def main():
parser = get_command_line_parser()
args = parser.parse_args()
torch.manual_seed(args.seed)
os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu
use_gpu = torch.cuda.is_available()
if use_gpu:
print("Currently using GPU: {}".format(args.gpu))
torch.backends.cudnn.benchmark = True
torch.cuda.manual_seed_all(args.seed)
else:
print("Currently using CPU")
trainloader, testloader = get_mnist_data(train_batch_size=args.batch_size,
workers=args.workers)
print("Creating model: {}".format(args.model))
feature_extractor = ConvNet(depth=6, input_channel=1)
model = BaseLine(feature_extractor=feature_extractor,
num_base_class=10,
embed_size=2)
if use_gpu:
model = model.cuda()
# optimizer_model = torch.optim.SGD(model.parameters(), lr=args.lr_model, weight_decay=5e-04, momentum=0.9)
optimizer_model = torch.optim.Adam(model.parameters(), lr=args.lr_model)
if args.stepsize > 0:
scheduler = torch.optim.lr_scheduler.StepLR(optimizer_model,
step_size=args.stepsize,
gamma=args.gamma)
start_time = time.time()
for epoch in range(args.max_epoch):
print("==> Epoch {}/{}".format(epoch + 1, args.max_epoch))
train(model, optimizer_model, trainloader, use_gpu, 10, epoch, args)
if args.stepsize > 0:
scheduler.step()
if args.eval_freq > 0 and (epoch + 1) % args.eval_freq == 0 or (
epoch + 1) == args.max_epoch:
print("==> Test")
acc, err = evaluate(model,
testloader,
use_gpu,
10,
epoch,
args=args)
print("Accuracy (%): {}\t Error rate (%): {}".format(acc, err))
elapsed = round(time.time() - start_time)
elapsed = str(datetime.timedelta(seconds=elapsed))
print("Finished. Total elapsed time (h:m:s): {}".format(elapsed))
[MNIST Dataset](http://yann.lecun.com/exdb/mnist/). Author: Aymeric Damien Project: https://github.com/aymericdamien/TensorFlow-Examples/ """ from __future__ import print_function import tensorflow as tf from tensorflow.contrib import rnn import numpy as np import pdb # Import MNIST data from prepare_data import get_mnist_data mnist = get_mnist_data() ''' To classify images using a bidirectional recurrent neural network, we consider every image row as a sequence of pixels. Because MNIST image shape is 28*28px, we will then handle 28 sequences of 28 steps for every sample. ''' # Training Parameters learning_rate = 0.001 training_steps = 10000 batch_size = 128 display_step = 200 # Network Parameters num_input = 28 # MNIST data input (img shape: 28*28)
import os
import sys
import numpy as np
from prepare_data import get_mnist_data
from prepare_nn import create_dense_autoencoder
from misc_tools import tools_save_model, tools_load_model
from make_imgs import get_image_from_one_sample, get_images_from_dataset, del_all_images, make_html_page
(x_train, y_train), (x_test, y_test) = get_mnist_data()
filenames = ["nnets/autoencoder.json", "nnets/autoencoder.bin"]
autoencoder = tools_load_model(filenames[0], filenames[1])
encoder = autoencoder.get_layer("encoder")
decoder = autoencoder.get_layer("decoder")
autoencoder.compile(optimizer="adam", loss="binary_crossentropy")
part_size = 20
num_of_parts = 3
batch_size = part_size
row_heads = []
del_all_images()
for part_index in range(0, num_of_parts):
x_test_part = x_test[0 + part_index * part_size:part_size *
(part_index + 1)]