceMoLy8nGAzidrvJycnpdL4DgQB1dXUAjBkzhg0bNuD3+3G5XOTn58dyNj8/n0gk0ilnN2/eHDvfxcXFnXJWUZTY+R4xYgSVlZV4vV5sNhtDhw7t8XzX1NTQ1tbW5fm22+1d5uyPz3dGRgZut7tTzjY3N9Pc3IyqqowdOzY2ZCUtLY20tLRO57utrY3Gxsadcrar893e3t5lzrrdbnJzc3vM2Y0bN+L3+3E6nQPeRng8HoYOHSpthE5txLbzLW1E921EbW0tmZmZ0kbo1EaAvI/YVRuxbt06XC6XtBFdtRF2O2M/+AAuu4ysjvXDzYsXk3LkkdTOmkX7tdfiHzp0z95HtLWR98UXDHv4YcwdMQD4J0xg2U030V5YSGl7+6BuI5YuXRrL9URuI+bOnUtVVRU5OTm8+uqrCdNG6PU+QtM0ekPRevvMOLj++uu56667enxOWVkZY8eOjT2uqKhgxowZHHrooTz22GPdfl8oFGLBggWUlpai7tBtx2azYbPZ9jx4YWgrVqxg/PjxeochRLckR4XRSY4Ko5Mc7R3zwoU4r7sOU0cxB9Fu5sGf/pTAL35B8LDDwGrt/Qu2tGB99VXsDz+MqaNggega5e1XXhmd5M1u788fIWElS45OmDCByspKCgoKWL58ud7hGF44HGbJkiXMmDEDs7n7++KGumN+7bXXcs455/T4nNLS0tjXW7duZebMmUyfPp1//vOfvTpGSkoKJpNpT8IUCcjhcOgdghA9khwVRic5KoxOcrR3QoccQssnn2B79FHsd9+N2tyMEolgnT8f6/z5RFJTCc2cSeiggwhNnEiktBQtMxPMZgiFUGprMa1fj2nJEswLFmBZuBDF5+t0jOBBB+H74x8JT56s009pTMmSo1OmTKGoqIjs7Gy9Q0kqhrpjvjsqKiqYOXMm06ZN49///vcui+1td8ynTJkihfkgFAwGsVgseochRLckR4XRSY4Ko5Mc7YOWFmxPPIH90UdRO7rjdkczmVDC4R6fE9p3X9qvvprgrFmDctb1XZEcHZx6e8c8ISd/q6io4NBDD2XIkCHcc8891NbWUlVVFRt7IcSPbRunI4RRSY4Ko5McFUYnOdoHqan4r7yS5u+/p+3ZZ/GfcgqR1NQun9pdUR7Jy6P9/PNpef/96Prps2dLUd4NyVHRE0N1Ze+t9957j/LycsrLyykuLu60L0E7AAghhBBCCKEPs5ngrFnRO92hEKYffsC8aBHqqlWoFRWo9fUQDoOqEsnJIVJURHjCBML77EN4wgQpxIXoBwlZmJ9zzjm7HIsuxI5yc3P1DkGIHkmOCqOTHBVGJznaT8xmwnvtRXivvfSOJOlIjoqeJGRhLsTuUuRKrjA4yVFhdJKjwugkR4XRJUuO/va3v6WpqYn09PRdrqglei8hx5gLsbu2rVsohFFJjgqjkxwVRic5KowuWXL0zTff5MUXX+TNN9/UO5SkIoW5EEIIIYQQQgihI+nKLgaFESNG6B2CED2SHBVGJzkqjE5yVBhdsuTovHnzCIfDsgR1P5M75mJQqKys1DsEIXokOSqMTnJUGJ3kqDC6ZMnRoUOHUlpaytChQ/UOJalIYS6Snt/v54EHHsDv9+sdihBdkhwVRic5KoxOclQYneSo2BUpzEXS8/v9PPHEE9IQCsOSHBVGJzkqjE5yVBid5KjYFRljLoQQQgghhBCiVz799FP8fj82m42DDz5Y73CShhTmQgghhBBCCCF65aKLLqKyspKCggKWL1+udzhJY9AU5pqmARAOh3WORMRbJBLB6XQSiUTk9y8MSXJUGJ3kqDA6yVFhdMmUo3a7HafTid1uT/ifJR62naNt9Wh3FG1Xz0gS7e3tfPbZZ3qHIYQQQgghhBBikDnooIOw2+3d7h80hXkkEiEQCGAymVAURe9whBBCCCGEEEIkOU3TCIfDWK1WVLX7udcHTWEuhBBCCCGEEEIYkSyXJoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0JIW5EEIIIYQQQgihIynMhRBCCCGEEEIIHUlhLoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0JIW5EEIIIYQQQgihIynMhRBCCCGEEEIIHUlhLoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0JIW5EEIIIYQQQgihIynMhRBCCCGEEEIIHUlhLoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0JIW5EEIIIYQQQgihIynMhRBCCCGEEEIIHUlhLoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0ZNY7gHiJRCIEAgFMJhOKougdjhBCCCGEEEKIJKdpGuFwGKvViqp2f1980BTmgUCAzz77TO8whBBCCCGEEEIMMgcddBB2u73b/YOmMDeZTABMnDgx9rUYPDweDy6XS+8whOiW5KgwOslRYXSSo8LoJEcHp3A4zLJly3ZZg+pSmDc3N+N2u+NaIG/rvm4ymaQwH4S2bNnC+PHj9Q5DiG5JjgqjkxwVRic5KoxOcnRw29Vw6rhN/rZo0SJmz56N0+kkKyuLBQsWAFBXV8dxxx3Hxx9/HK9QhBBCCCGEEEIIw4hLYf75559z8MEHs2bNGs4880wikUhsX3Z2Ns3NzTzyyCPxCEUMUsXFxXqHIESPJEeF0UmOCqOTHBVGJzkqehKXruw33HAD48aN48svv6S1tZXHHnus0/6ZM2fy1FNPxSMUMUh5vV5SU1P1DkOIbkmOCqOTHBVGJzkqjC4Rc1TTNJbULuHjTR+zuHoxm1s2s/qh1WgeDXuqneNuOo598vdhzvA5ZDoy9Q43ocWlMP/mm2/405/+hM1mo62tbaf9RUVFVFVVxSMUMUg1NDSQn5+vdxhCdEtyVBid5KgwOslRYXSJlKMt/haeXv40T/zwBOub13feuQlohfaUdp5a9hRPLXsKs2pm9vDZXL3P1eyVt5cuMSe6uBTmFoulU/f1H6uoqMDtdscjFCGEEEIIIYQQXYhoEZ5e/jS3fnYrTf6mnfYrKGiKttP2UCTEm2vf5M21b3LcyOO44yd3UOAuiEPEyUPRNG3nM9vPZs+eTVtbG59++in19fXk5OTw/vvvc9hhh+HxeJgwYQL77rsvL7744oDFEAqFWLBgAVOmTJFZ2YUQQgghhBBiB2X1ZVzz4TV8VflVp+2F7kLGZ41naOpQMu2ZsdnF/WE/NZ4aypvKWV63HG/IG/ueFGsKfz3sr5w0+qS4/gxGFA6HWbJkCTNmzMBs7v6+eFwmf7vllltYtGgRP/vZz3j77bcBWLJkCY899hjTpk2jtraWG2+8MR6hiEFq9erVeocgRI8kR4XRSY4Ko5McFUZn1Bz1hXz88Ys/cuizh3YqykdljOL8Sedz5vgz2Ttvb7IcWZ2W/LKZbJSkljBzyEwunnoxRw49EofZAUBroJVfzv8l1y+4nlAkFPefKRHFpSv7/vvvz7x587jkkkv4xS9+AcC1114LwIgRI5g3bx6TJ0+ORyhikAqFpEEQxiY5KoxOclQYneSoMDqj5agn6OHFVS/ywLcPdBpHnmpNZfbw2QxLG9br1zKrZvbK24uxmWN5f9P7lNWXAfDPJf9kc8tmHp39KE6Ls79/hKQy4IW5pmm0trYyffp0Vq1axffff8+aNWuIRCKMGDGCadOm7XKxdSH2VEpKit4hCNEjyVFhdJKjwugkR4XRGSVHvUEvjy19jAe+fYD69vrYdlVR2SdvHw4uPhiz2rcy0WFxcHTp0QxJGcJ7G98jokV4e/3bnPLaKbxw3AtSnPdgwAvzQCBAZmYmd9xxB9dddx1Tp05l6tSpA31YITrJzs7WOwQheiQ5KoxOclQYneSoMDoj5OiCzQu46oOr2NiysdP2AlcBs4fPJseZs8vX+OG9Hwi2B7HYLUw6ctJO+xVFYUruFNJt6byy5hUCkQCfb/2cM948g2ePeRa72d5vP08yGfAx5jabjfz8fGw220AfSohurV+/ftdPEkJHkqPC6CRHhdFJjgqj0zNHw5Ewt3x2Cye8ckKnonxk+kh+MeEXnDXhrF4V5QALn17I+/94n4VPL+zxeUPThnLq2FOxmqxA9KLAxe9eTETrfrWuwSwuk7+dc845PP300wQCgXgcTgghhBBCCCEE0BZo47Q3TuP+b++Pbct35XPuxHM5cfSJ5LsGbm31AncBJ48+GYtqAeD18te5/fPbB+x4iSwuk79NmjSJV199lQkTJnDOOecwbNgwHA7HTs878cQT4xGOGIQKCwv1DkGIHkmOCqOTHBVGJzkqjE6PHG1qb+KU109hUdUiILoO+cHFB3NAwQF9nufriIuPIOgPYrFZevX8opQijh15LP9b/T80NO779j7GZY3j5LEn9+n4ySouhfnPf/7z2NfdLYumKArhcDge4YhByO/36x2CED2SHBVGJzkqjE5yVBhdvHO0xd/Cia+eyPc13wNgNVk5fuTxuzXbeldGHzR6t79nRPoIDh96OO9vfB+Aqz68igk5ExifNX6PYkkmcSnMP/roo3gcRohu1dfXk5eXp3cYQnRLclQYneSoMDrJUWF08czR9lA7Z711Vqwot5vtnDrmVPJc+v2N7J23N9Wean6o+wFfyMc5b53Dxz//WGZq7xCXwnzGjBnxOIwQQgghhBBCDGqapnHlB1eycEt0cjabycbPx/6815O7DaQjhh1BtbeaGm8N5U3l3Pr5rdw54069wzKEuEz+tqMVK1bw9ttv8/bbb7NixYp4H14MUmPHjtU7BCF6JDkqjE5yVBid5Kgwunjl6IPfPciLq14EwKyYmTt6br8W5X6vP/axuyyqhWNGHINZid4f/ueSf7Jg84J+iy2Rxa0wf+211xgxYgSTJk3i6KOP5uijj2bSpEmMHDmS119/PV5hiEFq3bp1eocgRI8kR4XRSY4Ko5McFUYXjxxduHkht3x2S+zx7NLZFKUU9esxHr/oce6fez+PX/R4n74/y5HFjJLtPaove+8yWvwt/RVewopLYT5v3jxOOukkAO644w5eeeUVXnnlFe644w40TePEE09k/vz58QhFDFKyVJ8wOslRYXSSo8LoJEeF0Q10jtb76jutE75v/r6GnVxt77y9GZIyBICKtgp+98nvdI5If3EZY37bbbcxefJkFi5ciMvlim0/9thjueyyyzj44IO55ZZbmD17djzCEYPQjnknhBFJjgqjkxwVRic5KoxuIHNU0zQuf/9yKj2VABS5izi05NABOVbJpBJ8LT4cqTsvf91biqJwVOlR/GvZvwiEAzxb9izHjzqeI4cd2Y+RJpa43DFfunQpZ599dpfJ6HK5OOecc1i6dGk8QhGDlMzSKoxOclQYneSoMDrJUWF0A5mjL616ifnroz2Q7SY7x408rs/rlO/KMdcdwym3n8Ix1x2zR6+Takvl8CGHxx5f9/F1tIfa9zS8hBWXwtxut9PQ0NDt/oaGBux2ezxCEYOUjDsTRic5KoxOclQYneSoMLqBytE6b12nruCzhs3CbXUPyLH628TsiZSklACwsWUjf/v2bzpHpJ+4FOaHHXYY999/P1988cVO+7766iv+9re/ccQRR+z261ZUVHDmmWeSlZWFw+Fg0qRJLFq0qD9CFkIIIYQQQgjD+/3C39PQHr0JOiJ9BGOyxugcUe8pisKRQ49EIXp3/95F97KheYO+QekkLmPM//znP3PggQdy8MEHs99++zFmTDRZVq1axddff01ubi533XXXbr1mY2MjBx10EDNnzuTtt98mJyeHNWvWkJGRMRA/gkhwBQUFeocgRI8kR4XRSY4Ko5McFUY3EDn63ob3YkujWU1WZg9PvDm7sp3Z7JO/D99UfYM/7OeGT27gv8f8V++w4i4uhfnw4cNZunQpf/rTn3j77bd5/vnnARg6dChXXnkl119/Pbm5ubv1mnfddRclJSU88cQTnY6zK62trajq9o4CNpsNm822W8cWiScUCukdghA9khwVRic5KoxOclQYXX/naFugjWs/ujb2eGbJTFyWgZ8Ecd5f5+Fr9eFIcXDUNUf1y2tOL5pOWX0ZbcE25q+fz/x185ldmngXGfZEXApzgNzcXO69917uvffefnm9119/nVmzZnHyySezYMECioqK+NWvfsUvf/nLHr9v4sSJeL3e2ONzzz2Xyy+/nIKCAtauXQtEJ2bQNI2amhoARo0axZYtW/D5fNjtdkpKSlizZk3s51JVlaqqKgBGjBhBVVUVHo8Hm83GsGHDWLVqFQDZ2dlYrVa2bt0KRC8k1NbW0tbWhsViYeTIkZSVlQGQmZmJw+GgoqICgGHDhtHQ0EBLSwsmk4kxY8ZQVlaGpmmkp6eTkpLC5s2bARgyZAgtLS00NTWhKArjxo1j1apVhMNhUlNTycjIYOPGjQAUFxfj9XpjcwCMHz+e1atXEwqFSElJITs7m/Xr1wNQWFiI3++nvr4egLFjx7Ju3ToCgQAul4u8vLzY2JmCggJCoRC1tbUAjB49mk2bNtHe3o7D4aCoqIjy8vLY+Qaorq4GYOTIkVRUVMTO95AhQ1i9ejUAOTk5mM1mKiujM06WlpZSXV2Nx+PBarVSWlrKypUrAcjKysJms7F161ZaW1txu93U1dXR2tqK2Wxm9OjRrFixIna+nU4nW7ZsAaIXjRobG7s936mpqWzatAmAkpISWltbuz3fmZmZbNiwAYCioiJ8Pl/sfI8bN47y8nKCwSBut5ucnJxO5zsQCFBXVwfAmDFj2LBhA36/H5fLRX5+fixn8/PziUQinXJ28+bNsfNdXFzcKWcVRYmd7xEjRlBZWYnX68VmszF06NAez3dNTQ1tbW1dnm+73d5lzv74fGdkZOB2uzvlbHNzM83NzaiqytixY1m5ciWRSIS0tDTS0tI6ne+2tjYaGxt3ytmuznd7e3uXOet2u8nNze0xZzdu3Ijf78fpdA54G9Ha2oqmadJG6NRGbDvf0kZ030ZUV1fT0tIibYRObQTI+4hdtRHr1q2jtrZW2gh5H2HYNmJbjvZXG/HP9f9kS2s058a5x3F41uEEI0EKtUIA1qvrydFycGtuggQpV8sZFxkHQIPSgE/xURSJrnG+Qd1AppZJqpZKmDCrTasZGx6LgkKT0kSr0kpJJDoW/OHvHqalvoWMrAzGhsey0rSS0eHRmDDRorTQqDQyNDIUgC3qFpyak0wtE4AyUxmjwqMwY6ZVaaVOqWN4JHpz9WclP+P5ddEbuNd9cB0zS2ayeePmhH8foWkavaFovX3mHgiFQni9XlJTU7vc39LSgtPpxGzu/XWCbZPFXXPNNZx88sl88803XHnllTz88MOcffbZXcawYMECSktL5Y75ILRixQrGjzfmOo5CgOSoMD7JUWF0kqPC6PozR9c3refAfx9IIBJAVVTOn3Q+Gfb4DOl96KyHaKtvw53l5lfP/KrfXlfTNJ5d+WzsYsOfZ/yZC6Zc0G+vr5dwOMySJUuYMWNGj/VuXArzX/3qV3zyyScsW7asy/2TJk2KTRDXW1arlX322YfPP/88tu2KK67gm2++6XKSuW2F+ZQpUzCZTLv/Q4iEFgqFduvCjxDxJjkqjE5yVBid5Kgwuv7M0V+89QveXPsmANPypnH40MN38R39x9fiQ9M0FEXZo7XMu1LZVskzK54BIMuexbdnf0uqreubu4mit4V5XGZlnz9/PnPnzu12/9y5c5k3b95uvWZBQcFOV5zGjRsX66YixI4kL4TRSY4Ko5McFUYnOSqMrr9y9NMtn8aKcofZwcHFB/fL6/aWI9WBM83Z70U5QIG7gLGZYwGob6/ngcUP9PsxjCouhfnWrVspKirqdn9hYWFsPElvHXTQQbExV9usXr2aoUOH9ilGkdza29v1DkGIHkmOCqOTHBVGJzkqjK4/cjQcCXPDJzfEHh9SfAg2U3INyz2k+BBUJVqm/n3x36lsq9Q5oviIS2GelZW1UxG9o7Kysm7Hn3fn6quv5ssvv+SOO+6gvLyc//73v/zzn//k0ksv3dNwRRJyOPr/ip4Q/UlyVBid5KgwOslRYXT9kaP/XvFvltVFhwdn2bOYkjNlj1/TaDLsGeyVuxcA7eF27vzqTp0jio+4FOazZ8/mkUce4bvvvttp3+LFi/nnP//JnDlzdus19913X1555RWeffZZJk6cyG233cZ9993HGWec0V9hiyTSU48NIYxAclQYneSoMDrJUWF0e5qjW1q3cPOnN8ce/3TYT1EUZU/D2m3lX5WzcuFKyr8qH7BjHFh4IFaTFYD/rPgPqxtWD9ixjCIuM2TcdtttzJ8/n/32249jjz2WCRMmALBs2TLeeOMNcnNzue2223b7dY8++miOPvro/g5XJKHy8nKZqVUYmuSoMDrJUWF0kqPC6PYkRyNahEvfu5SWQAsAozJGUZJa0p/h9dq7D74bm5V95P4jB+QYTouTAwoO4JMtnxDRIvz56z/z2OzHBuRYRhGXO+aFhYUsWrSI008/nQ8++IDbb7+d22+/nQ8//JAzzjiDb775huLi4niEIoQQQgghhBAJ5e6v72bhloUAuCwu5gzfvd7GiWjvvL1xmp0AvLL6FVbUr9A5ooEVtzUlCgoKeOqpp9A0LbYQfE5Oji7dL8Tgk5eXp3cIQvRIclQYneSoMDrJUWF0fc3RF1e+yF1f3RV7/LPSn2E32/srrN02/fTpBH1BLA7LgB7HarKyX8F+fLz5YzQ07vryLp762VMDekw9xX2xR0VRyM3NBSAQCBAMBnG5XPEOQwghhBBCCCEM7cutX3L5+5fHHh9cdDDD0obpFxAwdc7UuB1rr9y9+KbqGzxBD2+sfYMfan9gUs6kuB0/nuLSlf25557j6quv7rTtlltuwe12k56ezgknnEBbW1s8QhGDVHV1td4hCNEjyVFhdJKjwugkR4XR7W6Orm9az5lvnkkgEgBgXNY4Diw8cCBCMyyLycIBBQfEHv/pyz/pGM3Aikth/pe//AWPxxN7/Pnnn3PLLbcwa9Ysrr76aubPn88f//jHeIQihBBCCCGEEIZW463h5NdOpqG9AYAidxFHDT9qUA4DnpI7BbfFDcD89fP5rnrnlb6SQVwK87Vr1zJ58uTY4//+97/k5+fzyiuv8Oc//5lLL72Ul19+OR6hiEFq5MiBmTFSiP4iOSqMTnJUGJ3kqDC63uZos7+Zua/OZV3zOgDSbemcNPokTKppIMMzLLNq7tRT4M4vk3Nd87gU5n6/H7t9+wQF7777LnPmzMFsjg5xHz9+PFu2bIlHKGKQqqio0DsEIXokOSqMTnJUGJ3kqDC63uSoN+jltNdPY1ndMiA6A/upY07VdbK3H3vkvEf4y7F/4ZHzHonbMSflTCLVmgrAexvf4+vKr+N27HiJS2E+fPhw3n//fQAWLVpEeXk5s2fPju2vrq7G7XbHIxQxSPl8Pr1DEKJHkqPC6CRHhdFJjgqj21WO+kN+zp53Nl9VfgWA3WTntDGnkWZPi0d4vRYOhgmHwoSD4bgdczDcNY9LYX7RRRfxwgsvMHnyZH76059SXFzM0UcfHdv/2WefMWHChHiEIgapHXtsCGFEkqPC6CRHhdFJjgqj6ylHvUEvZ7x5Bh9s/AAAi2rhlDGnkOXMild4vZYzLIfcEbnkDMuJ63EnZk8kzRa9SPHx5o/5ouKLuB5/oMVlubTLL78cu93OvHnzmDZtGr/97W9xOBwANDQ0UFVVxcUXXxyPUMQgNWTIEL1DEKJHkqPC6CRHhdFJjgqj6y5H2wJtnPHmGSzcshAAs2LmpNEnke/Oj2d4vXbybSfrclyTamJ64XTeXv82EJ2h/fWTXtclloEQlzvmAL/85S955ZVXeOKJJxg7dmxse2ZmJosWLeKCCy6IVyhiEFq9erXeIQjRI8lRYXSSo8LoJEeF0XWVoy3+Fk5+7eRYUW5RLcwdM5chqXKhqSsTsieQYcsA4NOKT1m4eaHOEfWfuBXmQgghhBBCCCGiNrdsZs5Lc2Jjyq2qlVPGnCJFeQ9URWV60fTY4zu+vANN03SMqP9IYS4GhZyc+I6BEWJ3SY4Ko5McFUYnOSqMbsccXVS1iCOeP4Ky+jIAbCYbp409jaKUIr3CSxjjssaRZY+Ovf+q8qvYDPaJLi5jzIXQ27al+YQwKslRYXSSo8LoJEeF0W3L0ZdXv8zl711Oe7gdgFRrKqeMOYVMR6ae4fXaR49/RHtrO/YUOzPPnxn346uKykHFB7GkZgk3Tb+JSTmT4h7DQJAWTAwKlZWVZGRk6B2GEN2SHBVGJzkqjE5yVBjd+i3rufW7W3lq2VOxbQWuAuaOmYvD7NAxst1T9nEZbfVtuLPcuhTmAGMzx5LtyKY0vVSX4w8EKcyFEEIIIYQQYgCtqFvBBV9fwAbPhti2MZljOLr0aEyqSb/AhGFIYS4GhdLS5LmaJpKT5KgwOslRYXSSo8KIfCEf9y+6n/u/vR9/2A9El0ObOWQmU3OnoiiKzhHuvlP/dCpaWEMxJV7sRha3wryxsZFnn32WdevW0djYuNPseYqi8Pjjj8crHDHIVFdXM3ToUL3DEKJbkqPC6CRHhdFJjgqjeXf9u1y/4Ho2tGyIbcu0Z3L8yOPJdmbrF9geyirO0juEpBSXwvydd95h7ty5eDweUlNTuxz/k4hXi/rb5pbNfF7xebf7d3WOFLrf39P39vR9Pe+Kfzx9/b4tW7ZQEirp2zHj/HPs6fcaKZ6+fp9JMWFWzZhVMxbVgkW1YFJNWFRLbPu2fdu+tqrWhO4K5vF49A5BiB5JjgqjkxwVRqBpGp9u+ZS/fPMXPtnySWy7gsKs3FmMHzIesyqdlsXO4pIV1157Lfn5+fzvf/9j0qTkmDVvIHxX8x2XvHeJ3mEkr6V6ByAGmlW14rA4cJqdOMwOHBYHDnP0sdPixG11k25LJ82WRrotnXR7Oum2dDLsGeQ6c8l35eO2uvWJ3WrV5bgi8UW0CG2BNloCLTT7m2nxd3wOtOAJegiEA/jDfoKRIIFwIPYRIYKKiqIoqIqKqqgoRL+2mWzYzfbo35LFgd1sp7mxmdrNtaTb0sm0Z5Jhz8BlccmFdWEY0o72H03T8Ia8eINePEEPnqCHYCRIRIsQjoSJECESiRDRoh+xC+YmS+yiuVW1YjFZcFvcuCwuLCaL3j/WgAqGg7y74V0eWPwAX1d+3WlfgauAOcPnsL9tf9aqa3WKUBhdXArz8vJy7r77binKhRADKhAJEPAHaPY39/k13BY3Be4C8px55LvzKXAVMDx9OKVppZSml1LoLkRV1H6MOkrGRoquhCNhKtoqWN+8nvXN69nSsoUqbxXVnurYR52vDg1t1y/WH77t/NCqWsl0RIv0HEcOhe5CilKKoh/uIopTiilNK8VmtsUnPjGoSTvasxZ/C5tbN7O5ZTNbWrdQ7Y22H/W++tjnZn8znqAHb9Db7+2KzWTDZXHhtkYLdZfFhdviJs2WRqYjk0z7Dh8d7UqWPYtMRyap1lRDXgTUNI0ltUt4fuXzvLzqZep8dZ32p1pTObjoYCZkT0BRFNZp63SKtH9VlFUQDoYxWUwUjZN11/tLXArzUaNG0draGo9DJbRJ2ZO4+9C7eW/DezuNwd+TxvHHr9Xr79uTY/b0vT3uGphYcyO51Kg1homnx++N8++rr8fbk2P2+JqahoZGWAtvvxpPxxV6LRLd3nGFftvX2z6HIiGC4SAhLUQoEv0Ia+HdOn5bsI01jWtY07imy/02k43hacOZkD2BSTmToh/Zk/Z4rNjKlSsZP378Hr2GSFwRLcKmlk0sq1vGstplLKtbxuqG1Wxs2UgwEtQ7vG4FIgGqPFVUeaooo6zL56iKyrDUYYzKGMWozFGMzhjN1NypjM0aK905Rb+SdnR7W7KyfiUrG1aysn4lqxpWsb55PS2BFl1j84f9+MN+Gtobdvt7LaqFbEc2WY4ssh3ZZDujX+c4cjpty3ZEPwaqkA9HwqxtWsu31d/y8aaP+XjTx9T6and6XrotnemF0xmfPb7TxfyxkbGUmbpuKxPJa3e8Flsu7VfP/ErvcJJGXP4j3n777Vx66aWcfvrpDBs2LB6HTEjD04dzfvr51PvqiWgRvcNJKuPC45KiIRS7T9M0gpEgwUgw+qYg5Kc91E57uL3T59ZAK62B1k5d9rriD/ujb3YaVvLy6pdj2wtcBexfuD8HFR3E9KLpjM0ca8ir+8IYvEEv31Z9y+dbP+eLii9YXL2YtmDbbr2GghIdqmFxYjfbsZvsnT7bTDasJismxYRJNUU/7/C1oiixC2Gw/aKYhha7sBWKhAhGgoQiITJCGVRoFfhCPrwhL76gD1/IF/sb6u4iWESLsK55Heua1/HOhndi2x1mB5NzJrNX3l5My5vG9KLpFLgL+n5ShRiEmv3NfFP5Dd9UfcPXlV/zbdW3u92WbGNRLVhN1s6fO7qjb2tLdhz2sm0YDLDzBfNImLAWJhQJEYgECIajQ2mCkSCBSIBQOERIC/U6tmAkSKWnkkpPZa9/ll0V8um2dExqdE6bbXPbbOvC7wl6aAu0Ue2tprKtksq2SsqbyimrL8MX8nV5TJNiYmjqUPbK3YvS9FJ5DyB2W1wK8w8++ICcnBzGjRvHkUceSUlJCSZT50maFEXh/vvvj0c4YhCqV+r1DkHoRFEUrCYrVpMVl8XV6+/zh/14Ah5aAi00+htpbI9+NPgaaA4073TxrNJTyatrXuXVNa8CkGXPYnrRdA4behizhs8i35Xf4/GysmSG02QW0SJ8V/0d72x4h082f8J31d/t8k64WTGTYkuJjenOtGeSZksjxZqCy+LCYXbE9Y1fjz2PiP7NtPpbaQm00BqIfm5qb6LOV0ejv5FQpPObcF/Ix1eVX/FV5VexbaMyRjGjZAaHDTmMGUNm4DA7BuznEclnMLSj7aF2vtz6Je9vfJ+PNn3EyvqVu+y9pqDgsrhItaWSZksjw5ZBmi0Nt9Udm4PFYXbEvQdLRIsQDAdpD7fjC/liF/t2/Ng2zt0b9Ma29aa33u4W8n1lUS0UuAoYkzmGcVnjsJvtPT4/Wd6P7n303vi9fmxOGabUnxRtT/qx9pKq7no8pqIohMO71+V0R3feeSe/+93vuPLKK7nvvvt22h8KhViwYAFTpkzZ6aKA0fz5qz/LHfN+lhZJo1nt+7hjIXYUjoSpb6+nxlNDtbdjrK+3usdCa++8vTmq9ChOGn0SQ9N2Xs6nqamJ9PT0AYxaxFsoEuKTzZ/w6ppXeXfDu9R4uy9qXRYX2Y5sClwF5LnyyHXmkm5LN9Qdlz1pRzVNoy3YRr2vnhpvDVWeKra2be2xa63T7OSwoYdxVOlRHFV6FKm21L6GLgaJZG1HWwOtvLP+HV4vf50PN36IN+Tt9rlOs5NcZy55zjxynDlkO7PJtGcmzbARTdPwh/2xgt0X8nWaoG7bhy/koz3U3utCvrdSramxtnpI2hAKXYW7tSKMvB/tX3W+Os6acBZTcqfoHUqPwuEwS5YsYcaMGZjN3f8txuWvNBIZ2CLzm2++4ZFHHmHy5MkDehyRuAq1QpqRhlD0D5NqIteZS64zl4lMBKJX/qs91bGJdTa3biYQCcS+Z3H1YhZXL+b2L25n3/x9OXnMycwdM5d0ezoAW7duTco3lIPR9zXf81zZc7yy+pUuxx4CpNnSKHYXMyxtGMUpxaTZ0uIc5e7bk3ZUURRSrCmkWFMYljYstt0b9FLlqaKirYINzRuo8lTF3kR7Q17eXPsmb659E4fZwVGlR3HK2FOYOWRm0hQZon8lUzsaioT4cOOH/HvFv3lvw3v4w/6dnqOgkOXIothdTElqCUXuIlKsKYa6oNffFEWJDtkx28m0Z+7y+d0V8t6Ql/ZQe6wLfkSLROfcUbZ36bearLjM0cnqtrVfVtOezfwv70dFTxL+P1tbWxtnnHEGjz76KLfffrve4QghBilVUSlwF1DgLmC/gv1ihXp5UznljeWdCrRvqqLjAW/+7GZOHH0i5006DxvSHSyReYNe/rf6fzy57EkWVy/eab9ZMVOcUsyYzDGUppeSYk3RIUrjcVqclKZHVzw4pPgQ/GE/m1o2Ud5YTnlTeWwspy/k4+XVL/Py6pcpdBdy3qTzOGvCWeQ4c3T+CYToXxWtFTy57En+u+K/XXbDtpvtsckUh6UNk+Eeu7C7hbwQeopLV/Zt1q9fz9tvv83GjRsBGDp0KHPmzGH48OF9fs2zzz6bzMxM7r33Xg499FCmTp3aY1f20tLSTl3rbTYbNpux3hBLV/b+Z9fstCvteochBrFmfzNl9WUsr19OvW/nMWbTcqdxzX7XMGv4rAFZjk0MjDpvHf9c+k8eW/IYTf6mTvu2TQQ0KXsSpRmlWNTEXsM33u1oRIuwtW0rZfVllNWX0R7ufGyrauWkMSdxxbQrGJM5Jm5xCePy+Xw4HIlZqC6tWcrfv/s7r6x5Zaf5GOxmO6PSRzEhewLFKcXyPyKByfvR/iVd2fvo2muv5f7779+pW7uqqlx11VXcc889u/2azz33HIsXL+abb77p9fdMnDgRr3f72Jxzzz2Xyy+/nIKCAtauXQtAXl4emqZRUxMdDzhq1Ci2bNmCz+fDbrdTUlLCmjXRpZRyc3NRVZWqqioARowYQVVVFR6PB5vNxrBhw1i1ahUA2dnZWK1Wtm7dCsDw4cOpra2lra0Ni8XCyJEjKSsrY2xoLPVKPT7FR1EkujbgBnUDmVomqVoqYcKsNq1mbHgsCgpNShOtSislkRIANqmbSNVSSdfS0dBYaVrJ6PBoTJhoUVpomDxzyQAAPEVJREFUVBoZGomOcd2ibsGpOcnUolcRy0xljAqPwoyZVqWVOqWO4ZHohZOtylZs2MjSopOrrFRXUhopxYoVDx6q1WpKI9E1RCuVSsyYydGidzNWq6sZEhmCHTs+fFSoFYyMjASgWqmOnnctD4BytZyiSBEOHLTTziZ1E6MjowGoVWoJEaJAi87cu05dR14kDxcuAgRYp65jbGQsEJ1gw4+fQq0Qh+ZghWkF2Vo2KVoKIUKsMa1hXHgcAA1KA17FS3GkGICN6kYytIxuz3eL0sKQyBAANqubSdFSuj3fDUoDwyLDAKhQK3Boju3nWy1jZGQkFiy0KW3UKrWdzrcVK9ladBmuVeoqhkWGYcOGBw9VahUjIiMAqFKqUFHJ1XIBWKOuoSRSEjvfW9QtjIqMAqBGqUFDi53vtepaCiIFOHHix89GdWOP5ztXy8Wtubs83+1Ke5c5++Pz3ag00qa0dcrZNC2NNC2NCBFWmVYxJjwGFZVmpZlmpbnT+XZrbjK0jJ1ytqvzbdfsXeZsm9JGjVLTY84OjQzFhg0vXirVytj5rlaqUVA6ne/iSHEsZzermzud7wgRxpnHcUDeAawtWEu4Lcw3dd/waf2ntEei/6C/rfmWM948g9Hpozmt6DQOzz+ckaUju2wjADIzM3E4HFRUVAAwbNgwGhoaaGlpwWQyMWbMGMrKytA0jfT0dFJSUti8eTMAQ4YMoaWlhaamJhRFYdy4caxatYpwOExqaioZGRmxC6jFxcV4vV4aGqLL24wfP57Vq1cTCoVISUkhOzub9evXA1BYWIjf76e+PnrhYezYsaxbt45AIIDL5SIvL49166JruBYUFBAKhaitjfYkGD16NJs2baK9vR2Hw0FRURHl5eXRtiEvmqvV1dG2YuTIkVRUVMTa5CFDhrB69WoAcnJyMJvNVFZG7zSVlpZSXV2Nx+PBarVSWlrKypUrgehEUTabrVObXFdXR2trK2azmdGjR7NixYrY+XY6nWzZsgUAa7aV+765j5fXv4w/0rmLaYGjgGk505iTOQenybm9jQgndhuRpWXRpDTFtY2Y4JzAkc4jWVeyjq1NW1lUu4ilLUvR0AhEAjxb9izPlT3HUcOO4uS8kxmZMpKSkhLa2tpobGzcKWdTU1PJzMxkw4YNABQVFdHe3t5lzrrdbnJzc3vM2Y0bN+L3+3E6nYZ6H7EtZwdbG7FhwwYcDoch2oihQ4fS2NjY7flOTU1l06ZNLGtexn8q/sPCrQvZkYrKpLRJ7Ju9LwUZBZRq0f9VFVTgiBizjZD3Ebt+HzE+PB6f4uvT+4h8LT92vvMj+bhw4cfPBnUDYyLRi5N1Sh0BAhRqhQCsV9eTo+Xg1twECVKuljMusv29b19rjd9f/Xs8jR4yMjK46d6bdKs1bIoNb4WXFXUrDP0+orf3weNyx/wvf/kLv/nNb5g7dy7XXnst48ZFE6KsrIx7772XF198kXvuuYerr76616+5efNm9tlnH957773Y2HK5Yy66I8ulCSMKhAOU1ZfxbfW31PnqOu0blzWOm6ffzJHDjkzq8YKJpqm9ifu/vZ9Hvn+k0x1cBYVRGaPYv2B/8l35Sfk7M0o72uRv4vvq71lSu2SncbfHjzqem6bf1Gkcuxg8VqxYkTDrmC+tWcqfvvxTpyUEAWwmG1NypjAtf5oMeUlCRmlH99RDZz1kiHXMk+2OeVwK87FjxzJ27FheffXVLvcff/zxrFy5Mnb1oTdeffVVTjjhhE4zrIfD4eiaiqqK3+/vtE9mZR/cRoVHsca0Ru8whOiSpmn4Gnz8r/p/bG3b2mnf9MLp3P6T25maO1Wf4AQAwXCQR5Y8wl+++QvN/u0T95gVMxNzJnJAwQFJP2u40drRQDjA9zXf83Xl151mqbaoFi6YfAG/3u/XZNgzdIxQxNvq1asZPXq03mH0qNpTza2f38qzZc922p5iTWG//P2YnDMZiymxh72I7hmtHe2rp654Ck+jB1eGi7P/drZucUhh3gd2u517772XSy65pMv9//jHP7j66qtpb+/9mIvW1tZYN6ptzj33XMaOHctvf/tbJk6c2GmfFOZCCKPTNI1NrZv4ZPMnnSb9UVA4b9J5/N/0/0uI2buTzadbPuU3H/+GVQ2rYttURWVy9mQOLj4Yp8WpY3QiGAmytGYpn2/9PDZZHECmPZPbDrmN08aelpQ9GERiCYQDPLLkEe7+6m7agm2x7S6Li+mF05mcM3m3lt0SQiRfYR6XMea5ubksWbKk2/1LliwhJ2f3ZlZNSUnZqfh2uVxkZWXttF2IZOk6JJLXthwdmjqUM8efyerG1Xyy5RMa2xvR0Hj8h8d5vfx1bj/kduaOmSuFRhzUeev4/cLf8+KqFzttH5M5hpklM5P+DvmPGbUdtagWpuVPY2LORL6u/JqvK78mrIVpaG/g0vcu5dkVz3LPzHsYnWnsO6lizxm1K/uiqkVc/v7lnS7uWU1WDiw4kGn502T5v0HEqO2oMIa4TOt48skn89hjj3HnnXfi8Xhi2z0eD3fddRePPfYYp556ajxCEUIIw1MUhTGZYzhv4nkcWnJobDbvWl8tF717Eb946xfUerteH1v0j7fWvsX0/0zvVJTnOHI4a/xZHDfyuEFXlCcCm8nGIcWHcOGUCxmbOTa2/dOKT/nJf3/C3779G+FIWMcIxWDjDXq5ceGNzH5xdqeifELWBC6cfCH7F+4vRbkQIiYuXdm9Xi/HHHMMH330EWazmcLC6EyBW7duJRQKMXPmTN544w2czoHrDihd2Qe3vEge1Wq13mEI0a2ecrTF38KHmz5kdePq2LYsexZ/PeyvHDPymHiFOCg0+5v53YLf8dzK52LbrCYrPyn+CVNzpw7qZYoSrR1d17SO9za+12lOgAMLD+ShIx9iaNpQHSMTA6Wqqor8/Hy9wwDgu+rvuPCdC1nbtDa2LduRzVHDjyLfbYwYRfwlWjtqdNKVvQ+cTicffPABr732Wqd1zGfPns1RRx3FMcccI90yxYDyKt5dP0kIHfWUo6m2VI4fdTyrG1bzzoZ38IV81LfXc/a8szlrwlncOeNOHObEXLvXSL6t+pbz3j6Pza2bY9uGpQ7jqNKjcFvdOkZmDInWjpaml3Jeynl8VvEZX1d9DcAXW7/g4P8ezD2H3sOp46SnXrIZyBs8vRXRIjy4+EFu/+L22HrkJsXEAYUHcGDhgYP64p5IvHa0O18+/yV+rx+b08YBpx6gdzhJI679Z4477jiOO+64eB5SCACKI8UypkcYWm9ydHTmaIpSinh3w7usaYzO6vrM8mdYXL2YJ+Y8wciMkfEINelomsbD3z/MHz77A8FIEIiOWz5syGFMzpksF447JGI7ajFZOHTIoYzIGMG8tfNoDjTjCXq45L1L+GLrF9w5407sZrveYYp+smXLFl3HmNd4a7j4nYv5ePPHsW05zhyOKT2GbGe2bnEJ40jEdrQri99cHFsuTQrz/iOX7YQQIoG4LC6OH3k8c4bPiY1NXF63nJnPzeTVNa/qG1wCag20cva8s/n9wt/HivI8Zx7nTzqfKblTpChPEiUpJZwz6RwmZm+fHPbp5U8z64VZrG9ar2NkIlksqlrEzGdndirKp+VN4xfjfyFFuRCiVwbkjvnw4cNRVZWVK1disVgYPnz4Lt/cKIrC2rVre3yOEH21Ud246ycJoaPdyVFFUZiUM4kCVwGvlb9GfXs9nqCH894+j7L6Mn67/2+lu2QvbGjewOlvnM7KhpWxbXvn7s3MITNl2aIuJHo7ajPZOKr0KIakDuHdDe8SioT4oe4HZj43k8fnPM7hQw/XO0Sxh4YO1WfugGeWP8NvPvoNgUgAAIfZwdGlRzM8fbgu8QjjSvR2dJvjbjiOcDCMySL/K/vTgBTmM2bMQFEUVFXt9FgIvWRoGUkzrkckp77kaLYzm7MmnMW7G95lRf0KAO7++m5W1q/koZ8+hMviGohQk8KnWz7lnHnn0NDeAIBVtXJU6VGypFYPkqUdnZg9kTxnHq+Vv0ZDewMtgRZOff1Ubj/kdi6acpG8X0lgjY2NuFzxa/eC4SC/++R3/OuHf8W25bvyOWn0SdL+ii4lSztaNK5I7xCS0oAU5k8++WSPj4WIt1QtlQoq9A5DiG71NUetJis/K/0Zuc7cWBfKN9a+wcaXNvLsMc9S4C7o30CTwL+X/5trPromNjFTmi2Nk0efTKYjU+fIjC2Z2tEcZw5nTTiLeevmsaZxDREtwg2f3MDK+pX8+dA/YzVZ9Q5R9EFLS0v8juVv4dy3z+WjTR/Ftk3KmcRPh/5UetyIbiVTOyr6X1z6Oj799NNs2LCh2/0bN27k6aefjkcoYpAKI2vXCmPbkxxVFIX9CvbjpNEnYVWjBcXS2qXMfnF2bJI4EZ3k7Z6v7+GKD66IFeUlKSWcPeFsKcp7IdnaUZvJxvEjj+eAgu0TFz29/GlOe/00WgOtOkYm+ipey+Fuad3CUS8dFSvKVUXlp8N+ypzhc6QoFz1KtnZU9K+4rGNuMpl45plnOP3007vc//zzz3P66acTDg9csso65kKIwaDOV8fLq16mORBduznTnsmzxzzLvgX76hyZvsKRMNctuI4nfngitm1SziRmDZsl4/EFy+uWM3/9fMJa9H3IlJwpPHfsc+S58nSObOCFI2HK6sv4cuuXLK5ezLrmdVS2VdLkbyIQjo6ZdlvcZDmyGJUxijGZYxiTOYb9C/dnSOoQnaOPv2W1yzjl9VOo8lQB0Qs8J4w6YVCeCzF41W+pRwtrKCaFrOIs3eKQdcz7YFe1v8fj6TFIIfbU2PBYVppW7vqJQuikv3I025HNGePP4KXVL1HjraGhvYHjXzmeJ496kiOHHdkPkSYef8jPL9/5JW+ufTO27eCigzmw8EAZT7wbkrkdnZA9gTRbGv9b/T/aw+0sqV3C7Bdn89LxLzEifYTe4fW79lA77254l5dWvcSCzQt22UPAH/ZT317P6sbVvLXurdj2sZljmTV8FnNK57Bv/r66/z2VlZUxbty4AXv9ryq/4rTXT6PZH73wmWpN5eQxJ5Pl0K8wEYklWdrR53/3fGy5tF898yu9w0kaA1YNL126lO+//z72eOHChYRCoZ2e19TUxMMPP8zo0TLhjhg4CvLmWxhbf+ao2+rm5+N+zitrXmFTyyZ8IR9nvnkmj85+lGNHHttvx0kEvpCPX7z1Cz7Y+AEQPc+zhs9ics5knSNLPMnejhanFHP6+NN5cdWLtAZa2diykTkvzuF/x/+PiTkTd/0CCWBVwyoe/v5hXl3zaqy47IrdZMdmsmFWzWhoBMNBvCFvrEfBNisbVrKyYSX3f3s/47PGc8HkC5g7Zi5uq3ugf5QuDWQn0I82fcRZb56FNxSduCvXmcspY07BaXEO2DFF8kn2dlTsmQErzF955RVuueUWIDr+8ZFHHuGRRx7p8rnp6ekyxlwMqCalSe8QhOhRf+eozWRj7ui5vLXuLVY1rCIYCXL+2+fzj5/+g7lj5vbrsYyqLdDGGW+ewcItCwEwKSaOH3k8IzKS7w5oPAyGdjTbkc2Z48/kpVUvUeurpc5Xx7H/O5aXj3+ZvfL20ju8Pvu+5nv++s1feWvtW2h0Ll7tJjtF7iKGpA6hOKWYLEdWl5PfRbQILf4W6nx11HhrWNu0lkpPZWz/ivoVXPPRNfzhsz/wyym/5PK9LyfVljrgP9uO0tPTB+R13yh/g1/O/2VsObQidxEnjzlZJgkUuy1Z2tFxh46jvbUde4pd71CSyoCNMa+srGTr1q1omsZ+++3Hrbfeypw5czofXFFwuVyMGDFiwLuyyxjzwc2lufAoHr3DEKJbA5WjES3C/PXzWVa3DIherb//8Ps5c8KZ/X4sI2nxt3DK66fwdeXXAJhVM3NHz5VxoHtgMLWj7aF2Xlr1Els9WwFIsabwwnEvsH/B/jpHtnvWN63nxk9vZN66eZ22m1UzI9JHMCVnCkNSh/R5ngVv0MvaprUsqVkSO1fbZNmzuGbfazhv0nnYzLY+/wy7o62tDbe7f+/W/3fFf7nigyti78uGpw3nhFEnYFZlCKbYfYOpHY0HGWPeSwUFBRQURJfp+eijjxg3bhy5ubkDdTghejQkMoQyU5neYQjRrYHKUVVRmTN8DmbVzPc136OhccUHV2BSTfx83M/7/XhG0BZo49TXT40V5VbVyiljTqEwpVDnyBLbYGpH7WY7J489mZdXv8yW1i20Blo56dWTeO6Y5zi4+GC9w9slT9DDfYvu48HFD+IP+2PbnWYn++bvy155e/XL3V6nxcmknElMyplElaeK76q/Y3n9ciJahPr2en6/8Pc8uuRR/nrYXzl0yKF7fLxd2bRpE+PHj++313vihye49qNrY4/HZI7hmBHHyISRos8GUzsqdl9cWpYZM2ZIUS6EEDpRFIUjhx7JtLxpsW2Xv385L69+WceoBoY36OX0N07nq8qvgGiX/tPGnSZFudhtNpONk0efzLDUYUA0t0557ZTYfAVG9fGmjznw3wfyl2/+EivKnWYnhw85nIunXsz+hfsPSBfsfFc+c0rncMGkCxiXtX0Ctg0tGzjx1RO59L1LafA19PtxB8qTy57sVJRPypnEsSOOlaJcCDFgBuSO+cyZM1FVlXfeeQez2cxhhx22y+9RFIUPPjD2PzuRuDarm/UOQYgeDXSOKorCYUMOQ0NjcfViIlqEi9+5GJvJxtEjjh7QY8dLe6ids946i08rPgWid8pPHXMq+a58nSNLDoOxHbWYLJw4+kReK3+NtU1raQ+3c8YbZ/CfY/7D4UMP1zu8TrxBL7d8dguPLn00tk1VVPbK3YuDiw/GZopPd/J0ezrHjDiG/fL348NNH7K5NZo3z5Y9y/sb3uf+/2/v3qOjKNPEj3+rO+nuXDoh9ysh5B5AYQVB11EQEBwEJgQIIILgDDgOIA47sqO/HWHOzupcvI3KqGfHwXVE1AEhgoJyEVjUKIKiGAIEhEDugdzv6arfH700EyHhYtJV6Tyfc3KOVV1d9eT1yUs/XW+975g/c2fCnd1y7b59+3bJeV479BrLdi5zbd8QfgNj+o3RfdZ50fP1xn5UXLlu+dpP0zRU9cIz0qqqomlapz//fLwQXc2u2fUOQYhOuSNHFUVhTNwY14zkDs3BT7f8lG0nt3X7tbtbi6OF+7bcx0cFHwHgbfImKzWLSH8pyrtKb+1HvUxeZCRlkBqUCkCL2sKczXPYc3qPzpFd8FXZV4xaO6pdUR7tH838QfMZ02+M24ryfxbhF8HMtJmMjx/vun55Yzl3b76bR/c8SnNb82XOcPVqaztf9u1KrMldwy93/tK1LUW56Eqe0o/+4zf/4NUlr/KP3/xD71A8SrfcMd+1a1en20K4Wx+tD8UUX/5AIXTirhxVFIXx8eNxqA6+PfstrWorc9+by9pJa93yDGh3aFPbWLB1AVu/2wpcmOhNhq93rd7cj5pNZiYlTULL1zhaeZQmRxOzNs1i3U/WcXPMzbrFpWkarx56lUd2P+KaMdysmLk19lZDrCuuKAqDwweT0CeBD09+yPGq4wC89NVLfFr4KX/98V+7dJ34qqoqoqOv/e9+7eG1PLj9QdfM9UPChkhRLrqUp/Sj5SfLXeuYi64jD8qIXuH7y8MIYTTuzFFFUfhxwo9JDXbeAWx2NDN782z2Fe9zWwxdRdM0frnzl2w6vglwFiWZyZn0DeiaIa3igt7ej5oUE5MSJ5HUJwmAxrZGst7N0u3vpr61ngc+fIB/++jfXEV5mE8Y8wbOY3jUcEMVk3aLnczkTMb2G4tZca6Mc7D8IGPeHNOlI3Z+yO+87sg6Fm9b7MrzwWGDuSP+DkO1o+j5PKUfNXubMXuZMXsbe6WrnsYthXlBQQF79+5tt+/gwYPMnTuXGTNmsHHjRneEIXqxPHOe3iEI0Sl356hJMTExYWK7ImPmuzM5cu6IW+P4oR7PeZw1uWsA5++UkZRBfGC8vkF5KOlHnXfOJydNJiEwAXAWx9Ozp/Nl6ZdujeN0zWnGvz2et4+87do3OGwwcwfOJcQ3xK2xXClFUbgh4gbuGXAPQbYgAGpaapj57kye2/8cXbF6b3p6+uUPuoStJ7bywIcPuIqm68KuY1z8OCnKRZfzlH70/r/dz7+9+2/c/7f79Q7Fo7ilMH/wwQdZuXKla7u0tJTbb7+dd955hz179jB16lTeeecdd4QieqkUR4reIQjRKT1y9HyRcX5t78rmSqZtnEZhbaHbY7kWq79ZzVP7nnJtT+g/gcSgrhsWK9qTftTJy+TFT5J/Qr+AfoCzuJy6cSrfVnzrlut/UfIFY98aS+7ZXMA5n8LEhImM7z8es8n4d68i/CKYO3AuyUHJgPMO4sqPV7Lwg4U0tTX9oHMfOXL1XyzuOb2H+Vvm49AcAAwIGcCd8XdKUS66hfSjojNuKcw///xz7rjjDtf2a6+9RmNjIwcPHqSwsJAxY8bw5JNPuiMU0UuZMf6HFdG76ZWjXiYvpiRPIdzXuaRlYV0h07OnU9VUpUs8V+q94+/x8K6HXduj+o5iQGjXrV8sLib96AXeJm8ykzOJtccCUNVcReaGTI5VHuvW6248tpHJ6ydT3lgOQIAlgDkD5vS43LearWQkZXBLzC2ufeuPridzYyaVTZXXfF6Hw3FVx39R8gWzN892LSuXHJTMXQl3SVEuuo30o6IzbinMz507124d882bNzNy5EgSExMxmUxkZmaSl+cZQzuEMdUoNXqHIESn9MxRq9nKtJRp9LH2ASDvXB6zNs2isa1Rt5g6k1OUw4KtC1A152oeQ8OHMjxquM5ReT7pR9vzNnszNWUq0X7OycbKG8vJeCeDU9WnuvxamqbxzL5nuG/LfTQ5nHeVo/yiuHfQvYT6hnb59dxBURRuibmFKclT8DZ5A86/7Tv/cScFNQXXdM6AgIArPjb3bC5Z2VnUt9YD0C+gH5MTJ0tRLrqV9KOiM24pzMPCwjh1yvkPVVVVFTk5OYwfP971eltbG21tbe4IRfRS55RzeocgRKf0zlF/iz/TU6fj6+ULwGfFn/GzLT+jTTVW33zk3BHu3nS3qzhJCUphdL/ROkfVO+ido0ZkNVuZljrNNeKkuL6YjA0ZFNUVddk1Wh2tLNmxhP/89D9d+9KC07g7/W58vHy67Dp6SQ5KZlb6LFffc6zyGOPfHs/XZV9f9bmCg4Ov6LgTVSeYumEqVc1VgHNpuakpU3vEowCiZ/OUfvSrLV+x7519fLXlK71D8ShuKczHjh3Lc889x9NPP83cuXNRVZWMjAzX67m5ufTtKzPoiu4Tr8brHYIQnTJCjgbZgpiWOs1192rLd1tYtnNZl0zK1BWK64qdw+z/78N0rD2WSYmT5A6XmxghR43I5mUjKzWLEJtz0rVTNaeYsmEK5Q3lP/jcDa0NzHlvDm/kvuHa968x/8qkxEkeVURG+kVyz8B7CLY5C+vShlLuWn8XO0/tvKrznDx58rLHFNYWMmXDFEobSgHnTPbTU6fjZeqWFYSFaMdT+tFP3viEj/76EZ+88YneoXgUtxTmv//970lPT+dXv/oVH374IU8++ST9+/cHoLm5mbfffpsxY8Zc1TmfeOIJbrzxRux2O+Hh4WRkZFzTpB9CCCEuiPSLJDM5E5Pi/Ofh9dzXeTzncZ2jgprmGrLezeJM7RkAQn1C5Q6XMAxfb1+y0rIItAYCzru+UzdO/UHPS1c3VzMtexofnvwQcC4FODFhIj+K+ZFHfhnVx9qH2emzifGPAZwz3s/cNJP1R9Z32TUqGirI3JjJ6drTgPPLyJnpM7GarV12DSGEuFZuKcwjIiL4+OOPqayspKamhqVLl7peU1WVHTt2tJu1/Urs3r2bRYsWkZOTw7Zt22htbWXcuHHU19d3cfTCExSaesYs06L3MlKO9gvsx8SEia7tp/Y9xepvVusWT3NbM3Pem+Oa9drf25+s1Cz5MO1mRspRI7Jb7MxMm4nd2w7AoYpDTM+eTm1L7VWfq7S+lEnrJ5FTlAM4J5ubljKtx03ydrV8vH3ISstyzdjeprax8IOFvHro1St6f0xMTIevnf9y7/wEfXaLnVlpszzicQDRc3hKPzpu8TgmPzKZcYvH6R2KR3FLYX5eYGAgFoul3T4fHx8GDx58xc8Fnbd161bmzZvHwIEDGTx4MK+++ioFBQXs37+/K0MWHsJHk394hbEZLUfTQtIYE3dhJNPDux7m/ePvuz0OVVNZtH0R/3vmfwHnM70z02bib/F3eyy9ndFy1IgCrYHMSJvhel76QOkBZr07i4bWhis+x6nqU0xYN4FDFYcAsJltzEqbRb/Aft0Ss9F4m7z5SdJPGBw2GHAup7Zs5zKeP/D8Zd/b2HjpCSsb2xq5e9PdfFX2FQC+Xr7SjwhdeEo/mjQiibRb00gakaR3KB6lWx6oee211wCYM2cOiqK4ti9n7ty513zN6upq4PITf9TW1mIyXfg+wmq1YrXKXRdPF6wFU0qp3mEI0SEj5ujQyKHUttbyefHnqJrKz7b+jI2ZG906A/pjex/jnaPvAM6hvNNSphHsc3Vf5IquYcQcNaJgn2BmpM1g7eG1NDma+KToE+a+N5c1E9dg9er880ZuRS7TsqdRUl8CgJ+3HzPTZhLiE+KO0A3DpJgYFz8Oi9nCvpJ9AKzYu4LalloeGfFIh0P5z507R2RkZLt9rY5W7nv/Pj4pcj4LazVbmZE2gyBbUPf+EkJcgvSjojOK1g2z+phMJhRFobGxEYvF0q4Q7jAQRbnq9SfPU1WVyZMnU1VVxd69ey95TFtbG7t372by5Mk0NFz45nr+/PksWbKEqKgojh8/DjiH3muaRllZGQDJycmcOXOGxsZGbDYbffv25dgx51Co8PBwTCYTJSXOf0QTExMpKSmhvr4eq9VKfHy869n30NBQLBYLRUXO2Vr79+9PeXk5dXV1eHt7k5SUxOHDh8mvzOescpZGpZEY1Tks66TpJMFaMAFaAA4cHDUfJc2RhoJClVJFrVJLX9U5gV6BqYAALYA+Wh80NPLMeaQ4UjBjpkapoVKppJ/q/Ob9jOkMvpovwZrzg+5h82GSHcl44UWtUkuFUkF/1TkfQJFShBUrIZrzA0KeKY8ENQELFuqpp9RUSoKaAECxUowXXoRpYQAcNR0lTo3Dho1GGik0FZKkOr9lK1WcHVSEFgFAvimfGDUGH3xoookCUwEpagoA5Uo5bbQRpUUBcMJ0ggg1Aj/8aKGFE6YTpKlpAJxVztJMM9FaNHbNztfmrwnVQrFrdtpo45j5GOmOdMA5S2aD0kCs6lyT9pTpFEFaUIftXaPUEKfGAXDadBq7Zu+wvc8p51yTfRSaCvHRfC60t+kwSWoS3nhTp9RRrpS3a28LFkI151I4R0xHiFfjsWKlnnpKTCUkqokAlCglmDARrjlnBj5mOkZfta+rvc+YzpCsOocGlillaGiu9j5uOk6UGoUvvjTTzCnTqU7bO1wLx1/zv2R7NylNl8zZ77d3pVJJnVLXLmcDtUACtUBUVI6Yj5DqSMWEiWqlmmqlul17+2v+BGlBF+XspdrbptkumbN1Sh1lSlmnOdtP7YcVKw00UGwqdrV3qVKKgtKuvWPVWFfOnjadbtfeKiqRWqSrvSPVSPzwo5lmTppOkqqmYtfsfGf6jhZaiNacyy99Z/qOMC0Mf82fVlrJN+WTrl7IWXf0EQVKAdknsjlw7gDgfCbzpWEvEWOLwW63ExoaynfffQdAdHQ0zc3NnD17FoC0tDROnDhBS0sLfn5+REREcOLECQCioqJoa2ujvNw5QVZKSgoFBQU0NTXh4+NDTEwMv9vxO1449gIACgoPJjzIkKAh3dJHnG9v6SM67iMitAgqlArpI66wjzhTf4a/HvkrTapzBYFRkaNYOWAl/r7+l/wcsS1vGw9/9TB1bXXOvxFbFA8mP0i1rZpUNRWACqXCcH1Ed36OMGtmcopz2FC0gfOm953O8sHL6Rvbl/z8fMD5uQ0gPz8fu91OUlIShYWF1DfU8/jhx9latBUAq8nKgpQFRPtHSx8hnyN06SOGOIZQq9R2y+cI6H19RGNLIzcF3kSgNfCSnyO+30eUljprjvN9xPnaLi4ujqNHjwLO1cS8vLwoLi4GICEhgdLSUurr67FYLCQkJLiW+Q4JCcFqtbar7SoqKqitrcXLy4uUlBRyc3PRNI3m5mZGjhyJl1fH98W7pTA/vzRav3792m1fzvnjr9YDDzzAli1b2Lt3L7GxsZc85nxhnpCQYPg75n/87I+u9XlFF9EAz5srR3gSA+eoQ3Ww7ug6TtU4+/K4gDg+mP4BEX4R3XbN9UfWs+CDBa7tO/rdwb9E/Eu3XU9cAQPnqFGdqT3D20fedi07OC11Gi/e8eJFkxZuO7mNee/Po7HNORQ73DecGakz8PH2jGGvP9T+kv3sKNjh2p49YDbPjn72onbUNM11N13TNJbvWs4r37wCOEfcZCZn0r9Pf/cFLsT3eUg/2ljT6Pp78wnQr5+qaKxgzsA5DA4frFsMV8LhcHDw4MHLFubdMpT9+wX2tRbcV2Lx4sVs3ryZPXv2dFiU/zO73Y7ZLLP49jZJahL55ny9wxCiQ0bOUbPJTEZyBmsPr6WsoYyCmgJmvDuDTVM3YbfYu/x6e07v4RfbfuHaHhE1QopyAzByjhpVrD2WzORM1h9dj0NzsO7IOny9fHlm9DOuAnL9kfU8sO0BV/Ee6x/LtNRpWMyWzk7dqwyNHIrFbGHrd1vR0FiTu4a6ljpeHv9yu3bKz88nOdl5l/HxnMddRbmCwsTEiVKUC915Sj+6etFq6s7W4R/izy/+/ovLv0FcEbdO/taVNE1j8eLFbNiwgZ07d7qWXxPiUrzx1jsEITpl9By1mq1MTZlKgCUAgK/Lv2be+/NodbR26XUOlR/ins330Ko6zzsgZAC3xd7WpdcQ18boOWpU8YHxTE6ajPJ/t8le+/Y1/t///j80TeOvB//Kwg8WuoryxD6JZKVlSVF+CdeFXcekpEmupRyz87O5Z/M97SbWa2119hurDqziqX1PufaPix9HanCqewMW4hKkHxWd6ZY75t83evToyx6jKAo7duy47HHnLVq0iDfeeIPs7GzsdrvrGe/AwEB8fGTol2ivTqnTOwQhOtUTctRusTM9dTprctfQ5Gjio4KPWLpjKavuWNUl6yoX1BSQ9W4Wda3OtugX0I8JCRM8cs3mnqgn5KhRJQclMzFxIpuObwLgpa9e4ouSL/ii5AvXMQNCBjAhYYKr8BQXSwtOw2KysPHYRtq0Nraf2k5WdhZvTHqDAGsA/v7+vP7t6/xm729c7xkZO9Lww1xF7+Ep/Wj8v8TTWNuIj11qrq7klt5fVVU0TWv309bWxvHjx9m1axdnzpxBVa/umeoXX3yR6upqRo0aRVRUlOvnrbfe6qbfQvRk5Uq53iEI0amekqMhPiFkpmRiVpyPBL2Z9yb/9el//eDznms8x/Ts6a7ZqMN9w8lMzpQixUB6So4aVXpIOnf2v9O1/c9F+bCIYdyVcJfk+xVI6JPgHOpvco4q+KToE6ZsmEJFQwWf1nzKQzsfch07PHI4I6JH6BSpEBfzlH50wrIJTF0xlQnLJugdikdxyx3zXbt2dfja5s2bWbhwIU8//fRVnbMb5qwTHqy/2p/D5sN6hyFEh3pSjsbaY5mUOImN+RsBePqLp4n2j+a+6++7pvPVt9Yza9MsjlU6Z6kOtASSlZqFt1mG/BlJT8pRo7o+7HpaHa2uicxMiokxcWNkDoWrFBcQx4y0GfzjyD9ocjTxZdmX/OiNH1HeUI6G8/Ph9WHXM7LvSJ0jFaI96UdFZ3T/anbixIncc889PPTQQ3qHIoQQ4gqlBKcwtt9Y1/by3ct57/h7V32eVkcr89+f71qr2MfLh6y0LHy9fbssViGMZGjkUCYlTmJQ6CBmp8+WovwaRflHcXf63fh7+wNQ1lDmKsrTQ9IZHz9eHoMRQvQouhfm4Fz7e9++fXqHITxYkVKkdwhCdKon5ugNETcwIso5TFTVVBZsXcBnxZ9d8ftVTeXBHQ+y/dR2ALxN3kxPnU6QLahb4hU/TE/MUaNKD0lnQsIEovyj9A6lRwv1DWXOwDlE+0W79t0QfgMTEyZKUS4MSfpR0Rm3DGXvTFtbG2+//TahoaF6hyI8mAWZ4VYYW0/N0dtib6O2pZbcs7k0OZq4e9PdbJm2hZTglMu+d8XeFbyV55wX5Pwaw5F+kd0dsrhGPTVHhWezW+zMTJ/J4bOHifGKITgoWO+QhOiQp/Sjm/64icaaRnwCfJi0fJLe4XgMtxTm99136ecOq6qqyMnJoaSk5KqfMRfiaoRqoZTjGRNuCM/UU3NUURR+3P/H1LfWc6rmFJVNlWRuzOT9ae8TFxDX4fue2/8cq75c5TwHChMSJtAvsJ+7whbXoKfmqPB8XiYvrgu7jnRHOoeR53eFcXlKP3r6m9OudcxF13FLYb5z586LhhQpikJQUBA/+tGP+NnPfsa4cePcEYoQQoguZjaZyUjOYO3htZQ1lFFUV0TGOxlsnraZaP/oi45fk7uGlR+vdG2PjhtNeki6GyMWQgghhDAWtxTmJ0+edMdlhOjQEdMRvUMQolM9PUetZivTU6ez9vBazjWd42TNSaZsmMLmqZsJ8w1zHbf+yHqW7ljq2r4p6iaGRg7VI2RxlXp6jgrPJzkqjM5TcvSnL/9U7xA8kiEmfxOiu8Wr8XqHIESnPCFH/bz9mJE2g0BrIADHKo8xdeNUqpqqANiUv4mff/hzVE0FnMsZ3Rp7q17hiqvkCTkqPJvkqDA6T8lRq6/V9SO6jlvumO/Zs+ea3nfbbbd1cSSit7IiHYcwNk/JUbvFzoy0GazNXUttay2HKg4xLXsas9Jn8eieR3FoDgAGhAyQ5Yx6GE/JUeG5JEeF0UmOis64pTAfNWrUZT98aZrmOub8fzscDneEJ3qBeur1DkGITnlSjvax9iErLYu1h9fS0NbAgdIDHCg94Ho9NTiVuxLukqK8h/GkHBWeSXJUGJ3kqOiMWwrzrVu3snz5choaGli4cCGpqakA5OXl8d///d/4+fnxhz/8AatVvkUS3aPEVKJ3CEJ0ytNyNMQnxHnn/PBamhxNrv1JfZKYlDhJivIeyNNyVHgeyVFhdJ6So0c/PkprcyveVm9Sbrn88qjiyritMLdYLOTk5GCz2Vz7J02axKJFixg5ciQffPABTz31lDvCEb1QoprIYbMsoSKMyxNzNMw3jNkDZpN7NhdvkzehPqEk9kmUoryH8sQcFZ5FclQYnafk6PaXtruWS5PCvOu4ZfK3NWvWMGfOnHZF+Xm+vr7MmTOH119/3R2hCCGEcKMQnxBujb2Vm6JvIikoSYpyIYQQQohLcMsd8/r6eoqLizt8vbi4mIaGBneEInqpEsUzhg4JzyU5KoxOclQYneSoMDpPydFb595Ka1Mr3jZvvUPxKG4pzMeOHcuf//xnhg0bRmZmZrvX1q9fz5///GfGjx/vjlBEL2WSlQGFwUmOCqOTHBVGJzkqjM5TcvS6O67TOwSP5JbCfNWqVYwePZrp06cTFRVFUlISAMePH6eoqIjExESef/55d4QieqlwLZyznNU7DCE6JDkqjE5yVBid5KgwOslR0Rm3fG0TExPDwYMHefrppxk0aBClpaWUlpYycOBAnnnmGQ4ePEhsbKw7QhFCCCGEEEIIIQzFLXfMAWw2G0uXLmXp0qXuuqQQLsdMx/QOQYhOSY4Ko5McFUYnOSqMTnJUdMYtd8zPnTvH119/3eHr33zzDZWVle4IRfRSfdW+eocgRKckR4XRSY4Ko5McFUbnKTn6lzl/4Y8T/shf5vxF71A8ilsK81/+8pcsXLiww9fvv/9+fvWrX7kjFNFL2bh4qT4hjERyVBid5KgwOslRYXSSo6IzbinMd+7cyeTJkzt8fdKkSWzfvt0doYheqpFGvUMQolOSo8LoJEeF0UmOCqPzlByNSIogOi2aiKQIvUPxKG55xry8vJzQ0NAOXw8JCaGsrMwdoYhe6ozpjN4hCNEpyVFhdJKjwugkR4XReUqOTl0xVe8QPJJb7phHRUXx5Zdfdvj6/v37CQsLc0coopdKVpP1DkGITkmOCqOTHBVGJzkqjE5yVHTGLYV5RkYGr7zyCu++++5Fr2VnZ7N69WqmTJnijlCEEEIIIYQQQghDcctQ9pUrV7J9+3amTJnC4MGDGTRoEACHDh3i4MGDpKen89vf/tYdoYheqkyRRyWEsUmOCqOTHBVGJzkqjE5yVHTGLXfMAwMDycnJ4T/+4z9obW1l3bp1rFu3jtbWVn7zm9/w2Wef0adPn2s696pVq4iPj8dmszFixAg+//zzrg1eeAQNTe8QhOiU5KgwOslRYXSSo8LoPCVHt7+4nc1/2sz2F2Xy7q7klsIcwM/Pj9/+9rd88803NDQ00NDQwDfffMPKlSvx8/O7pnO+9dZbLFu2jBUrVnDgwAEGDx7M+PHjZSI5cZEITWaNFMYmOSqMTnJUGJ3kqDA6T8nRo58cJfejXI5+clTvUDyK2wrz7vD000+zYMEC5s+fz4ABA3jppZfw9fXlb3/7m96hCSGEEEIIIYQQV8Qtz5gDNDU1sX79eg4cOEB1dTWqqrZ7XVEUXnnllSs+X0tLC/v37+eRRx5x7TOZTIwdO5ZPP/20w/fV1tZiMl34PsJqtWK1Wq/iN+l+GhqtaqveYXiUPPKkTYWhSY4Ko5McFUYnOSqMzlNyNOsPWWiqhmJSdP19HKpDt2t3B7cU5qdOneL222/n5MmT9OnTh+rqaoKDg6mqqsLhcBAaGoq/v/9VnbOiogKHw0FERPshIREREeTl5XX4vkGDBtHQ0ODanj9/PkuWLCEqKorjx4+7zqFpmmtIfHJyMmfOnKGxsRGbzUbfvn05duwYAOHh4ZhMJkpKSgBITEykpKSE+vp6rFYr8fHxHDlyBIDQ0FAsFgtFRUUA9O/fn/Lycurq6vD29iYpKYnDhw9zs/lmysxltJhaCG4Jdv6+1gr82vzwcfigKiolPiVENUShoFDvVU+TuYmQ5hAAzlrP4uPwwbfNFw2NYt9iIhsiMWGi0dxIvVc9oc3OdeXPWc5hVa34tTkfJyjyLSKiMQKzZqbR3EidVx1hzc6l7CotlXipXtjb7AAU+xQT1hSGl+ZFs6mZaks14U3hAFRZqjBpJgJaAwAo8SkhpDkEb9WbFlMLlZZKIpqc/++qvasBCGwNBKDUVkpQSxAW1UKrqZWz1rNENkYCUONdg6qo9GnpA0CZrYzAlkCsqpU2pY1yWzlRjVEA1HrV0mZqc57LYaHQpxD/Nn98HD44FAelPqVEN0QDUO9VT7Op+dLtjUqJ74X2bvBqoNHc2K69bQ4bfm1+F9q7MRKTdun2tqgW/Nuc+V7kU0REk7O9m8xN1HrXEtZ09e1d7V2NgtJhe5+znCOy6UIbamjt2rtPSx+sqpVWpZUKW8WFNvSuxaE42rV3QGsANoeNNqWNMlsZ0Y3ONqzzqqPV1EpQSxAA5dbyK27vs9az+LT54Ou4kLOu9jY30Oh1ob2vNme9Ve927R3eFI6X5kWTuYka75p2OWvWzNhbL7R3aFMo3po3zaZmqixV7XK2XXvbSghuCe4wZ/+5vTvKWYvDwlnLWVfOApTbyrG32rE5bM42tJW2a2/pI7quj7janO2NfYS91U69uV76CJ36iEvmrPQRrvYGCGsKo8XUIn2EfI4wbB/Rt74vLaaWnt9H9DdIH2E14ShxkFuRS0pKCgUFBTQ1NeHj40NMTAz5+fkArnqxtLQUgKSkJAoLC121XVxcHEePOoflh4WF4eXlRXFxMQAJCQmUlpZSX1+PxWIhISHBVWuGhIRgtVrb1XYVFRXU1tbi5eVFSkoKubm5aNqVzS2gaFd65A+QlZXFjh072LJlCwkJCYSHh7N9+3ZuueUWnnvuOV544QW2b99OcvKVr+1XVFRETEwMn3zyCTfffLNr//Lly9m9ezefffZZu+Pb2trYvXs3CQkJhr9jLrpWc3Mzjz76KI8//rj8vxaGJDkqjE5yVBid5KgwOsnR3svhcHDw4EFGjhyJl1fH98Xd8oz5zp07+cUvfsHw4cNdRbGmaVitVh5++GHGjBnDQw89dFXnDA0NxWw2u775OK+0tJTIyMgO32e32wkICHD9yB+G52tubmb16tU0NzfrHYoQlyQ5KoxOclQYneSoMDrJUXE5binMGxoaiI+PByAgIABFUaiurna9fvPNN7N3796rOqfFYmHo0KHs2LHDtU9VVXbs2NHuDroQQgghhBBCiK6xd+9eduzYcdX1m+icW54xj4uL48yZM84LenkRExNDTk4OmZmZAOTm5mKz2a76vMuWLePee+9l2LBhDB8+nGeffZb6+nrmz5/fpfELIYQQQgghhID777+f4uJioqKi+Pbbb/UOx2O4pTAfPXo02dnZrFixAoB58+bxxBNPUFlZiaqq/P3vf2fu3LlXfd4ZM2ZQXl7OY489RklJCUOGDGHr1q0XTQgHuB66dzg8a/Y+cXmqquLr64uqqvL/XxiS5KgwOslRYXSSo8LoPClHbTYbvr6+2Gy2Hv+7uMP5Nrrc1G5umfytoKCAffv2MXHiRKxWK01NTSxevJj169djNpuZOHEizz33HAEBAd0WQ1NTEx9//HG3nV8IIYQQQgghhLiUW265pdNR4m4pzI1AVVVaWlowm80oiqJ3OEIIIYQQQgghPJymaTgcDiwWS7vVwb6v1xTmQgghhBBCCCGEEbllVnYhhBBCCCGEEEJcmhTmQgghhBBCCCGEjqQwF0IIIYQQQgghdCSFuRBCCCGEEEIIoSMpzEWv1NzczJAhQ1AUha+++krvcIQA4OTJk/z0pz+lf//++Pj4kJiYyIoVK2hpadE7NNGLrVq1ivj4eGw2GyNGjODzzz/XOyQhXJ544gluvPFG7HY74eHhZGRkcOTIEb3DEuKSfv/736MoCg899JDeoQgDksJc9ErLly8nOjpa7zCEaCcvLw9VVXn55Zf59ttveeaZZ3jppZd49NFH9Q5N9FJvvfUWy5YtY8WKFRw4cIDBgwczfvx4ysrK9A5NCAB2797NokWLyMnJYdu2bbS2tjJu3Djq6+v1Dk2Idvbt28fLL7/M9ddfr3cowqBkuTTR62zZsoVly5axfv16Bg4cyJdffsmQIUP0DkuIS/rTn/7Eiy++yIkTJ/QORfRCI0aM4MYbb+SFF14AQFVV+vbty5IlS/j1r3+tc3RCXKy8vJzw8HB2797Nbbfdpnc4QgBQV1fHDTfcwF/+8hd+97vfMWTIEJ599lm9wxIGI3fMRa9SWlrKggUL+Pvf/46vr6/e4QhxWdXV1QQHB+sdhuiFWlpa2L9/P2PHjnXtM5lMjB07lk8//VTHyIToWHV1NYD0m8JQFi1axF133dWuPxXi+7z0DkAId9E0jXnz5vHzn/+cYcOGcfLkSb1DEqJT+fn5PP/88zz55JN6hyJ6oYqKChwOBxEREe32R0REkJeXp1NUQnRMVVUeeughbrnlFgYNGqR3OEIA8Oabb3LgwAH27dundyjC4OSOuejxfv3rX6MoSqc/eXl5PP/889TW1vLII4/oHbLoZa40R/9ZYWEhd955J9OnT2fBggU6RS6EED3HokWLOHToEG+++abeoQgBwOnTp1m6dClr1qzBZrPpHY4wOHnGXPR45eXlnD17ttNjEhISyMrKYtOmTSiK4trvcDgwm83Mnj2b//mf/+nuUEUvdaU5arFYACgqKmLUqFHcdNNNvPrqq5hM8h2qcL+WlhZ8fX1Zt24dGRkZrv333nsvVVVVZGdn6xecEN+zePFisrOz2bNnD/3799c7HCEA2LhxI1OmTMFsNrv2ORwOFEXBZDLR3Nzc7jXRu0lhLnqNgoICampqXNtFRUWMHz+edevWMWLECGJjY3WMTginwsJCbr/9doYOHcrrr78u/2ALXY0YMYLhw4fz/PPPA86hwnFxcSxevFgmfxOGoGkaS5YsYcOGDezatYvk5GS9QxLCpba2llOnTrXbN3/+fNLS0vj3f/93eeRCtCPPmIteIy4urt22v78/AImJiVKUC0MoLCxk1KhR9OvXjyeffJLy8nLXa5GRkTpGJnqrZcuWce+99zJs2DCGDx/Os88+S319PfPnz9c7NCEA5/D1N954g+zsbOx2OyUlJQAEBgbi4+Ojc3Sit7Pb7RcV335+foSEhEhRLi4ihbkQQhjEtm3byM/PJz8//6Ivi2Rwk9DDjBkzKC8v57HHHqOkpIQhQ4awdevWiyaEE0IvL774IgCjRo1qt3/16tXMmzfP/QEJIcQ1kqHsQgghhBBCCCGEjmRGISGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNCRFOZCCCGEEEIIIYSOpDAXQgghhBBCCCF0JIW5EEIIIYQQQgihIynMhRBCCCGEEEIIHUlhLoQQQgghhBBC6EgKcyGEEEIIIYQQQkdSmAshhBBCCCGEEDqSwlwIIYQQQgghhNDR/wcll6fLYf2yCQAAAABJRU5ErkJggg==",
       "text/plain": [
        "<Figure size 1200x600 with 2 Axes>"
       ]
@@ -534,16 +535,14 @@
     "        distributions.mvn_to_norm,\n",
     "    )\n",
     "\n",
-    "    # Selecting\n",
-    "    acqf = acquisition_fn(model)\n",
-    "    \n",
-    "    next_x, _ = bfgs(\n",
-    "        random.fold_in(key, i),\n",
-    "        acqf,\n",
-    "        bounds,\n",
-    "        batch_size,\n",
-    "        num_samples,\n",
-    "        num_restarts,\n",
+    "    # Selecting    \n",
+    "    acqf = models.outcome_transformed(\n",
+    "        vmap(model),\n",
+    "        acquisitions.upper_confidence_bound(2.0)\n",
+    "    )\n",
+    "\n",
+    "    next_x, _ = optimizer_fn(acqf)(\n",
+    "        random.fold_in(key, i)\n",
     "    )\n",
     "\n",
     "    # Evaluating\n",
@@ -578,7 +577,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.7"
   }
  },
  "nbformat": 4,
diff --git a/docs/source/guides/Untitled.ipynb b/docs/source/guides/Untitled.ipynb
deleted file mode 100644
index 20cc81c..0000000
--- a/docs/source/guides/Untitled.ipynb
+++ /dev/null
@@ -1,494 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "61b0cf2e-48b7-425f-b32b-e90c7e301f89",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from jax import config\n",
-    "\n",
-    "config.update(\"jax_enable_x64\", True)\n",
-    "\n",
-    "from jax import numpy as jnp\n",
-    "from jax import jit, lax, nn, random, value_and_grad, vmap\n",
-    "\n",
-    "import optax\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "plt.style.use('bmh')\n",
-    "\n",
-    "from boax.core import distributions, samplers\n",
-    "from boax.prediction import models, objectives\n",
-    "from boax.optimization import acquisitions, optimizers"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "55860138-87d3-4ce2-8ddc-bf7faed007ba",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "data_key, optimizer_key = random.split(random.key(0))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "087ac062-9584-40e3-ad0f-3d1e707d9bd5",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def objective(x):\n",
-    "    return -((x[..., 0] + 1) ** 2) * jnp.sin(2 * x[..., 0] + 2) / 5 + 1"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "id": "a23cb1db-ca8d-4568-b391-4038352632b4",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def approximate(x):\n",
