BERT_SentenceTransformer_self.sbert_model = sentencetransformer("cardiffnlp

1.总体结构理解：

第一步：生成 word_embedding

word_embedding_model = models.Transformer(model_name)

第二步：池化 pooling

pooling_model = models.Pooling(word_embedding_model.get_word_embedding_dimension(),
                                pooling_mode_mean_tokens=True,
                                pooling_mode_cls_token=False,
                                pooling_mode_max_tokens=False

第三步：句子相似度比较模型

model = SentenceTransformer(modules=[word_embedding_model, pooling_model])

第四步：数据形式InputExample

csv文件里面的数据为：

inp_example = InputExample(texts=[texts=[row['sentence1'], row['sentence2']]], label=score)

其中，InputExample是一个普通的类，他有两属性texts和label， texts就是两个句子， label就是score

困惑点1：下面这段代码已经完全将csv文件里面的内容读取到内存了，在使用DataLoader的作用适合，仅仅只是为了获取做batch分割和数据顺序扰乱吗？

train_samples = []
dev_samples = []
test_samples = []
with gzip.open(sts_dataset_path, 'rt', encoding='utf-8') as fIn:
    reader = csv.DictReader(fIn, delimiter='\t', quoting=csv.QUOTE_NONE)
    for row in reader:
        score = float(row['score']) / 5.0
        inp_example = InputExample(texts=[row['sentence1'], row['sentence2']], label=score)
 
        if row['split'] == 'dev':
            dev_samples.append(inp_example)
        
        elif row['split'] == 'test':
            test_samples.append(inp_example)
        
        else:
            train_samples.append(inp_example)
train_dataloader = DataLoader(train_samples, shuffle=True, batch_size=train_batch_size)   

第五步：相似度损失CosineSimilarityLoss

train_loss = losses.CosineSimilarityLoss(model=model）

第六步：训练，evaluator是训练过程的评估器， 暂时不予考虑

model.fit(train_objectives=[(train_dataloader, train_loss)],
            evaluator=evaluator,
            epochs=num_epochs,
            evaluation_steps=1000,
            warmup_steps=warmup_steps,
            output_path=model_save_path
        )

2.结构细节理解：

代码的关键文件：

configuration_auto.py

AutoConfig类：用于加载模型参数

CONFIG_MAPPING: 这是一个常量dict，用于定义选择模型

2.1 word_embedding_model的结构：

2.2 CosineSimilarityLoss的结构：

训练的这个loss这个模型， 这里的loss是一个model， 不仅仅是计算l损失的功能

CosineSimilarityLoss(
  (model): SentenceTransformer(
    (0): Transformer(
      (auto_model): DistilBertModel(
        (embeddings): Embeddings(
          (word_embeddings): Embedding(30522, 768, padding_idx=0)
          (position_embeddings): Embedding(512, 768)
          (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
          (dropout): Dropout(p=0.1, inplace=False)
        )
        (transformer): Transformer(
          (layer): ModuleList(
            (0): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
            (1): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
            (2): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
            (3): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
            (4): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
            (5): TransformerBlock(
              (attention): MultiHeadSelfAttention(
                (dropout): Dropout(p=0.1, inplace=False)
                (q_lin): Linear(in_features=768, out_features=768, bias=True)
                (k_lin): Linear(in_features=768, out_features=768, bias=True)
                (v_lin): Linear(in_features=768, out_features=768, bias=True)
                (out_lin): Linear(in_features=768, out_features=768, bias=True)
              )
              (sa_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (ffn): FFN(
                (dropout): Dropout(p=0.1, inplace=False)
                (lin1): Linear(in_features=768, out_features=3072, bias=True)
                (lin2): Linear(in_features=3072, out_features=768, bias=True)
              )
              (output_layer_norm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            )
          )
        )
      )
    )
    (1): Pooling()
  )
  (loss_fct): MSELoss()
  (cos_score_transformation): Identity()
)