from tensorflow.keras.layers import Dropout
from tensorflow.keras.layers import Flatten
from tensorflow.keras.layers import GlobalAveragePooling2D
from tensorflow.keras.layers import Input
from tensorflow.keras.layers import LeakyReLU
from tensorflow.keras.layers import MaxPool2D
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam
from mnistData import MNIST
random.seed(0)
np.random.seed(0)
tf.random.set_seed(0)
data = MNIST()
data.data_augmentation(augment_size=10000)
data.data_preprocessing(preprocess_mode="MinMax")
(
x_train_splitted,
x_val,
y_train_splitted,
y_val,
) = data.get_splitted_train_validation_set()
x_train, y_train = data.get_train_set()
x_test, y_test = data.get_test_set()
num_classes = data.num_classes
# Save Path
dir_path = os.path.abspath(
"C:/Users/Jan/Dropbox/_Programmieren/UdemyTF/models/")
x = Flatten()(x)
x = Dense(units=num_classes)(x)
y_pred = Activation("softmax")(x)
model = Model(
inputs=[input_img],
outputs=[y_pred]
)
model.summary()
return model
if __name__ == "__main__":
data = MNIST(with_normalization=False)
data.data_augmentation(augment_size=5_000)
x_train_, x_val_, y_train_, y_val_ = data.get_splitted_train_validation_set()
model = build_model(data.img_shape, data.num_classes)
model.compile(
loss="categorical_crossentropy",
optimizer=Adam(learning_rate=0.0005),
metrics=["accuracy"]
)
tb_callback = TensorBoard(
log_dir=MODEL_LOG_DIR,
write_graph=True
x = Activation("relu")(x)
x = MaxPool2D()(x)
x = Flatten()(x)
x = Dense(units=num_classes)(x)
y_pred = Activation("softmax")(x)
model = Model(inputs=[input_img], outputs=[y_pred])
model.summary()
return model
if __name__ == "__main__":
data = MNIST(with_normalization=True)
x_train_, x_val_, y_train_, y_val_ = data.get_splitted_train_validation_set(
)
model = build_model(data.img_shape, data.num_classes)
model.compile(loss="categorical_crossentropy",
optimizer=Adam(learning_rate=0.0005),
metrics=["accuracy"])
tb_callback = TensorBoard(log_dir=MODEL_LOG_DIR, write_graph=True)
model.fit(x=x_train_,
y=y_train_,
epochs=40,
model = Model(
inputs=[input_img],
outputs=[y_pred]
)
model.compile(
loss="categorical_crossentropy",
optimizer="Adam",
metrics=["accuracy"]
)
return model
if __name__ == "__main__":
data = MNIST(with_normalization=True)
x_train, y_train = data.get_train_set()
param_distributions = {
"filters_1": randint(8, 64),
"kernel_size_1": randint(3, 8),
"filters_2": randint(8, 64),
"kernel_size_2": randint(3, 8),
"filters_3": randint(8, 64),
"kernel_size_3": randint(3, 8),
}
keras_clf = KerasClassifier(
build_fn=build_model,
epochs=3,
x = Flatten()(x)
x = Dense(units=128)(x)
x = Activation("relu")(x)
x = Dense(units=num_classes)(x)
y_pred = Activation("softmax")(x)
model = Model(inputs=[input_img], outputs=[y_pred])
model.summary()
return model
if __name__ == "__main__":
data = MNIST()
x_train, y_train = data.get_train_set()
x_test, y_test = data.get_test_set()
img_shape = data.img_shape
num_classes = data.num_classes
model = build_model(img_shape, num_classes)
model.compile(loss="categorical_crossentropy",
optimizer="Adam",
metrics=["accuracy"])
model.fit(x=x_train / 255.0,
y=y_train,
epochs=3,