-    "    return 0.5 * objective(x) + x[..., 0] / 4 + 2"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "dc827755-2117-4d1a-ae36-f3b54eb29ce9",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "bounds = jnp.array([[-5.0, 5.0]])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "id": "36aeac5b-ede2-49fc-845e-1b4bd87970e8",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "fidelities = jnp.array([0.5, 1.0])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "id": "4c59393e-f1e3-46a5-83a9-1707ccbc363c",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "x_train_values = random.uniform(random.fold_in(data_key, 0), minval=bounds[:, 0], maxval=bounds[:, 1], shape=(10, 1))\n",
-    "x_train_fidelities = random.randint(random.fold_in(data_key, 1), minval=0, maxval=2, shape=(10, 1))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "id": "99e42b33-fc9c-4945-ae87-b876593257b8",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "x_train = jnp.concatenate(\n",
-    "    [x_train_values, fidelities[x_train_fidelities]], axis=-1,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "id": "e3a5325a-9dbb-4f8e-8d9f-28be117b1656",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "y_train = jnp.where(\n",
-    "    x_train_fidelities[..., 0],\n",
-    "    objective(x_train_values),\n",
-    "    approximate(x_train_values)\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "id": "94fb7e2e-b4b1-4e29-a32e-5ea7b1ec4afe",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "xs = jnp.linspace(bounds[:, 0], bounds[:, 1], 501)\n",
-    "ys_true = objective(xs)\n",
-    "ys_approx = approximate(xs)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "id": "b99309b2-51cb-44bd-a915-607e0b33607c",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1200x400 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "fig, ax = plt.subplots(figsize=(12, 4))\n",
-    "\n",
-    "ax.plot(xs, ys_true, c='r', label='objective')\n",
-    "ax.plot(xs, ys_approx, c='k', linestyle='--', label='approximation')\n",
-    "ax.scatter(x_train_values, y_train, marker='x', c='k', label='observations')\n",
-    "ax.legend()\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "id": "c94c2231-34fb-42de-b29d-5863ba4c5864",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def model_fn(params, x_train, y_train):\n",
-    "    return models.multi_fidelity(\n",
-    "        models.means.zero(),\n",
-    "        lambda fid1, fid2: models.kernels.scaled(\n",
-    "            models.kernels.linear_truncated(\n",
-    "                models.kernels.matern_five_halves(nn.softplus(params['unbiased'])),\n",
-    "                models.kernels.matern_five_halves(nn.softplus(params['biased'])),\n",
-    "                nn.softplus(params['power']),\n",
-    "            )(fid1, fid2),\n",
-    "            nn.softplus(params['amplitude']),\n",
-    "        ),\n",
-    "        models.likelihoods.gaussian(nn.softplus(params['noise'])),\n",
-    "        x_train,\n",
-    "        y_train,\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "id": "3059ac02-6e3e-4553-9445-9a1a2aa28b69",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def loss_fn(params, x_train, y_train):\n",
-    "    y_hat = model_fn(params, None, None)(x_train)\n",
-    "    \n",
-    "    objective = objectives.penalized(\n",
-    "        objectives.negative_log_likelihood(\n",
-    "            distributions.multivariate_normal.logpdf\n",
-    "        ),\n",
-    "        -jnp.sum(\n",
-    "            distributions.gamma.logpdf(\n",
-    "                distributions.gamma.gamma(2.0, 0.15),\n",
-    "                nn.softplus(params['amplitude']),\n",
-    "            )\n",
-    "        ),\n",
-    "        -jnp.sum(\n",
-    "            distributions.gamma.logpdf(\n",
-    "                distributions.gamma.gamma(3.0, 3.0),\n",
-    "                nn.softplus(params['power']),\n",
-    "            )\n",
-    "        ),\n",
-    "        -jnp.sum(\n",
-    "            distributions.gamma.logpdf(\n",
-    "                distributions.gamma.gamma(3.0, 6.0),\n",
-    "                nn.softplus(params['unbiased']),\n",
-    "            )\n",
-    "        ),\n",
-    "        -jnp.sum(\n",
-    "            distributions.gamma.logpdf(\n",
-    "                distributions.gamma.gamma(6.0, 2.0),\n",
-    "                nn.softplus(params['biased']),\n",
-    "            )\n",
-    "        )\n",
-    "    )\n",
-    "\n",
-    "    return objective(y_hat, y_train)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "id": "c6ab3460-015c-4963-99e0-d24fa45fcc83",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "params = {\n",
-    "    'biased': jnp.zeros(()),\n",
-    "    'unbiased': jnp.zeros(()),\n",
-    "    'power': jnp.zeros(()),\n",
-    "    'amplitude': jnp.zeros(()),\n",
-    "    'noise': jnp.zeros(()),\n",
-    "}"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "id": "bf78868e-dbe1-444d-b71b-e3d2f6cb8248",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "adam = optax.adam(0.01)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "id": "ba5b1bd4-c2b0-4f48-83de-baf72512866c",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def fit(x_train, y_train):\n",
-    "    def step(state, i):\n",
-    "        loss, grads = value_and_grad(loss_fn)(state[0], x_train, y_train)\n",
-    "        updates, opt_state = adam.update(grads, state[1])\n",
-    "        params = optax.apply_updates(state[0], updates)\n",
-    "\n",
-    "        return (params, opt_state), loss\n",
-    "\n",
-    "    (next_params, _), _ = lax.scan(\n",
-    "        jit(step),\n",
-    "        (params, adam.init(params)),\n",
-    "        jnp.arange(500),\n",
-    "    )\n",
-    "\n",
-    "    return next_params"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "id": "d403af5d-872a-4cac-b0eb-e1be0c33fc73",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "y_mean, y_var = jnp.mean(y_train), jnp.var(y_train)\n",
-    "y_norm = nn.standardize(y_train, mean=y_mean, variance=y_var)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "id": "4b41a926-7235-4335-8767-11164e14ac58",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "next_params = fit(x_train, y_norm)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "id": "44291f9c-614c-4b3b-97aa-5a84cd3f9dca",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "model = models.scaled(\n",
-    "    models.outcome_transformed(\n",
-    "        model_fn(next_params, x_train, y_norm),\n",
-    "        distributions.mvn_to_norm,\n",
-    "    ),\n",
-    "    distributions.normal.scale,\n",
-    "    loc=y_mean,\n",
-    "    scale=jnp.sqrt(y_var),\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "id": "05a0bc3d-151b-488f-ada7-a7a1eba3f7a6",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "y_hat = model(jnp.hstack([xs, jnp.ones((501, 1))]))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "id": "633e5615-7028-4d1d-81f2-f9a516ca6ee4",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1200x400 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "fig, ax = plt.subplots(figsize=(12, 4))\n",
-    "\n",
-    "ax.plot(xs, ys_true, c='r', label='objective')\n",
-    "ax.plot(xs, ys_approx, c='k', linestyle='--', label='approximation')\n",
-    "ax.scatter(x_train_values, y_train, marker='x', c='k', label='observations')\n",
-    "\n",
-    "ax.plot(xs, y_hat.loc, label='mean')\n",
-    "ax.fill_between(\n",
-    "    xs.flatten(),\n",
-    "    y_hat.loc - 2 * y_hat.scale,\n",
-    "    y_hat.loc + 2 * y_hat.scale,\n",
-    "    alpha=0.3,\n",
-    "    label='95% CI'\n",
-    ")\n",
-    "\n",
-    "ax.legend(loc='upper left')\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "id": "b9d81dbd-2ea0-4e15-a346-e7f7117e31b0",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def cost(x):\n",
-    "    return 10 + x[..., 1]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 32,
-   "id": "d90d542b-4e0c-4194-bf8d-0a6dd1b505c6",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "model_cost = models.joined(\n",
-    "    model,\n",
-    "    cost,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 42,
-   "id": "c9e415c2-e969-45e5-9de5-42b9f27c2b8d",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "key1, key2 = random.split(random.key(0))\n",
-    "s, n = 32, 10\n",
-    "\n",
-    "best = 0.0\n",
-    "loc, scale = random.uniform(key1, (2, n, s, 1))\n",
-    "cost = random.uniform(key2, (n,))\n",
-    "preds = distributions.normal.normal(loc, scale)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 47,
-   "id": "8d532346-278c-4cbb-a5b7-4b17ef0430af",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "(10,)"
-      ]
-     },
-     "execution_count": 47,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "cost.shape"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 46,
-   "id": "4f79deff-d624-4c47-a390-40bfc6561301",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "(10, 32)"
-      ]
-     },
-     "execution_count": 46,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "jnp.squeeze(preds.loc - best, axis=-1).shape"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 53,
-   "id": "4968d650-3b19-4b22-ba3f-ac32940c69b0",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Array([0.73711553, 1.26476817, 0.92278978, 0.68727337, 0.50606529,\n",
-       "       0.67834267, 0.86699063, 0.61253229, 1.11807063, 2.28338359],      dtype=float64)"
-      ]
-     },
-     "execution_count": 53,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "jnp.mean((jnp.squeeze(preds.loc - best, axis=-1) / cost[..., jnp.newaxis]), axis=-1)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "286b4fba-4de9-4eee-8112-3883bd7505e1",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.5"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}

