about Embedding

[Fyhyuky-FONTA]

class EmbeddingLayer ：
...
if len(sparse_emb) > 0:
sparse_exists = True
sparse_emb = torch.cat(sparse_emb, dim=1) # (batch_size, num_features, embed_dim)

    # Note: if the emb_dim of sparse features is different, we must squeeze_dim
    if squeeze_dim:  
        if dense_exists and not sparse_exists:  # only input dense features
            return dense_values
        elif not dense_exists and sparse_exists:
            return sparse_emb.flatten(start_dim=1)  # squeeze dim to : (batch_size, num_features*embed_dim)

如果sparse_emb = torch.cat(sparse_emb, dim=1) # (batch_size, num_features, embed_dim)能够正确执行，那么说明所有Embedding向量维度是相同的，为这么这里需要特别说明# Note: if the emb_dim of sparse features is different, we must squeeze_dim，作者团队是怎么设计的-