Brain_Tumor_Classification/CNN_Model.py at main · ArjunT13/Brain_Tumor_Classification · GitHub

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
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
from tensorflow.keras.callbacks import EarlyStopping
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout
from tensorflow.keras.layers import BatchNormalization
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.optimizers import Adam, SGD
from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint

generator_train = ImageDataGenerator(rescale=1./255,
                                    featurewise_center=False,
                                    samplewise_center=False,
                                    featurewise_std_normalization=False,
                                    samplewise_std_normalization=False,
                                    zca_whitening=False,
                                    rotation_range=0,
                                    zoom_range = 0,
                                    width_shift_range=0,
                                    height_shift_range=0,
                                    horizontal_flip=True,
                                    vertical_flip=False)

generator_test = ImageDataGenerator(rescale=1./255,
                                    featurewise_center=False,
                                    samplewise_center=False,
                                    featurewise_std_normalization=False,
                                    samplewise_std_normalization=False,
                                    zca_whitening=False,
                                    rotation_range=0,
                                    zoom_range = 0,
                                    width_shift_range=0,
                                    height_shift_range=0,
                                    horizontal_flip=True,
                                    vertical_flip=False)


train = generator_train.flow_from_directory('./Training', target_size=(64,64),
                                              batch_size=32, class_mode= "categorical", color_mode='grayscale')

test = generator_test.flow_from_directory('./Testing', target_size=(64,64),
                                              batch_size=32, class_mode= "categorical", color_mode='grayscale')

model1 = Sequential()

# Convolutional layer 1
model1.add(Conv2D(32,(3,3), input_shape=(64, 64, 1), activation='relu'))
model1.add(BatchNormalization())
model1.add(MaxPooling2D(pool_size=(2,2)))

# Convolutional layer 2
model1.add(Conv2D(32,(3,3), activation='relu'))
model1.add(BatchNormalization())
model1.add(MaxPooling2D(pool_size=(2,2)))

model1.add(Flatten())

model1.add(Dense(units= 252, activation='relu'))
model1.add(Dropout(0.2))
model1.add(Dense(units=252, activation='relu'))
model1.add(Dropout(0.2))
model1.add(Dense(units=4, activation='softmax'))

optimizer = tf.keras.optimizers.Adam(learning_rate=0.001, decay=0.0001, clipvalue=0.5)
model1.compile(optimizer=optimizer, loss='categorical_crossentropy',
                   metrics= ['categorical_accuracy'])

#model1.summary()

model1_es = EarlyStopping(monitor = 'loss', min_delta = 1e-11, patience = 12, verbose = 1)
model1_rlr = ReduceLROnPlateau(monitor = 'val_loss', factor = 0.2, patience = 6, verbose = 1)

# Automatically saves the best weights of the model, based on best val_accuracy
model1_mcp = ModelCheckpoint(filepath = 'model1_weights.h5', monitor = 'val_categorical_accuracy',
                      save_best_only = True, verbose = 1)

# Fiting the model.
history1 = model1.fit(train, steps_per_epoch=5712//32, epochs=50, validation_data=test, validation_steps= 1311//32,
                     callbacks=[model1_es, model1_rlr, model1_mcp])

# Model held for testing
model_evaluation = model1.evaluate(test)