def train_ch8(model, train_iter, vocab, lr, num_epochs, strategy, use_random_iter=False):
"""Train a model (defined in Chapter 8)."""
params = model.params # ---------------------------------------------------------------------------------------------------------
with strategy.scope():
loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
updater = tf.keras.optimizers.Adam(lr, amsgrad=True)
animator = d2l.Animator(xlabel='epoch', ylabel='perplexity',
legend=['train'], xlim=[10, num_epochs])
# def predict(self,vocab=vocab,prefix='generated ',num_tokens=50):
predict = lambda prefix: model.predict(vocab=vocab, prefix=prefix, num_tokens=50)
# Train and predict
for epoch in range(num_epochs):
ppl, speed = train_epoch_ch8(
model, train_iter, loss, updater, params, use_random_iter)
if (epoch + 1) % 10 == 0:
print(predict('time traveller'))
animator.add(epoch + 1, [ppl])
device = d2l.try_gpu()._device_name
print(f'perplexity {ppl:.1f}, {speed:.1f} tokens/sec on {str(device)}')
print(predict('time traveller'))
print(predict('traveller'))
device_name = d2l.try_gpu()._device_name
strategy = tf.distribute.OneDeviceStrategy(device_name)
num_epochs, num_hiddens, lr = 500, 512, 0.0001
model=RnnModelScrach(num_hiddens,len(vocab))
train_ch8(model, train_iter, vocab,lr, num_epochs, strategy)
def train(net, train_iter, test_iter, loss, num_epochs, updater):
animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs], ylim=[0.3, 0.9],
legend=['train loss', 'train acc', 'test acc'])
for epoch in range(num_epochs):
train_metrics = train_epoch(net, train_iter, loss, updater)
test_acc = evaluate_accuracy(net, test_iter)
animator.add(epoch + 1, train_metrics + (test_acc,))
train_loss, train_acc = train_metrics
def __init__(self, net, train_iter, test_iter, num_epochs, device_name):
self.timer = d2l.Timer()
self.animator = d2l.Animator(
xlabel='epoch', xlim=[0, num_epochs], legend=[
'train loss', 'train acc', 'test acc'])
self.net = net
self.train_iter = train_iter
self.test_iter = test_iter
self.num_epochs = num_epochs
self.device_name = device_name
self.loss_arr = []
self.train_acc_arr = []
self.test_acc_arr = []
def train_ch8(model, train_iter, vocab, num_hiddens, lr, num_epochs,
use_random_iter=False):
"""Train a model (defined in Chapter 8)."""
params = get_params(len(vocab), num_hiddens)
loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
animator = d2l.Animator(xlabel='epoch', ylabel='perplexity',
legend=['train'], xlim=[1, num_epochs])
updater = tf.keras.optimizers.SGD(lr)
predict = lambda prefix: predict_ch8(prefix, 50, model, vocab, params)
# Train and predict
for epoch in range(num_epochs):
ppl, speed = train_epoch_ch8(
model, train_iter, loss, updater, params, use_random_iter)
if epoch % 10 == 0:
print(predict('time traveller'))
animator.add(epoch + 1, [ppl])
device = d2l.try_gpu()._device_name
print(f'perplexity {ppl:.1f}, {speed:.1f} tokens/sec on {str(device)}')
print(predict('time traveller'))
print(predict('traveller'))