Module model_tools.model_plot_tools

A module for calculating various statistics not included in the sklearn or scipy packages.

# -*- coding: utf-8 -*-
"""
A module for calculating various statistics not included
in the sklearn or scipy packages.
"""
import matplotlib.pyplot as plt
import numpy as np

def plot_ROC_curve(fpr, tpr):
    fig, ax = plt.subplots(figsize=(9, 8))
    ax.plot(fpr, tpr)
    ax.plot([0, 1], [0, 1], linestyle='--', linewidth=2)
    ax.set_ylabel('True Positive Rate', fontsize=15)
    ax.set_xlabel('False Positive Rate', fontsize=15)
    return fig, ax

def plot_variance_covariance_matrix(variance_covariance_matrix, feature_labels):
    """
    returns a matplotlib figure and axis of a pcolormesh
    plot of the input matrix
    """
    varcov = np.array(variance_covariance_matrix)
    fig, ax = plt.subplots(figsize=(9, 8))
    cbar = ax.pcolormesh(np.array(varcov), vmin=-0.01, vmax=0.01, cmap='seismic')
    fig.colorbar(cbar, label='covariance')
    ax.set_xticks(np.arange(0.5, 7, 1))
    ax.set_yticks(np.arange(0.5, 7, 1))
    ax.set_xticklabels(np.concatenate([('intercept',),np.array(feature_labels)]), rotation=90)
    ax.set_yticklabels(np.concatenate([('intercept',),np.array(feature_labels)]), rotation=0)

    for n,mc in enumerate(np.array(varcov)):
          for i,m in enumerate(mc):
                varcov_value = np.format_float_scientific(np.round(m, 5))
                ax.text(x=n+0.15, y=i+0.35, s=varcov_value, color='black', fontsize=10)
    
    return fig, ax

Functions

def plot_ROC_curve(fpr, tpr)

Source code

def plot_ROC_curve(fpr, tpr):
    fig, ax = plt.subplots(figsize=(9, 8))
    ax.plot(fpr, tpr)
    ax.plot([0, 1], [0, 1], linestyle='--', linewidth=2)
    ax.set_ylabel('True Positive Rate', fontsize=15)
    ax.set_xlabel('False Positive Rate', fontsize=15)
    return fig, ax

def plot_variance_covariance_matrix(variance_covariance_matrix, feature_labels)

returns a matplotlib figure and axis of a pcolormesh plot of the input matrix

Source code

def plot_variance_covariance_matrix(variance_covariance_matrix, feature_labels):
    """
    returns a matplotlib figure and axis of a pcolormesh
    plot of the input matrix
    """
    varcov = np.array(variance_covariance_matrix)
    fig, ax = plt.subplots(figsize=(9, 8))
    cbar = ax.pcolormesh(np.array(varcov), vmin=-0.01, vmax=0.01, cmap='seismic')
    fig.colorbar(cbar, label='covariance')
    ax.set_xticks(np.arange(0.5, 7, 1))
    ax.set_yticks(np.arange(0.5, 7, 1))
    ax.set_xticklabels(np.concatenate([('intercept',),np.array(feature_labels)]), rotation=90)
    ax.set_yticklabels(np.concatenate([('intercept',),np.array(feature_labels)]), rotation=0)

    for n,mc in enumerate(np.array(varcov)):
          for i,m in enumerate(mc):
                varcov_value = np.format_float_scientific(np.round(m, 5))
                ax.text(x=n+0.15, y=i+0.35, s=varcov_value, color='black', fontsize=10)
    
    return fig, ax