loss = criterion(output, category_tensor)
optimizer.zero_grad()
loss.backward()
optimizer.step()
return output, loss.item()
current_loss = 0
all_losses = []
plot_steps, print_steps = 1000, 5000
n_iter = 100000
for i in range(n_iter):
category, line, category_tensor, line_tensor = random_training_example(
category_lines, all_categories)
output, loss = train(line_tensor, category_tensor)
current_loss += loss
if (i + 1) % plot_steps == 0:
all_losses.append(current_loss / plot_steps)
current_loss = 0
if (i + 1) % print_steps == 0:
guess = category_from_output(output)
correct = "CORRECT" if guess == category else f"WRONG ({category})"
print(f"{i} {i / n_iter * 100} {loss:.4f} {line}/ {guess} {correct}")
plt.figure()
plt.plot(all_losses)
loss = 0
# sum of losses for an epoch
total_loss = 0
# now!
start = time.time()
print()
print("----------------")
print("training starts!")
print("----------------")
print()
for i in range(1, n_iter + 1):
output, loss = train(*random_training_example())
total_loss += loss
if i % print_every == 0:
avg_loss = total_loss / plot_every
sys.stdout.write(
"%d %d%% (%s) %.4f\n" %
(i, i / n_iter * 100, time_since(start), avg_loss))
losses.append(avg_loss)
total_loss = 0
plt.figure()
plt.plot(losses)
plt.show()
sys.stdout.write("train finished\n")