batchA_images, batchB_images = img_process.shuffle_data(
trainA_paths, trainB_paths)
train_feed_dict = {
image_u: batchA_images,
image_r: batchB_images,
learning_rate: epoch_learning_rate,
training_flag: True
}
_, summary_str = sess.run([train_op, all_sum],
feed_dict=train_feed_dict)
train_summary_writer.add_summary(summary=summary_str,
global_step=counter)
# batch_loss = sess.run(errG, feed_dict=train_feed_dict)
# total_loss.append(batch_loss)
counter += 1
end_time = time.time()
# train_loss = np.mean(total_loss)
line = "epoch: %d/%d, time cost: %.4f\n" % (
epoch, total_epochs, float(end_time - start_time))
# line = "epoch: %d/%d, train loss: %.4f, time cost: %.4f\n" % (epoch, total_epochs, float(train_loss), float(end_time - start_time))
print(line)
if epoch % 10 == 0:
U_NET.save(sess=sess,
model_path=ckpt_path + str(epoch) + 'u_net.ckpt')
print('--------------Loaded Data----------------')
# Import Unet model
model = UNet()
# Train the model
history = model.fit(trainx_cropped,
trainy_hot,
validation_data=(validationx_cropped, validationy_hot),
epochs=50,
batch_size=16,
verbose=1)
# Save the model
model.save("/home/data/IIRS/PS1-Project/src/trained_model.h5")
print('--------------Training Completed----------------')
# list all data in history
print(history.history.keys())
# summarize history for accuracy
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('Model accuracy')
plt.ylabel('Accuracy')
plt.xlabel('Epoch')
plt.legend(['train', 'val'], loc='upper left')
plt.savefig('Accuracy_Plot.png')
###
print('--------------Loaded Data----------------')
# Import Unet model
model = UNet()
# Load trained weights
# model.load_weights(weights_file)
# Train the model
history = model.fit(trainx_cropped, trainy_hot, validation_data=(
validationx_cropped, validationy_hot), epochs=30, batch_size=16, verbose=1)
# Save the model
model.save("/home/data/IIRS/December-Testing/TrainedModels/trained_model_sar-v2.h5")
print('--------------Training Completed----------------')
# list all data in history
print(history.history.keys())
# summarize history for accuracy
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('Model accuracy')
plt.ylabel('Accuracy')
plt.xlabel('Epoch')
plt.legend(['train', 'val'], loc='upper left')
plt.savefig('Accuracy_Plot-SAR-2.png')
