px.scatter(cz_df, x='travel_lt15', y='aum', width=350, height=250,
- labels={'travel_lt15':'Commute time under 15 min', 'aum':'Upward mobility'})
+ labels={'travel_lt15':'Commute time under 15 min',
+ 'aum':'Upward mobility'})
fig = px.scatter(x=prediction, y=error,
- labels=dict(x='prediction for AUM', y='error'),
+ labels=dict(x='Prediction for AUM', y='Error'),
width=350, height=250)
fig.add_hline(0, line_width=2, line_dash='dash', opacity=1)
@@ -1806,7 +1807,7 @@ 15.6.2. Relating Upward Mobility Using M
fig = px.scatter(x=prediction_ct_sm, y=error_ct_sm,
- labels=dict(x='two-variable prediction for AUM', y='error'),
+ labels=dict(x='Two-variable prediction for AUM', y='Error'),
width=350, height=250)
fig.add_hline(0, line_width=2, line_dash='dash', opacity=1)
@@ -1844,7 +1845,7 @@ 15.6.2. Relating Upward Mobility Using M
fig = px.scatter(x=prediction_7var, y=error_7var,
- labels=dict(x='7-variable prediction for AUM', y='error'),
+ labels=dict(x='Seven-variable prediction for AUM', y='Error'),
width=350, height=250)
fig.add_hline(0, line_width=2, line_dash='dash', opacity=1)
diff --git a/ch/15/linear_categorical.html b/ch/15/linear_categorical.html
index 5b5e3ad3..64ea45e9 100644
--- a/ch/15/linear_categorical.html
+++ b/ch/15/linear_categorical.html
@@ -1446,7 +1446,7 @@ 15.8. Feature Engineering for Categorica
fig = px.scatter(sfh, x='log_bsqft', y='log_price',
facet_col='city', facet_col_wrap=2,
- labels={'log_bsqft':'Building sie (log ft^2)',
+ labels={'log_bsqft':'Building size (log ft^2)',
'log_price':'Sale price (log USD)'},
width=500, height=400)
diff --git a/ch/15/linear_feature_eng.html b/ch/15/linear_feature_eng.html
index 75ab5fcc..175c426f 100644
--- a/ch/15/linear_feature_eng.html
+++ b/ch/15/linear_feature_eng.html
@@ -1459,7 +1459,7 @@ 15.7. Feature Engineering for Numeric Me
fig = px.scatter(x=prediction, y=error,
- labels=dict(x='Predicted Sale Price (log USD)', y='error'),
+ labels=dict(x='Predicted sale price (log USD)', y='Error'),
width=350, height=250)
fig.add_hline(0, line_width=2, line_dash='dash', opacity=1)
@@ -1483,7 +1483,7 @@ 15.7. Feature Engineering for Numeric Me
px.box(sfh, x='city', y='errors1_log',
category_orders={"city":["Piedmont","Lamorinda","Berkeley", "Richmond"]},
- labels=dict(errors1_log='Prrediction Error (log USD)', city=''), width=450, height=250)
+ labels=dict(errors1_log='Prediction error (log USD)', city=''), width=450, height=250)
diff --git a/ch/15/linear_multi.html b/ch/15/linear_multi.html
index e1f5b74c..66a6959c 100644
--- a/ch/15/linear_multi.html
+++ b/ch/15/linear_multi.html
@@ -1269,7 +1269,7 @@ Multiple Linear Model
fig = px.scatter(GA, x='pm25aqs', y='pm25pa',
facet_col=rh_cat, facet_col_wrap=2,
- labels={'pm25aqs':'AQS PM 2.5', 'pm25pa':'PurpleAir PM 2.5'},
+ labels={'pm25aqs':'AQS PM2.5', 'pm25pa':'PurpleAir PM2.5'},
width=550, height=350)
fig.update_layout(margin=dict(t=30))
@@ -1286,7 +1286,7 @@ Multiple Linear Model
fig = px.scatter_matrix(GA[['pm25pa', 'pm25aqs', 'rh']],
labels={'pm25aqs':'AQS', 'pm25pa':'PurpleAir',
- 'rh':'humidity'},
+ 'rh':'Humidity'},
width=550, height=400)
fig.update_traces(diagonal_visible=False)
@@ -1393,7 +1393,7 @@ Multiple Linear Model
fig = px.scatter(y = error_2var, x=predicted_2var,
- labels={"y": "error", "x": "predicted PurpleAir measurement"},
+ labels={"y": "Error", "x": "Predicted PurpleAir measurement"},
width=350, height=250)
fig.update_yaxes(range=[-12, 12])
diff --git a/ch/15/linear_pa.html b/ch/15/linear_pa.html
index 828394aa..d6f6d740 100644
--- a/ch/15/linear_pa.html
+++ b/ch/15/linear_pa.html
@@ -1400,7 +1400,7 @@ 15.2. Example: A Simple Linear Model for
px.scatter(GA, x="pm25aqs", y="pm25pa", trendline='ols',
trendline_color_override="darkorange",
- labels={'pm25aqs':'AQS PM 2.5', 'pm25pa':'PurpleAir PM 2.5'},
+ labels={'pm25aqs':'AQS PM2.5', 'pm25pa':'PurpleAir PM2.5'},
width=350, height=250)
@@ -1468,7 +1468,9 @@ 15.2. Example: A Simple Linear Model for
Let’s plot these errors against the predicted values:
-fig = px.scatter(fit, y='error', x='prediction', width=350, height=250)
+fig = px.scatter(fit, y='error', x='prediction',
+ labels={"prediction":"Predicition","error":"Error"},
+ width=350, height=250)
fig.add_hline(0, line_width=2, line_dash='dash', opacity=1)
diff --git a/ch/16/ms_cv.html b/ch/16/ms_cv.html
index b32f57c3..7d873036 100644
--- a/ch/16/ms_cv.html
+++ b/ch/16/ms_cv.html
@@ -1270,7 +1270,7 @@ 16.4. Cross-Validation\( (~)^+ \) stands for “positive part,” so when \( x \) is less than 65 it evaluates to 0, and when \( x \) is 65 or greater it is just \( x-65 \). We create this new feature and add it to the design matrix:
-y = heat_df["ccf"]
+y = heat_df["ccf"]
X = heat_df[["temp"]]
X["temp65p"] = (X["temp"] - 65) * (X["temp"] >= 65)
@@ -1280,7 +1280,7 @@ 16.4. Cross-Validation
-bend_index = LinearRegression().fit(X, y)
+bend_index = LinearRegression().fit(X, y)
@@ -1288,7 +1288,7 @@ 16.4. Cross-Validation
-temp = np.arange(10, 80, 1)
+temp = np.arange(10, 80, 1)
temp65p = (temp - 65) * (temp >= 65)
Xs = pd.DataFrame({"temp": temp, "temp65p": temp65p})
@@ -1299,9 +1299,9 @@ 16.4. Cross-Validation
-fig = px.scatter(
+fig = px.scatter(
heat_df, x="temp", y="ccf",
- labels={"temp": "Temperature (Farenheit)", "ccf": "Gas (cubic ft)"},
+ labels={"temp": "Temperature (<sup>o</sup>F)", "ccf": "Gas (ft^3)"},
width=350, height=250
)
@@ -1318,7 +1318,7 @@ 16.4. Cross-Validation\(k\)-fold cross-validation to choose the temperature value at which the line bends. Let’s consider models with bends at \( 40, 41, 42, \ldots, 68, 69 \) degrees. For each of these, we need to create the additional feature to enable the line to bend there:
-bends = np.arange(40, 70, 1)
+bends = np.arange(40, 70, 1)
for i in bends:
col = "temp" + i.astype("str") + "p"
@@ -1450,7 +1450,7 @@ 16.4. Cross-Validation
-y = heat_df['ccf']
+y = heat_df['ccf']
X = heat_df.drop(columns=['ccf'])
test_size = 22
@@ -1464,7 +1464,7 @@ 16.4. Cross-ValidationKFold method in scikit-learn:
-from sklearn.model_selection import KFold
+from sklearn.model_selection import KFold
kf = KFold(n_splits=3, shuffle=True, random_state=42)
@@ -1498,17 +1498,17 @@ 16.4. Cross-Validation
-totals = validation_errors.mean(axis=0)
+totals = validation_errors.mean(axis=0)
-fig = px.line(x=bends, y=totals, markers=True, width=350, height=250)
+fig = px.line(x=bends, y=totals, markers=True, width=350, height=250)
fig.add_annotation(
- x=60, y=350, text="3-fold Cross-Validation", showarrow=False
+ x=60, y=350, text="3-fold cross-validation", showarrow=False
)
fig.update_layout(
@@ -1525,7 +1525,7 @@ 16.4. Cross-Validation
-bent_final = LinearRegression().fit(
+bent_final = LinearRegression().fit(
X_train.loc[:, ["temp", "temp58p"]], y_train
)
@@ -1535,14 +1535,14 @@ 16.4. Cross-Validation
-y_pred_test = bent_final.predict(X_test.loc[:, ["temp", "temp58p"]])
+y_pred_test = bent_final.predict(X_test.loc[:, ["temp", "temp58p"]])
mean_squared_error(y_test, y_pred_test)
-71.40781435952444
+71.40781435952441
diff --git a/ch/16/ms_overfitting.html b/ch/16/ms_overfitting.html
index 81c1aa0d..f9bae894 100644
--- a/ch/16/ms_overfitting.html
+++ b/ch/16/ms_overfitting.html
@@ -1373,14 +1373,14 @@ 16.2. Example: Energy Consumptiongo.Scatter(x=heat_df['temp'], y=heat_df['ccf'], mode="markers"),
row=1, col=1)
-fig.update_xaxes(title='Temperature (F)', row=1, col=1)
-fig.update_yaxes(title='Gas Consumption (ft^3)', row=1, col=1)
+fig.update_xaxes(title='Temperature (<sup>o</sup>F)', row=1, col=1)
+fig.update_yaxes(title='Gas consumption (ft^3)', row=1, col=1)
fig.add_trace(
go.Scatter(x=heat_df['temp'], y=heat_df['ccf'], mode="markers"),
row=1, col=2)
-fig.update_xaxes(type="log", title='Temperature (F)', row=1, col=2)
+fig.update_xaxes(type="log", title='Temperature (<sup>o</sup>F)', row=1, col=2)
fig.update_layout(height=250, width=500, showlegend=False)
fig.show()
diff --git a/ch/16/ms_regularization.html b/ch/16/ms_regularization.html
index fa85f78d..4bb29949 100644
--- a/ch/16/ms_regularization.html
+++ b/ch/16/ms_regularization.html
@@ -1721,7 +1721,7 @@ 16.6. Example: A Market Analysis
px.line(x=alphas, y=mses,
- labels={"x": "lambda", "y": "MSE"},
+ labels={"x": "Lambda", "y": "MSE"},
width=350, height=250)
diff --git a/ch/16/ms_train_test.html b/ch/16/ms_train_test.html
index 50268c2d..aea539aa 100644
--- a/ch/16/ms_train_test.html
+++ b/ch/16/ms_train_test.html
@@ -1321,7 +1321,7 @@ 16.3. Train-Test Split