def plot_one_univariate(ax, df, x_var_name, y_var_name, mask=None):
""" A linear spline regression of two columns in the dataframe.
INPUT:
ax:
Matplotlib axis to plot on.
(use 'fig, ax = matplotlib.pyplot.subplots(1,1)')
df:
Dataframe of floats or ints.
x_var_name:
the column name of the x var from dataframe
y_var_name:
string, the column name of the dependent y variable in
the dataframe
OUTPUT:
A linear regression, with light blue bootstrapped lines showing the
instability of the regression
"""
min_y = min(df[y_var_name])
max_y = max(df[y_var_name])
ax.set_ylim(min_y - .1 * np.abs(min_y), max_y + .1 * np.abs(max_y))
if mask is None:
plot_univariate_smooth(ax,
df[x_var_name].values.reshape(-1, 1),
df[y_var_name],
bootstrap=200)
else:
plot_univariate_smooth(ax,
df[x_var_name].values.reshape(-1, 1),
df[y_var_name],
mask=mask,
bootstrap=200)
return None
def plot_one_univariate(ax, dataframe, x_var_name, y_var_name, mask=None):
""" A linear spline regression of two columns in the dataframe. of string 'y_var' across the named string 'xvar' in the dataframe var_name on matplotlib axis 'ax'
INPUT:
ax:
matplotlib axis
(use 'fig, ax = matplotlib.pyplot.subplots(1,1)')
dataframe:
dataframe of floats or ints
x_var_name:
the index name of the x var from dataframe
y_var_name:
the index name of the y var from dataframe
OUTPUT:
A linear regression, with light blue bootstrapped lines showing the instability of the regression
"""
min_y = min(dataframe[y_var_name])
max_y = max(dataframe[y_var_name])
ax.set_ylim(min_y - .1 * np.abs(min_y), max_y + .1 * np.abs(max_y))
if mask is None:
plot_univariate_smooth(ax,
dataframe[x_var_name].values.reshape(-1, 1),
dataframe[y_var_name],
bootstrap=200)
else:
plot_univariate_smooth(ax,
dataframe[x_var_name].values.reshape(-1, 1),
dataframe[y_var_name],
mask=mask,
bootstrap=200)
def plot_univariate_bynumeric(target_column):
fig, axs = plt.subplots(len(numeric_predictors), figsize=(14, 14))
for name, ax in zip(numeric_predictors, axs.flatten()):
plot_univariate_smooth(ax,
data_df[name].values.reshape(-1, 1),
data_df[target_column],
bootstrap=100)
ax.set_title(name)
fig.tight_layout()
plt.show()
from regression_tools.plotting_tools import (
plot_univariate_smooth,
bootstrap_train,
display_coef,
plot_bootstrap_coefs,
plot_partial_depenence,
plot_partial_dependences,
predicteds_vs_actuals)
fig, axs = plt.subplots(10, 1, figsize=(14, 25))
univariate_plot_names = clean_df.columns[:-1]
for name, ax in zip(univariate_plot_names, axs.flatten()):
plot_univariate_smooth(ax,
clean_df[name].values.reshape(-1, 1),
clean_df['SalesPrice'],
bootstrap=100)
ax.set_title(name)
fig.tight_layout()
#Plotting Neural Networks
def plot_results(model, ax=None, function=np.sin, xlim=(-7, 7), data_lim=(-5, 5)):
x_actual = np.linspace(*xlim, 500)
y_actual = function(x_actual)
if ax is None:
fig, ax = plt.subplots(figsize=(12, 5))
y_pred = model.predict(x_actual)
ax.plot(x_actual,y_pred, 'r-', label='model predictions')