train_loss = 0
logging.info('Epoch {} train_loss {:.4f}'.format(i, train_loss))
for j in range(num_batches):
# logging.info("Epoch %d (%d/%d)" % (i+1, j+1, num_batches))
train_X_mb = train_X_shuffled[j * batch_size:(j + 1) *
batch_size] #mini-batch
train_y_mb = train_y_shuffled[j * batch_size:(j + 1) * batch_size]
for k in range(num_groups): # number_gropu = batch_size
train_X_mg = fixed(
train_X_mb[k * group_size:(k + 1) * group_size], 16,
FL_A_input
) # convert dataset to fixed-point, 16+FL_A_input
train_y_mg = train_y_mb[k * group_size:(k + 1) * group_size]
predictions, loss = cnn.feed_forward(train_X_mg,
train_y_mg,
train_or_test=1)
# logging.info('Epoch {} Batch {} Loss {:.4f}'.format(i, j, loss))
cnn.feed_backward()
cnn.weight_gradient()
# import pdb; pdb.set_trace()
if k == num_groups - 1:
cnn.apply_weight_gradients(Learning_Rate, args.momentum,
batch_size, True)
else:
cnn.apply_weight_gradients(Learning_Rate, args.momentum,
batch_size, False)
#import pdb; pdb.set_trace()
wrong_predictions += torch.sum(
predictions.cpu() != train_y_mg).numpy()
train_loss += loss
# Testing
logging.info('loading trained weights...')
cnn.load_params_mat(
'./result/result_{}classes/Best_epoch_CIFAR10_W.mat'.format(
task_division[0]))
# cnn.load_params_mat('./result/test_W.mat')
# print('test_W')
batch_size_valid = 40
num_batches_valid = int(cloud_image_valid / batch_size_valid)
valid_error = 0.
valid_loss = 0.
for j in range(num_batches_valid): # testing
predictions, valid_loss_batch = cnn.feed_forward(
fixed(
valid_cloud_x[j * batch_size_valid:(j + 1) * batch_size_valid],
16, FL_A_input),
valid_cloud_y[j * batch_size_valid:(j + 1) * batch_size_valid],
train_or_test=0)
valid_error += torch.sum(
predictions.cpu() != valid_cloud_y[j * batch_size_valid:(j + 1) *
batch_size_valid]).numpy()
valid_loss += valid_loss_batch
valid_error /= cloud_image_valid
valid_loss /= num_batches_valid
valid_acc = (100 - (valid_error * 100))
logging.info(" valid accuracy: %.2f%%" % valid_acc)
logging.info(" valid loss: %.4f" % valid_loss)
threshold = task_division[0] / 10.0
print('\n Generating mask for top %.3f params' % threshold)
W = sio.loadmat(
logging.info('Epoch {} train_loss {:.4f}'.format(i, train_loss))
print('\nTraining.....Mask applied\n') # weight update function
for j in range(int(num_batches)):
# logging.info("Epoch %d (%d/%d)" % (i+1, j+1, num_batches))
train_X_mb = train_X_shuffled[j * batch_size:(j + 1) *
batch_size] # mini-batch
train_y_mb = train_y_shuffled[j * batch_size:(j + 1) * batch_size]
for k in range(num_groups): # number_gropu = batch_size
train_X_mg = fixed(
train_X_mb[k * group_size:(k + 1) * group_size], 16,
FL_A_input
) # convert dataset to fixed-point, 16+FL_A_input
train_y_mg = train_y_mb[k * group_size:(k + 1) * group_size]
predictions, loss = cnn.feed_forward(train_X_mg,
train_y_mg,
train_or_test=1)
# logging.info('Epoch {} Batch {} Loss {:.4f}'.format(i, j, loss))
cnn.feed_backward()
cnn.weight_gradient()
if k == num_groups - 1:
cnn.apply_weight_gradients(Learning_Rate, args.momentum,
batch_size, True, mask)
else:
cnn.apply_weight_gradients(Learning_Rate, args.momentum,
batch_size, False, mask)
# import pdb; pdb.set_trace()
wrong_predictions += torch.sum(
predictions.cpu() != train_y_mg).numpy()