In [1]:

    

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline


    

In [2]:

    

sns.set_context('poster')
sns.set(font_scale=1.3)


    

In [3]:

    

list_factors = [
    'empo_1',
    'empo_2',
    'empo_3',
    'env_feature',
    'env_material',
    'envo_biome_1',
    'envo_biome_2',
    'envo_biome_3']


    

In [4]:

    

def plot_cm(path_in, path_out, factor):

    plt.figure()
    
    df = pd.read_csv(path_in % factor, sep='\t', index_col=0)

    df.dropna(inplace=True)

    df = df*100

    sns.heatmap(df, cmap='Blues', annot=True, fmt='.0f')
    plt.ylabel('True label')
    plt.xlabel('Predicted label')
    plt.tight_layout()
    plt.savefig(path_out % factor)


    

In [5]:

    

for factor in list_factors:

    plot_cm(
        '../../data/random-forest/%s_cm.txt', # values for roc_cm_full_set_qc_filtered
        'cm_full_set_qc_filtered.%s.pdf',
        factor)