-
Notifications
You must be signed in to change notification settings - Fork 0
/
plotting.py
63 lines (48 loc) · 2.65 KB
/
plotting.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
"""
Analysis code for article: Image-Based & Machine Learning-Guided Multiplexed Serology Test for SARS-CoV-2
Methods for plotting
@author: Christian Guckelsberger, Lassi Paavolainen
"""
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
def plot_confusion_matrix(cm, classes):
import pretty_print_cm as pp
df_cm = pd.DataFrame(cm, index = [c for c in classes.values()], columns = [c for c in classes.values()])
df_cm.rename(columns={"small bright":"small\nbright", "double positive": "double\npositive", "His positive":"HIS positive", "double negative": "double\nnegative"}, inplace=True)
df_cm.index=df_cm.columns
pp.pretty_plot_confusion_matrix(df_cm, percentages=False, fmt='d')
def plot_correlation(data, protein, antibody):
fig_corr = plt.figure(figsize=(6,6))
ax_corr = fig_corr.gca() #Get Current Axis
ax_corr.cla()
# Plot correlations
data_cov = data[data['Sample Name'].str.contains('S2020')]
data_uu = data[data['Sample Name'].str.contains('S2017')]
data_uu_reduced = data_uu.sample(n=data_cov.shape[0])
# Export raw data used for correlation plot
col_pred = "Ratio positive {}".format(antibody)
col_gt_raw = "Raw {} {}".format(protein, antibody)
ax_corr.plot(data_cov[col_gt_raw], data_cov[col_pred], 'o', alpha=0.5, color="red", label='COVID-19')
ax_corr.plot(data_uu_reduced[col_gt_raw], data_uu_reduced[col_pred], 'o', alpha=0.5, color="darkcyan", label='Negative')
ax_corr.set_xlabel('Raw ELISA Values',fontsize=14)
ax_corr.set_ylabel('Predicted Positive Ratio',fontsize=14)
#ax_corr.set_title('Positives={}, Negatives={}, Total={}'.format(data_cov.shape[0], data_uu_reduced.shape[0], data_cov.shape[0]+data_uu_reduced.shape[0]),fontsize=14)
ax_corr.legend()
fig_corr.show()
def plot_ROC_curve(scores_val):
fig_roc = plt.figure(figsize=(6,6))
ax_roc = fig_roc.gca() #Get Current Axis
ax_roc.cla()
ax_roc.plot((1-scores_val['specificity']), scores_val['sensitivity'], color="green", alpha=0.7, linewidth=3)
ax_roc.plot([0, 1], [0, 1], transform=ax_roc.transAxes, color='black', linestyle='--')
ax_roc.spines['right'].set_visible(False)
ax_roc.spines['top'].set_visible(False)
ax_roc.set_xlim([0.0, 1.01])
ax_roc.set_ylim([0.0, 1.01])
ax_roc.xaxis.set_ticks(np.arange(0, 1.01, 0.25))
ax_roc.yaxis.set_ticks(np.arange(0, 1.01, 0.25))
ax_roc.set_xlabel('False Positive Rate (1-Specificity)', fontsize=14)
ax_roc.set_ylabel('True Positive Rate (Sensitivity)', fontsize=14)
#ax_roc.set_title('{0}; GT: {1} p, {2} n, {3} total; AUC: {4:.2f}'.format(ch, counts[1],counts[0],y_gt.shape[0], area))
fig_roc.show()