From b7cab402a18761c39a6f5420e1a4641554a635fa Mon Sep 17 00:00:00 2001
From: Lando Loeper <lando.loeper@deutschebahn.com>
Date: Sun, 21 Apr 2024 16:19:39 +0200
Subject: [PATCH 9/9] fix: fix formatting

---
 boax/core/samplers/__init__.py                |  4 ++--
 boax/core/samplers/alias.py                   |  8 +++----
 boax/core/samplers/functions/iid.py           |  2 +-
 boax/core/samplers/functions/quasi_random.py  |  6 +++--
 boax/optimization/acquisitions/__init__.py    |  4 +++-
 boax/optimization/acquisitions/alias.py       |  4 ++--
 boax/optimization/optimizers/alias.py         | 12 ++++------
 boax/optimization/optimizers/base.py          |  2 +-
 .../optimizers/initializers/alias.py          |  4 ++--
 boax/prediction/models/__init__.py            |  2 +-
 .../models/functions/multi_fidelity.py        |  8 +++----
 boax/prediction/models/gaussian_process.py    | 23 +++++++++++--------
 boax/prediction/models/transformed.py         |  2 +-
 tests/core/samplers_test.py                   |  2 +-
 .../acquisitions/acquisitions_test.py         |  7 +++---
 .../optimizers/initializers_test.py           | 12 ++++++++--
 16 files changed, 57 insertions(+), 45 deletions(-)

diff --git a/boax/core/samplers/__init__.py b/boax/core/samplers/__init__.py
index 499384a..cd52a36 100644
--- a/boax/core/samplers/__init__.py
+++ b/boax/core/samplers/__init__.py
@@ -14,8 +14,8 @@
 
 """The samplers sub-package."""
 
-from .alias import normal as normal
-from .alias import uniform as uniform
 from .alias import halton_normal as halton_normal
 from .alias import halton_uniform as halton_uniform
+from .alias import normal as normal
+from .alias import uniform as uniform
 from .base import Sampler as Sampler
diff --git a/boax/core/samplers/alias.py b/boax/core/samplers/alias.py
index 7bb1f31..e9501fb 100644
--- a/boax/core/samplers/alias.py
+++ b/boax/core/samplers/alias.py
@@ -16,7 +16,7 @@
 
 from functools import partial
 
-from jax import lax, random
+from jax import lax
 from jax import numpy as jnp
 
 from boax.core import distributions
@@ -43,12 +43,12 @@ def uniform(
   Returns:
     The corresponding `Sampler`.
   """
-  
+
   out_shape = lax.broadcast_shapes(uniform.a.shape, uniform.b.shape)
 
   return compose(
     partial(partial, distributions.uniform.sample)(uniform),
-    partial(functions.iid.uniform, ndims=out_shape[0])
+    partial(functions.iid.uniform, ndims=out_shape[0]),
   )
 
