中文文本分类 pytorch实现--学习笔记_train loss: 0.02191843092441559 at step: 1600-优快云博客

from tensorboardX import SummaryWriter

writer=SummaryWriter(comment='TextCNN')
dummy_input = torch.rand(4762,300)
writer.add_graph(model, (dummy_input,))

TextCNN ---epoc=20 结果

Loading data...
Vocab size: 4762
180000it [00:00, 241581.78it/s]
10000it [00:00, 280066.51it/s]
10000it [00:00, 281899.90it/s]
Time usage: 0:00:01
<bound method Module.parameters of Model(
(embedding): Embedding(4762, 300)
(convs): ModuleList(
    (0): Conv2d(1, 256, kernel_size=(2, 300), stride=(1, 1))
    (1): Conv2d(1, 256, kernel_size=(3, 300), stride=(1, 1))
    (2): Conv2d(1, 256, kernel_size=(4, 300), stride=(1, 1))
)
(dropout): Dropout(p=0.5, inplace=False)
(fc): Linear(in_features=768, out_features=10, bias=True)
)>
Epoch [1/20]
Iter:      0, Train Loss:   2.3, Train Acc: 11.72%, Val Loss:   2.7, Val Acc: 10.00%, Time: 0:00:01 *
Iter:    100, Train Loss: 0.77, Train Acc: 72.66%, Val Loss: 0.69, Val Acc: 78.69%, Time: 0:00:02 *
Iter:    200, Train Loss: 0.71, Train Acc: 73.44%, Val Loss: 0.55, Val Acc: 83.29%, Time: 0:00:02 *
Iter:    300, Train Loss: 0.45, Train Acc: 85.16%, Val Loss: 0.49, Val Acc: 84.73%, Time: 0:00:03 *
Iter:    400, Train Loss: 0.74, Train Acc: 80.47%, Val Loss: 0.47, Val Acc: 85.22%, Time: 0:00:03 *
Iter:    500, Train Loss: 0.39, Train Acc: 88.28%, Val Loss: 0.43, Val Acc: 86.33%, Time: 0:00:03 *
Iter:    600, Train Loss: 0.49, Train Acc: 83.59%, Val Loss: 0.42, Val Acc: 86.65%, Time: 0:00:04 *
Iter:    700, Train Loss: 0.43, Train Acc: 82.81%, Val Loss:   0.4, Val Acc: 87.32%, Time: 0:00:04 *
Iter:    800, Train Loss: 0.45, Train Acc: 86.72%, Val Loss:   0.4, Val Acc: 87.88%, Time: 0:00:05 *
Iter:    900, Train Loss: 0.43, Train Acc: 86.72%, Val Loss: 0.39, Val Acc: 88.09%, Time: 0:00:05 *
Iter:   1000, Train Loss: 0.33, Train Acc: 88.28%, Val Loss: 0.38, Val Acc: 88.07%, Time: 0:00:06 *
Iter:   1100, Train Loss: 0.39, Train Acc: 89.84%, Val Loss: 0.37, Val Acc: 88.54%, Time: 0:00:06 *
Iter:   1200, Train Loss: 0.37, Train Acc: 87.50%, Val Loss: 0.37, Val Acc: 88.73%, Time: 0:00:07 *
Iter:   1300, Train Loss: 0.41, Train Acc: 85.16%, Val Loss: 0.36, Val Acc: 88.74%, Time: 0:00:07 *
Iter:   1400, Train Loss: 0.45, Train Acc: 85.94%, Val Loss: 0.35, Val Acc: 89.03%, Time: 0:00:07 *
Epoch [2/20]
Iter:   1500, Train Loss: 0.37, Train Acc: 89.06%, Val Loss: 0.35, Val Acc: 89.08%, Time: 0:00:08 *
Iter:   1600, Train Loss: 0.32, Train Acc: 92.19%, Val Loss: 0.35, Val Acc: 89.16%, Time: 0:00:08 *
Iter:   1700, Train Loss:   0.4, Train Acc: 89.06%, Val Loss: 0.35, Val Acc: 89.60%, Time: 0:00:09 *
Iter:   1800, Train Loss: 0.34, Train Acc: 90.62%, Val Loss: 0.36, Val Acc: 88.76%, Time: 0:00:09
Iter:   1900, Train Loss: 0.41, Train Acc: 87.50%, Val Loss: 0.34, Val Acc: 89.43%, Time: 0:00:10 *
Iter:   2000, Train Loss: 0.38, Train Acc: 83.59%, Val Loss: 0.34, Val Acc: 89.36%, Time: 0:00:10
Iter:   2100, Train Loss: 0.38, Train Acc: 90.62%, Val Loss: 0.34, Val Acc: 89.47%, Time: 0:00:11 *
Iter:   2200, Train Loss: 0.25, Train Acc: 89.84%, Val Loss: 0.34, Val Acc: 89.75%, Time: 0:00:11 *
Iter:   2300, Train Loss:   0.3, Train Acc: 92.19%, Val Loss: 0.34, Val Acc: 89.73%, Time: 0:00:11
Iter:   2400, Train Loss: 0.25, Train Acc: 93.75%, Val Loss: 0.33, Val Acc: 89.77%, Time: 0:00:12 *
Iter:   2500, Train Loss: 0.15, Train Acc: 95.31%, Val Loss: 0.33, Val Acc: 90.32%, Time: 0:00:12 *
Iter:   2600, Train Loss: 0.39, Train Acc: 86.72%, Val Loss: 0.33, Val Acc: 89.86%, Time: 0:00:13
Iter:   2700, Train Loss: 0.27, Train Acc: 89.06%, Val Loss: 0.33, Val Acc: 89.83%, Time: 0:00:13
Iter:   2800, Train Loss: 0.34, Train Acc: 90.62%, Val Loss: 0.33, Val Acc: 89.99%, Time: 0:00:14
Epoch [3/20]
Iter:   2900, Train Loss: 0.32, Train Acc: 91.41%, Val Loss: 0.32, Val Acc: 90.19%, Time: 0:00:14 *
Iter:   3000, Train Loss: 0.27, Train Acc: 92.19%, Val Loss: 0.33, Val Acc: 90.11%, Time: 0:00:15
Iter:   3100, Train Loss: 0.26, Train Acc: 92.97%, Val Loss: 0.33, Val Acc: 89.99%, Time: 0:00:15
Iter:   3200, Train Loss:   0.4, Train Acc: 89.06%, Val Loss: 0.33, Val Acc: 90.22%, Time: 0:00:15
Iter:   3300, Train Loss:   0.3, Train Acc: 92.19%, Val Loss: 0.33, Val Acc: 90.29%, Time: 0:00:16
Iter:   3400, Train Loss: 0.31, Train Acc: 90.62%, Val Loss: 0.33, Val Acc: 90.14%, Time: 0:00:16
Iter:   3500, Train Loss:   0.2, Train Acc: 92.97%, Val Loss: 0.33, Val Acc: 90.30%, Time: 0:00:17
Iter:   3600, Train Loss: 0.16, Train Acc: 94.53%, Val Loss: 0.33, Val Acc: 90.31%, Time: 0:00:17
Iter:   3700, Train Loss: 0.35, Train Acc: 86.72%, Val Loss: 0.33, Val Acc: