Upload model

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a813c542acf675006e2699e2a1070a166e99f9c1cb225ff613a76c0fda09df2
+oid sha256:3c6565fcbb6c375fc4b9c112ed3d73602e2741f45d6f7bea766a81c603e2f0be
 size 250204

modeling_transformer.py

@@ -71,7 +71,7 @@ def masked_softmax(X, valid_lens):
             valid_lens = valid_lens.reshape(-1)
         # 最后一轴上被掩蔽的元素使用一个非常大的负值替换，从而其softmax输出为0
         X = sequence_mask(X.reshape(-1, shape[-1]), valid_lens,
-                              value=-1e6)
+                              value=-1e4)
         return nn.functional.softmax(X.reshape(shape), dim=-1)
 
 class DotProductAttention(nn.Module):
@@ -411,4 +411,38 @@ class transformerModel(PreTrainedModel):
     def forward(self, enc_X, dec_X, *args):
         enc_outputs = self.encoder(enc_X, *args)
         dec_state = self.decoder.init_state(enc_outputs, *args)
-        return self.decoder(dec_X, dec_state)
+        return self.decoder(dec_X, dec_state)
+
+def predict_seq2seq(net, src_sentence, src_vocab, tgt_vocab, num_steps,
+                    device, save_attention_weights=False):
+    """序列到序列模型的预测
+
+    Defined in :numref:`sec_seq2seq_training`"""
+    # 在预测时将net设置为评估模式
+    net.eval()
+    src_tokens = src_vocab[src_sentence.lower().split(' ')] + [
+        src_vocab['<eos>']]
+    enc_valid_len = torch.tensor([len(src_tokens)], device=device)
+    src_tokens = d2l.truncate_pad(src_tokens, num_steps, src_vocab['<pad>'])
+    # 添加批量轴
+    enc_X = torch.unsqueeze(
+        torch.tensor(src_tokens, dtype=torch.long, device=device), dim=0)
+    enc_outputs = net.encoder(enc_X, enc_valid_len)
+    dec_state = net.decoder.init_state(enc_outputs, enc_valid_len)
+    # 添加批量轴
+    dec_X = torch.unsqueeze(torch.tensor(
+        [tgt_vocab['<bos>']], dtype=torch.long, device=device), dim=0)
+    output_seq, attention_weight_seq = [], []
+    for _ in range(num_steps):
+        Y, dec_state = net.decoder(dec_X, dec_state)
+        # 我们使用具有预测最高可能性的词元，作为解码器在下一时间步的输入
+        dec_X = Y.argmax(dim=2)
+        pred = dec_X.squeeze(dim=0).type(torch.int32).item()
+        # 保存注意力权重（稍后讨论）
+        if save_attention_weights:
+            attention_weight_seq.append(net.decoder.attention_weights)
+        # 一旦序列结束词元被预测，输出序列的生成就完成了
+        if pred == tgt_vocab['<eos>']:
+            break
+        output_seq.append(pred)
+    return ' '.join(tgt_vocab.to_tokens(output_seq)), attention_weight_seq