xhendyagsx · November 18, 2022 08:43
diff --git a/vgg16_attempt_2.py b/vgg16_attempt_2.py
 # --- import the required packages
 import tensorflow 
 from tensorflow.keras.preprocessing.image import ImageDataGenerator
 from tensorflow.keras.callbacks import EarlyStopping
 from tensorflow.keras.models import Model, Sequential
 from tensorflow.keras.layers import Conv2D, GlobalAveragePooling2D, MaxPool2D, Dense, Flatten, Dropout
 from tensorflow.keras.optimizers import Adam 
 from tensorflow.keras.applications import VGG16 
 from keras.preprocessing import image
 import os
 import pathlib
 import numpy as np
 import matplotlib.pyplot as plt

 # --- Data Information
 # --- cek jumlah images and classes dalam dataset.
 train_dir = pathlib.Path("/content/folderOnColab/myDataset/aug/train/")
 test_dir = pathlib.Path("/content/folderOnColab/myDataset/aug/test/")
 print("Terdapat ", len(list(train_dir.glob('*/*.jpg'))), "images di train set.")
 print("Terdapat ", len(list(test_dir.glob('*/*.jpg'))), "images di test set.")

 class_names = list([item.name for item in train_dir.glob('*')])
 print("Adapun classnya adalah:", class_names)

 image_generator = ImageDataGenerator()
 train_generator = image_generator.flow_from_directory(train_dir, class_mode='categorical')
 test_generator = image_generator.flow_from_directory(test_dir, class_mode='categorical')

 # --- number of patterns
 NO_CLASSES = max(train_generator.class_indices.values()) + 1
 print("Jumlah class (NO_CLASSES): ", NO_CLASSES)

 # --- Transfer Learning
 base_model = VGG16(include_top=False, weights='imagenet', input_shape=(224, 224, 3))
 print("Jumlah layers pada model basis VGG16: ", len(base_model.layers))

 # --- add our own custom layer
 x = base_model.output
 x = GlobalAveragePooling2D()(x)
 x = Dense(1024,activation='relu')(x)       # add dense layers so
                                           # that the model can 
                                           # learn more complex 
                                           # functions and 
                                           # classify for better 
                                           # results.
 x = Dense(1024,activation='relu')(x)       # dense layer 2
 x = Dense(512,activation='relu')(x)        # dense layer 3
 preds = Dense(5,
              activation='softmax')(x)     # final layer with 
                                           # softmax activation
 # --- create a new model with the base model's original 
 # input and the new model's output
 model = Model(inputs = base_model.input, outputs = preds)
 # --- don't train the first 19 layers - 0..18
 for layer in model.layers[:19]:
    layer.trainable=False
 # --- train the rest of the layers - 19 onwards
 for layer in model.layers[19:]:
    layer.trainable=True

 # --- compile the model    
 model.compile(optimizer=Adam(learning_rate=2e-5),
          loss='categorical_crossentropy',
          metrics=['accuracy'])

 model.summary()

 batch_size = 10

 # --- Fitting model and save weights
 # --- To use the augmented images for training, pass the train_generator into the fit() method of the model:
 # ---train the model---
 step_size_train = train_generator.n // train_generator.batch_size
 trained_FT = model.fit(train_generator,
                       steps_per_epoch = step_size_train,
                       epochs=10)

 # --- Create a function to plot the accuracy and the loss over the epochs.
 def plot_acc_loss(trained):
    fig, ax = plt.subplots(1, 2, figsize=(15,5))
    ax[0].set_title('loss')
    ax[0].plot(trained.epoch, trained.history["loss"], label="Train loss")
    #ax[0].plot(trained.epoch, trained.history["val_loss"], label="Validation loss")
    ax[1].set_title('acc')
    ax[1].plot(trained.epoch, trained.history["accuracy"], label="Train acc")
    #ax[1].plot(trained.epoch, trained.history["val_accuracy"], label="Validation acc")
    ax[0].legend()
    ax[1].legend()

