AvielMak

def plot_calibration(y_true, y_probability,model_name='LGBM'):
    # compute calibration
    prob_df_clf = pd.DataFrame({'y': y_true, 'y_hat': y_probability})
    # binning the dataframe, so we can see success rates for bins of probability
    bins = np.arange(0.05, 1.00, 0.05)
    prob_df_clf.loc[:,'prob_bin'] = np.digitize(prob_df_clf['y_hat'], bins)
    prob_df_clf.loc[:,'prob_bin_val'] = prob_df_clf['prob_bin'].replace(dict(zip(range(len(bins)), bins)))

    # opening figure
    plt.figure(figsize=(12,7), dpi=150)

AvielMak

# Def plot distribution
def plot_distribution(data_select,data,h,limit=()):
    """
    Plot_Distribution to see how 2 classes are seperated on different features
    data_select = which feature we want to see
    data = the DataFrame
    h = column of the two classes
    limit = non mendatory, tuple to limit the x_axis
    """
    figsize =( 15, 8)