def train():
num_epochs, lr, batch_size = 5, 0.5, 256
loss = gloss.SoftmaxCrossEntropyLoss()
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
trainer = gluon.Trainer(net.collect_params(), 'sgd', {'learning_rate': lr})
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size,
params, lr)
def train_resnet_18():
net = residual_network()
net.initialize(force_reinit=True, init=init.Xavier())
lr, batch_size, ctx, epoch_num = 0.05, 256, mxnet.cpu(), 5
trainer = gluon.Trainer(net.collect_params(), 'sgd', {'learning_rate': lr})
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size,
resize=96)
d2l.train_ch5(net, train_iter, test_iter, batch_size, trainer, ctx,
epoch_num)
def train_small_vgg():
ratio = 4
small_vgg_11 = [(pair[0], pair[1] // ratio) for pair in vgg_11]
net = vgg(small_vgg_11)
lr, num_epochs, batch_size, ctx = 0.01, 20, 256, mx.cpu()
net.initialize()
# mx.gluon 训练器中，定义需要更新的参数 net.collect_params(), 更新算法 sgd, 以及学习率 lr
trainer = mx.gluon.Trainer(net.collect_params(), 'sgd',
{'learning_rate': lr})
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)
d2l.train_ch5(net, train_iter, test_iter, batch_size, trainer, ctx,
num_epochs)
'''
conv0 output shape: (1, 6, 24, 24)
pool0 output shape: (1, 6, 12, 12)
conv1 output shape: (1, 16, 8, 8)
pool1 output shape: (1, 16, 4, 4)
dense0 output shape: (1, 120)
dense1 output shape: (1, 84)
dense2 output shape: (1, 10)
Process finished with exit code 0
'''
# 5.5.2
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)
def try_gpu():
try:
ctx = mx.gpu()
_ = nd.zeros((1, ), ctx=ctx)
except mx.base.MXNetError:
ctx = mx.cpu()
return ctx
ctx = try_gpu()
def evaluate_accuracy(data_iter, net, ctx):
def train():
num_epochs, lr, batch_size = 5, 0.5, 256
loss = gloss.SoftmaxCrossEntropyLoss()
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size,
params, lr)