training_code_original.py

import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)

# Input data files are available in the read-only "../input/" directory
# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory

import os
for dirname, _, filenames in os.walk('/kaggle/input/face-mask-detection/images'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

import os
import xml.etree.ElementTree as et
import re
import pandas as pd

dic = {"image": [],"Dimensions": []}
for i in range(1,116):
	dic[f'Object {i}']=[]
print("Generating data in CSV format....")

for file in os.listdir("/kaggle/input/face-mask-detection/annotations"):
    row = []
    xml = et.parse("/kaggle/input/face-mask-detection/annotations/"+file)
    root = xml.getroot()
    img = root[1].text
    row.append(img)
    h,w = root[2][0].text,root[2][1].text
    row.append([h,w])

    for i in range(4,len(root)):
        temp = []
        temp.append(root[i][0].text)
        for point in root[i][5]:
            temp.append(point.text)
        row.append(temp)
    for i in range(len(row),119):
        row.append(0)
    for i,each in enumerate(dic):
        dic[each].append(row[i])
df = pd.DataFrame(dic)

from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.applications import MobileNetV2
from tensorflow.keras.layers import AveragePooling2D
from tensorflow.keras.layers import Dropout
from tensorflow.keras.layers import Flatten
from tensorflow.keras.layers import Dense
from tensorflow.keras.layers import Input
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.applications.mobilenet_v2 import preprocess_input
from tensorflow.keras.preprocessing.image import img_to_array
from tensorflow.keras.preprocessing.image import load_img
from tensorflow.keras.utils import to_categorical
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report
import matplotlib.pyplot as plt
import numpy as np
import argparse
import pandas as pd
import glob
import os
import cv2
import random as rand

image_directories = sorted(glob.glob(os.path.join("/kaggle/input/face-mask-detection/images", "*.png")))
j = 0
classes = ["without_mask", "mask_weared_incorrect", "with_mask"]
labels = []
data = []

print("Extracting each data into respective label folders....")
for idx, image in enumerate(image_directories):
    img = cv2.imread(image)
    # scale to dimension
    X, Y = df["Dimensions"][idx]
    cv2.resize(img, (int(X), int(Y)))
    # find the face in each object
    for obj in df.columns[3:]:
        info = df[obj][idx]
        if info != 0:
            label = info[0]
            info[0] = info[0].replace(str(label), str(classes.index(label)))
            info = [int(each) for each in info]
            face = img[info[2]:info[4], info[1]:info[3]]
            if ((info[3] - info[1]) > 40 and (info[4] - info[2]) > 40):
                try:
                    face = cv2.resize(face, (224, 224))
                    face = img_to_array(face)
                    face = preprocess_input(face)
                    data.append(face)
                    labels.append(label)
                    if (label == "mask_weared_incorrect"):
                        data.append(face)
                        labels.append(label)

                except:
                    pass

print("Done!")

data = np.array(data, dtype="float32")
labels = np.array(labels)

lb = LabelEncoder()
labels = lb.fit_transform(labels)
labels = to_categorical(labels)

aug = ImageDataGenerator(
    zoom_range=0.1,
    rotation_range=25,
    width_shift_range=0.1,
    height_shift_range=0.1,
    shear_range=0.15,
    horizontal_flip=True,
    fill_mode="nearest"
    )

baseModel = MobileNetV2(weights="imagenet", include_top=False,
	input_tensor=Input(shape=(224, 224, 3)))

# construct the head of the model that will be placed on top of the
# the base model
headModel = baseModel.output
headModel = AveragePooling2D(pool_size=(7, 7))(headModel)
headModel = Flatten(name="flatten")(headModel)
headModel = Dense(64, activation="relu")(headModel)
headModel = Dropout(0.5)(headModel)
headModel = Dense(3, activation="softmax")(headModel)

# place the head FC model on top of the base model (this will become
# the actual model we will train)
model = Model(inputs=baseModel.input, outputs=headModel)

# loop over all layers in the base model and freeze them so they will
# *not* be updated during the first training process
for layer in baseModel.layers:
	layer.trainable = False

INIT_LR = 1e-4
EPOCHS = 50
BS = 1

(trainX, testX, trainY, testY) = train_test_split(data, labels,
	test_size=0.3, stratify=labels, random_state=42)

print("[INFO] compiling model...")
opt = Adam(lr=INIT_LR, decay=INIT_LR / EPOCHS)
model.compile(loss="categorical_crossentropy", optimizer=opt,
	metrics=["accuracy"])

# train the head of the network
print("[INFO] training head...")
H = model.fit(
	aug.flow(trainX, trainY, batch_size=BS),
	steps_per_epoch=len(trainX) // BS,
	validation_data=(testX, testY),
	validation_steps=len(testX) // BS,
	epochs=EPOCHS)

print("[INFO] evaluating network...")
predIdxs = model.predict(testX, batch_size=32)

# for each image in the testing set we need to find the index of the
# label with corresponding largest predicted probability
predIdxs = np.argmax(predIdxs, axis=1)

# show a nicely formatted classification report
print(classification_report(testY.argmax(axis=1), predIdxs,
	target_names=lb.classes_))

# serialize the model to disk
print("[INFO] saving mask detector model...")

# plot the training loss and accuracy
N = EPOCHS
plt.style.use("ggplot")
plt.figure()
plt.plot(np.arange(0, N), H.history["loss"], label="train_loss")
plt.plot(np.arange(0, N), H.history["val_loss"], label="val_loss")
plt.plot(np.arange(0, N), H.history["accuracy"], label="train_acc")
plt.plot(np.arange(0, N), H.history["val_accuracy"], label="val_acc")
plt.title("Training Loss and Accuracy")
plt.xlabel("Epoch #")
plt.ylabel("Loss/Accuracy")
plt.legend(loc="lower left")
plt.show()