XLM/src/model/transformer.py [180:215]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        mask_reshape = (bs, 1, qlen, klen) if mask.dim() == 3 else (bs, 1, 1, klen)

        def shape(x):
            """  projection """
            return x.view(bs, -1, self.n_heads, dim_per_head).transpose(1, 2)

        def unshape(x):
            """  compute context """
            return x.transpose(1, 2).contiguous().view(bs, -1, self.n_heads * dim_per_head)

        q = shape(self.q_lin(input))                                          # (bs, n_heads, qlen, dim_per_head)
        if kv is None:
            k = shape(self.k_lin(input))                                      # (bs, n_heads, qlen, dim_per_head)
            v = shape(self.v_lin(input))                                      # (bs, n_heads, qlen, dim_per_head)
        elif cache is None or self.layer_id not in cache:
            k = v = kv
            k = shape(self.k_lin(k))                                          # (bs, n_heads, qlen, dim_per_head)
            v = shape(self.v_lin(v))                                          # (bs, n_heads, qlen, dim_per_head)

        if cache is not None:
            if self.layer_id in cache:
                if kv is None:
                    k_, v_ = cache[self.layer_id]
                    k = torch.cat([k_, k], dim=2)                             # (bs, n_heads, klen, dim_per_head)
                    v = torch.cat([v_, v], dim=2)                             # (bs, n_heads, klen, dim_per_head)
                else:
                    k, v = cache[self.layer_id]
            cache[self.layer_id] = (k, v)

        q = q / math.sqrt(dim_per_head)                                       # (bs, n_heads, qlen, dim_per_head)
        scores = torch.matmul(q, k.transpose(2, 3))                           # (bs, n_heads, qlen, klen)
        mask = (mask == 0).view(mask_reshape).expand_as(scores)               # (bs, n_heads, qlen, klen)
        scores.masked_fill_(mask, -float('inf'))                              # (bs, n_heads, qlen, klen)

        weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
        weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



pytorch-transformers/pytorch_transformers/modeling_xlm.py [311:346]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        mask_reshape = (bs, 1, qlen, klen) if mask.dim() == 3 else (bs, 1, 1, klen)

        def shape(x):
            """  projection """
            return x.view(bs, -1, self.n_heads, dim_per_head).transpose(1, 2)

        def unshape(x):
            """  compute context """
            return x.transpose(1, 2).contiguous().view(bs, -1, self.n_heads * dim_per_head)

        q = shape(self.q_lin(input))                                          # (bs, n_heads, qlen, dim_per_head)
        if kv is None:
            k = shape(self.k_lin(input))                                      # (bs, n_heads, qlen, dim_per_head)
            v = shape(self.v_lin(input))                                      # (bs, n_heads, qlen, dim_per_head)
        elif cache is None or self.layer_id not in cache:
            k = v = kv
            k = shape(self.k_lin(k))                                          # (bs, n_heads, qlen, dim_per_head)
            v = shape(self.v_lin(v))                                          # (bs, n_heads, qlen, dim_per_head)

        if cache is not None:
            if self.layer_id in cache:
                if kv is None:
                    k_, v_ = cache[self.layer_id]
                    k = torch.cat([k_, k], dim=2)                             # (bs, n_heads, klen, dim_per_head)
                    v = torch.cat([v_, v], dim=2)                             # (bs, n_heads, klen, dim_per_head)
                else:
                    k, v = cache[self.layer_id]
            cache[self.layer_id] = (k, v)

        q = q / math.sqrt(dim_per_head)                                       # (bs, n_heads, qlen, dim_per_head)
        scores = torch.matmul(q, k.transpose(2, 3))                           # (bs, n_heads, qlen, klen)
        mask = (mask == 0).view(mask_reshape).expand_as(scores)               # (bs, n_heads, qlen, klen)
        scores.masked_fill_(mask, -float('inf'))                              # (bs, n_heads, qlen, klen)

        weights = F.softmax(scores.float(), dim=-1).type_as(scores)           # (bs, n_heads, qlen, klen)
        weights = F.dropout(weights, p=self.dropout, training=self.training)  # (bs, n_heads, qlen, klen)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -