incremental updates v3

.gitignore

@@ -228,3 +228,8 @@ pip-log.txt
 *.ai
 twoStep_output.csv
 Saltykov_output.csv
+
+#############
+## Notebook checkpoints
+#############
+.ipynb_checkpoints

DOCS/getting_started.md

@@ -153,7 +153,7 @@ dataset.head()
 
 ![](https://github.com/marcoalopez/GrainSizeTools/blob/master/FIGURES/dataframe_diameters.png?raw=true)
 
-> 👉 Note that we define the square root and pi as ``np.sqrt`` and ``np.pi``. In this case, ``np.`` stems for numerical Python, a basic package for scientific computing with Python, and the word after the dot the method or scientific variables to be applied. If you write in the console ``np.`` and then press press the TAB key, you will see a huge list of methods available. In general, the name of the methods used are equivalent to those used in MATLAB but always adding the ``np.`` first.
+> 👉 Note that we define the square root and pi as ``np.sqrt`` and ``np.pi``. In this case, ``np.`` stems for Numpy or numerical Python, a basic package for scientific computing with Python, and the word after the dot the method or scientific variables to be applied. If you write in the console ``np.`` and then press press the TAB key, you will see a huge list of methods available. In general, the name of the methods used are equivalent to those used in MATLAB but always adding the ``np.`` first.
 
 
 

grain_size_tools/GrainSizeTools_script.py

@@ -134,39 +134,47 @@ def summarize(data, avg=('amean', 'gmean', 'median', 'mode'), ci_level=0.95,
     """
 
     # check and remove for negative values
-    if data[data < 0].size > 0:
+    if data[data <= 0].size > 0:
         print('Warning: I found negative and/or zero values in your dataset!')
         data = data[data > 0]
         print('Negative/zero values automatically removed')
         print('')
 
     std = np.std(data)
 
+    # estimate Shapiro-Wilk test to check normality and lognormality
+    W, p_value = shapiro(data)
+    W2, p_value2 = shapiro(np.log(data))
+
     if 'amean' in avg:
-        amean, __, ci, length = averages.amean(data, ci_level, method='ASTM')
-        if len(data) > 99:
-            __, __, (low_ci, high_ci), length2 = averages.amean(data, ci_level, method='mCox')
+        if p_value2 < 0.05:
+            amean, __, ci, length = averages.amean(data, ci_level, method='ASTM')
         else:
-            __, __, (low_ci, high_ci), length2 = averages.amean(data, ci_level, method='GCI')
+            if len(data) > 99:
+                amean, __, (low_ci, high_ci), length2 = averages.amean(data, ci_level, method='mCox')
+            else:
+                amean, __, (low_ci, high_ci), length2 = averages.amean(data, ci_level, method='GCI')
 
-        # estimate coefficients of variation
-        lower_cvar = 100 * (amean - low_ci) / amean
-        upper_cvar = 100 * (high_ci - amean) / amean
+            # estimate coefficients of variation
+            lower_cvar = 100 * (amean - low_ci) / amean
+            upper_cvar = 100 * (high_ci - amean) / amean
 
         print(' ')
         print('============================================================================')
         print('CENTRAL TENDENCY ESTIMATORS')
         print('============================================================================')
         print('Arithmetic mean = {:0.2f} microns' .format(amean))
         print('Confidence intervals at {:0.1f} %' .format(ci_level * 100))
-        print('ASTM method: {:0.2f} - {:0.2f}, (±{:0.1f}%), length = {:0.3f}'
-              .format(ci[0], ci[1], 100 * (ci[1] - amean) / amean, length))
-        if len(data) > 99:
-            print('mCox method: {:0.2f} - {:0.2f} (-{:0.1f}%, +{:0.1f}%), length = {:0.3f}'
-                  .format(low_ci, high_ci, lower_cvar, upper_cvar, length2))
+        if p_value2 < 0.05:
+            print('ASTM method: {:0.2f} - {:0.2f}, (±{:0.1f}%), length = {:0.3f}'
+                  .format(ci[0], ci[1], 100 * (ci[1] - amean) / amean, length))
         else:
-            print('GCI method: {:0.2f} - {:0.2f} (-{:0.1f}%, +{:0.1f}%), length = {:0.3f}'
-                  .format(low_ci, high_ci, lower_cvar, upper_cvar, length2))
+            if len(data) > 99:
+                print('mCox method: {:0.2f} - {:0.2f} (-{:0.1f}%, +{:0.1f}%), length = {:0.3f}'
+                      .format(low_ci, high_ci, lower_cvar, upper_cvar, length2))
+            else:
+                print('GCI method: {:0.2f} - {:0.2f} (-{:0.1f}%, +{:0.1f}%), length = {:0.3f}'
+                      .format(low_ci, high_ci, lower_cvar, upper_cvar, length2))
 
     if 'gmean' in avg:
         m = 'CLT' if len(data) > 99 else 'bayes'  # choose optimal method to estimate confidence intervals
@@ -207,10 +215,6 @@ def summarize(data, avg=('amean', 'gmean', 'median', 'mode'), ci_level=0.95,
         else:
             print('KDE bandwidth =', bandwidth)
 
-    # estimate Shapiro-Wilk test to check normality and lognormality
-    W, p_value = shapiro(data)
-    W2, p_value2 = shapiro(np.log(data))
-
     print(' ')
     print('============================================================================')
     print('DISTRIBUTION FEATURES')

grain_size_tools/notebook_example.ipynb

@@ -836,7 +836,6 @@
       "============================================================================\n",
       "Arithmetic mean = 22.13 microns\n",
       "Confidence intervals at 95.0 %\n",
-      "ASTM method: 21.33 - 22.93, (±3.6%), length = 1.596\n",
       "mCox method: 21.35 - 22.98 (-3.5%, +3.8%), length = 1.623\n",
       "============================================================================\n",
       "Geometric mean = 20.44 microns\n",
@@ -892,7 +891,6 @@
       "Arithmetic mean = 34.79 microns\n",
       "Confidence intervals at 95.0 %\n",
       "ASTM method: 34.09 - 35.48, (±2.0%), length = 1.393\n",
-      "mCox method: 34.43 - 36.04 (-1.0%, +3.6%), length = 1.616\n",
       "============================================================================\n",
       "Geometric mean = 30.10 microns\n",
       "Confidence interval at 95.0 %\n",
@@ -1033,7 +1031,7 @@
      "output_type": "stream",
      "text": [
       "=======================================\n",
-      "Normalized SD = 0.16\n",
+      "Normalized SD = 0.165\n",
       "KDE bandwidth =  0.04\n",
       "=======================================\n"
      ]