 
@@ -73,7 +73,7 @@ def normal(
 
   return compose(
     partial(partial, distributions.normal.sample)(normal),
-    partial(functions.iid.normal, ndims=out_shape[0])
+    partial(functions.iid.normal, ndims=out_shape[0]),
   )
 
 
diff --git a/boax/core/samplers/functions/iid.py b/boax/core/samplers/functions/iid.py
index 3355b91..58dd348 100644
--- a/boax/core/samplers/functions/iid.py
+++ b/boax/core/samplers/functions/iid.py
@@ -18,7 +18,7 @@
 
 from jax import random
 
-from boax.utils.typing import PRNGKey, Array
+from boax.utils.typing import Array, PRNGKey
 
 
 def uniform(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Array:
diff --git a/boax/core/samplers/functions/quasi_random.py b/boax/core/samplers/functions/quasi_random.py
index 624557c..5ec38f4 100644
--- a/boax/core/samplers/functions/quasi_random.py
+++ b/boax/core/samplers/functions/quasi_random.py
@@ -28,7 +28,9 @@
 assert len(PRIMES) == MAX_DIMENSION
 
 
-def halton_sequence(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Array:
+def halton_sequence(
+  key: PRNGKey, sample_shape: Sequence[int], ndims: int
+) -> Array:
   shuffle_key, correction_key = random.split(key)
   num_samples = jnp.prod(jnp.asarray(sample_shape))
 
@@ -53,7 +55,7 @@ def halton_sequence(key: PRNGKey, sample_shape: Sequence[int], ndims: int) -> Ar
 
   return jnp.reshape(
     base_values + (zero_correction / (radixes**max_sizes_by_axes)).flatten(),
-    sample_shape + (ndims,)
+    sample_shape + (ndims,),
   )
 
 
diff --git a/boax/optimization/acquisitions/__init__.py b/boax/optimization/acquisitions/__init__.py
index 8648f5e..987659c 100644
--- a/boax/optimization/acquisitions/__init__.py
+++ b/boax/optimization/acquisitions/__init__.py
@@ -25,10 +25,12 @@
 from .alias import probability_of_improvement as probability_of_improvement
 from .alias import q_expected_improvement as q_expected_improvement
 from .alias import q_knowledge_gradient as q_knowledge_gradient
+from .alias import (
+  q_multi_fidelity_knowledge_gradient as q_multi_fidelity_knowledge_gradient,
+)
 from .alias import q_probability_of_improvement as q_probability_of_improvement
 from .alias import q_upper_confidence_bound as q_upper_confidence_bound
 from .alias import upper_confidence_bound as upper_confidence_bound
-from .alias import q_multi_fidelity_knowledge_gradient as q_multi_fidelity_knowledge_gradient
 from .base import Acquisition as Acquisition
 from .transformed import constrained as constrained
 from .transformed import log_constrained as log_constrained
diff --git a/boax/optimization/acquisitions/alias.py b/boax/optimization/acquisitions/alias.py
index 2ab9aaa..48de837 100644
--- a/boax/optimization/acquisitions/alias.py
+++ b/boax/optimization/acquisitions/alias.py
@@ -388,7 +388,7 @@ def q_multi_fidelity_knowledge_gradient(
           attrgetter('loc'),
           itemgetter(0),
         ),
-        itemgetter(1)
-      )
+        itemgetter(1),
+      ),
     )
   )
diff --git a/boax/optimization/optimizers/alias.py b/boax/optimization/optimizers/alias.py
index 3091635..9a79df1 100644
--- a/boax/optimization/optimizers/alias.py
+++ b/boax/optimization/optimizers/alias.py
@@ -67,14 +67,10 @@ def sequential(initializer: Initializer, solver: Solver, q: int) -> Optimizer:
   inner = batch(initializer, solver)
 
   def optimizer(key):
-    next_candidates, values = zip(*(
-      inner(random.fold_in(key, i))
-      for i in range(q)
-    ))
-
-    return (
-      jnp.concatenate(list(next_candidates)),
-      jnp.array(list(values))
+    next_candidates, values = zip(
+      *(inner(random.fold_in(key, i)) for i in range(q))
     )
 
+    return (jnp.concatenate(list(next_candidates)), jnp.array(list(values)))
+
   return optimizer
diff --git a/boax/optimization/optimizers/base.py b/boax/optimization/optimizers/base.py
index 37bb568..f35322e 100644
--- a/boax/optimization/optimizers/base.py
+++ b/boax/optimization/optimizers/base.py
@@ -14,7 +14,7 @@
 
 """Base interface for optimizers."""
 
-from typing import Callable, Protocol, Tuple
+from typing import Protocol, Tuple
 
 from boax.utils.typing import Array, PRNGKey
 
diff --git a/boax/optimization/optimizers/initializers/alias.py b/boax/optimization/optimizers/initializers/alias.py
index e38d2b0..826b121 100644
--- a/boax/optimization/optimizers/initializers/alias.py
+++ b/boax/optimization/optimizers/initializers/alias.py
@@ -52,10 +52,10 @@ def q_batch(
 
   def initializer(key):
     key1, key2 = random.split(key)
-    
+
     x = sampler(key1, (num_results, q))
     y = fun(x)
-    
+
     return random.choice(
       key2,
       x,
diff --git a/boax/prediction/models/__init__.py b/boax/prediction/models/__init__.py
index b0b0163..6b1a433 100644
--- a/boax/prediction/models/__init__.py
+++ b/boax/prediction/models/__init__.py
@@ -14,10 +14,10 @@
 
 """The models sub-package."""
 
+from . import gaussian_process as gaussian_process
 from . import kernels as kernels
 from . import likelihoods as likelihoods
 from . import means as means
-from . import gaussian_process as gaussian_process
 from .base import Model as Model
 from .transformed import input_transformed as input_transformed
 from .transformed import joined as joined
diff --git a/boax/prediction/models/functions/multi_fidelity.py b/boax/prediction/models/functions/multi_fidelity.py
index 72ceba5..0623f05 100644
--- a/boax/prediction/models/functions/multi_fidelity.py
+++ b/boax/prediction/models/functions/multi_fidelity.py
@@ -14,7 +14,7 @@
 
 """Multi fidelity functions."""
 
-from typing import Callable, Tuple
+from typing import Callable
 
 from jax import numpy as jnp
 from jax import scipy
@@ -56,14 +56,12 @@ def posterior(
   Kxx = kernel_fn(observation_fidelities, observation_fidelities)(
     observation_index_points, observation_index_points
   )
-  
+
   Kxz = kernel_fn(observation_fidelities, fidelities)(
     observation_index_points, index_points
   )
 
-  Kzz = kernel_fn(fidelities, fidelities)(
-    index_points, index_points
-  )
+  Kzz = kernel_fn(fidelities, fidelities)(index_points, index_points)
 
   K = Kxx + jitter * jnp.identity(Kxx.shape[-1])
   chol = scipy.linalg.cholesky(K, lower=True)
diff --git a/boax/prediction/models/gaussian_process.py b/boax/prediction/models/gaussian_process.py
index 43348e7..338d10e 100644
--- a/boax/prediction/models/gaussian_process.py
+++ b/boax/prediction/models/gaussian_process.py
@@ -85,7 +85,7 @@ def exact(
         ),
       )
     )
-  
+
 
 def fantasy(
   mean_fn: Mean,
@@ -102,7 +102,7 @@ def fantasy(
           jitter=jitter,
         )
       ),
-      in_axes=(0, None, 0)
+      in_axes=(0, None, 0),
     )
   )
 
@@ -136,9 +136,9 @@ def multi_fidelity(
   """
 
   if (
-    observation_index_points is None or
-    observation_fidelities is None or
-    observations is None
+    observation_index_points is None
+    or observation_fidelities is None
+    or observations is None
   ):
     return jit(
       compose(
@@ -167,7 +167,7 @@ def multi_fidelity(
         ),
       )
     )
-  
+
 
 def multi_fidelity_fantasy(
   mean_fn: Mean,
@@ -184,11 +184,16 @@ def multi_fidelity_fantasy(
           jitter=jitter,
         )
       ),
-      in_axes=(0, 0, None, None, 0)
+      in_axes=(0, 0, None, None, 0),
     )
   )
 
-  def fn(fantasy_points, observation_index_points, observation_fidelities, observations):
+  def fn(
+    fantasy_points,
+    observation_index_points,
+    observation_fidelities,
+    observations,
+  ):
     return fantasy_fn(
       fantasy_points,
       jnp.ones_like(fantasy_points),
@@ -196,5 +201,5 @@ def fn(fantasy_points, observation_index_points, observation_fidelities, observa
       observation_fidelities,
       observations,
     )
-  
+
   return fn
diff --git a/boax/prediction/models/transformed.py b/boax/prediction/models/transformed.py
index 4b9a288..585f832 100644
--- a/boax/prediction/models/transformed.py
+++ b/boax/prediction/models/transformed.py
@@ -20,7 +20,7 @@
 from jax import vmap
 
 from boax.prediction.models.base import Model
-from boax.utils.functools import apply, call, compose, identity, unwrap
+from boax.utils.functools import apply, call, compose
 from boax.utils.typing import Array
 
 A = TypeVar('A')
diff --git a/tests/core/samplers_test.py b/tests/core/samplers_test.py
index 7413e4e..13b865d 100644
--- a/tests/core/samplers_test.py
+++ b/tests/core/samplers_test.py
@@ -27,7 +27,7 @@ def test_uniform(self):
     result = samplers.uniform(uniform)(key3, (2, 5, 3))
 
     assert_shape(result, (2, 5, 3, 10))
-  
+
   def test_halton_normal(self):
     key1, key2, key3 = random.split(random.key(0), 3)
 
diff --git a/tests/optimization/acquisitions/acquisitions_test.py b/tests/optimization/acquisitions/acquisitions_test.py
index 44ced13..bca2140 100644
--- a/tests/optimization/acquisitions/acquisitions_test.py
+++ b/tests/optimization/acquisitions/acquisitions_test.py
@@ -112,7 +112,7 @@ def test_q_knowledge_gradient(self):
     qkg = acquisitions.q_knowledge_gradient(best)(preds)
 
     assert_shape(qkg, (n,))
-  
+
   def test_q_multi_fidelity_knowledge_gradient(self):
     key1, key2 = random.split(random.key(0))
     s, n = 32, 10
@@ -123,7 +123,9 @@ def test_q_multi_fidelity_knowledge_gradient(self):
     preds = distributions.normal.normal(loc, scale)
     costs = random.uniform(key2, (s,))
 
-    qmfkg = acquisitions.q_multi_fidelity_knowledge_gradient(best, cost_fn)((preds, costs))
+    qmfkg = acquisitions.q_multi_fidelity_knowledge_gradient(best, cost_fn)(
+      (preds, costs)
+    )
 
     assert_shape(qmfkg, (n,))
 
@@ -147,7 +149,6 @@ def test_constrained(self):
       constraints.log_less_or_equal(1.0),
     )(model)
 
-
     assert_shape(cei, (n,))
     assert_shape(clei, (n,))
     assert_trees_all_close(jnp.log(cei), clei, atol=1e-4)
diff --git a/tests/optimization/optimizers/initializers_test.py b/tests/optimization/optimizers/initializers_test.py
index d79853b..8c5bb30 100644
--- a/tests/optimization/optimizers/initializers_test.py
+++ b/tests/optimization/optimizers/initializers_test.py
@@ -17,7 +17,11 @@ def test_q_batch(self):
     s, n, q, d = 10, 5, 3, 1
 
     initializer = optimizers.initializers.q_batch(
-      fun, sampler, q, s, n,
+      fun,
+      sampler,
+      q,
+      s,
+      n,
     )
 
     result = initializer(key)
@@ -32,7 +36,11 @@ def test_q_nonnegative(self):
     s, n, q, d = 10, 5, 3, 1
 
     initializer = optimizers.initializers.q_batch_nonnegative(
-      fun, sampler, q, s, n,
+      fun,
+      sampler,
+      q,
+      s,
+      n,
     )
 
     result = initializer(key)