Get list from dir and separate train and test (python function)

from sklearn.model_selection import train_test_split

import random

import os

import glob

def train_test_split_from_dir(origin_dir, test_size=0.2):

os.chdir(origin_dir)

#get list

data_list = []

for file in glob.glob("*.jpg"):

data_list.append(file)

random.shuffle(data_list)

train_json_list, test_json_list = train_test_split(data_list, test_size=test_size)

return train_json_list, test_json_list

ROC & AUC example code in face detector model case

..

#https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html
import numpy as np
from sklearn import metrics
import matplotlib.pyplot as plt

#model #1
y = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0])
scores = np.array([0.64, 0.47, 0.46, 0.77, 0.72, 0.9, 0.85, 0.7, 0.87, 0.92, 0.89, 0.93, 0.85, 0.81, 0.88, 0.48, 0.1, 0.35, 0.68, 0.47])
fpr, tpr, thresholds = metrics.roc_curve(y, scores)
roc_auc = metrics.auc(fpr, tpr)

# plot
plt.title('Receiver Operating Characteristic')
plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)
plt.legend(loc = 'lower right')
plt.plot([0, 1], [0, 1],'r--')
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()

..

Example model metrics using sklearn in face detector case

..

from sklearn.metrics import classification_report
#model 1
y_true = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
y_pred = [0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
target_names = ['Non Face', 'Face']
print(classification_report(y_true, y_pred, target_names=target_names, digits=3))

..

..

#model 2
y_true = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
y_pred = [0, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
target_names = ['Non Face', 'Face']
print(classification_report(y_true, y_pred, target_names=target_names, digits=3))

..