 plot_acc_loss(trained_FT)
	# --- import the required packages
	import tensorflow
	from tensorflow.keras.preprocessing.image import ImageDataGenerator
	from tensorflow.keras.callbacks import EarlyStopping
	from tensorflow.keras.models import Model, Sequential
	from tensorflow.keras.layers import Conv2D, GlobalAveragePooling2D, MaxPool2D, Dense, Flatten, Dropout
	from tensorflow.keras.optimizers import Adam
	from tensorflow.keras.applications import VGG16
	from keras.preprocessing import image
	import os
	import pathlib
	import numpy as np
	import matplotlib.pyplot as plt

	# --- Data Information
	# --- cek jumlah images and classes dalam dataset.
	train_dir = pathlib.Path("/content/folderOnColab/myDataset/aug/train/")
	test_dir = pathlib.Path("/content/folderOnColab/myDataset/aug/test/")
	print("Terdapat ", len(list(train_dir.glob('/.jpg'))), "images di train set.")
	print("Terdapat ", len(list(test_dir.glob('/.jpg'))), "images di test set.")

	class_names = list([item.name for item in train_dir.glob('*')])
	print("Adapun classnya adalah:", class_names)

	image_generator = ImageDataGenerator()
	train_generator = image_generator.flow_from_directory(train_dir, class_mode='categorical')
	test_generator = image_generator.flow_from_directory(test_dir, class_mode='categorical')

	# --- number of patterns
	NO_CLASSES = max(train_generator.class_indices.values()) + 1
	print("Jumlah class (NO_CLASSES): ", NO_CLASSES)

	# --- Transfer Learning
	base_model = VGG16(include_top=False, weights='imagenet', input_shape=(224, 224, 3))
	print("Jumlah layers pada model basis VGG16: ", len(base_model.layers))

	# --- add our own custom layer
	x = base_model.output
	x = GlobalAveragePooling2D()(x)
	x = Dense(1024,activation='relu')(x) # add dense layers so
	# that the model can
	# learn more complex
	# functions and
	# classify for better
	# results.
	x = Dense(1024,activation='relu')(x) # dense layer 2
	x = Dense(512,activation='relu')(x) # dense layer 3
	preds = Dense(5,
	activation='softmax')(x) # final layer with
	# softmax activation
	# --- create a new model with the base model's original
	# input and the new model's output
	model = Model(inputs = base_model.input, outputs = preds)
	# --- don't train the first 19 layers - 0..18
	for layer in model.layers[:19]:
	layer.trainable=False
	# --- train the rest of the layers - 19 onwards
	for layer in model.layers[19:]:
	layer.trainable=True

	# --- compile the model
	model.compile(optimizer=Adam(learning_rate=2e-5),
	loss='categorical_crossentropy',
	metrics=['accuracy'])

	model.summary()

	batch_size = 10

	# --- Fitting model and save weights
	# --- To use the augmented images for training, pass the train_generator into the fit() method of the model:
	# ---train the model---
	step_size_train = train_generator.n // train_generator.batch_size
	trained_FT = model.fit(train_generator,
	steps_per_epoch = step_size_train,
	epochs=10)

	# --- Create a function to plot the accuracy and the loss over the epochs.
	def plot_acc_loss(trained):
	fig, ax = plt.subplots(1, 2, figsize=(15,5))
	ax[0].set_title('loss')
	ax[0].plot(trained.epoch, trained.history["loss"], label="Train loss")
	#ax[0].plot(trained.epoch, trained.history["val_loss"], label="Validation loss")
	ax[1].set_title('acc')
	ax[1].plot(trained.epoch, trained.history["accuracy"], label="Train acc")
	#ax[1].plot(trained.epoch, trained.history["val_accuracy"], label="Validation acc")
	ax[0].legend()
	ax[1].legend()

	plot_acc_loss(trained_FT)