def custom_forward()

in src/open-r1-multimodal/src/open_r1/grpo_rec.py [0:0]

• 32 lines of code
• 2 McCabe index (conditional complexity)

def custom_forward(
        self,
        hidden_states: torch.Tensor,
        cu_seqlens: torch.Tensor,
        rotary_pos_emb: Optional[torch.Tensor] = None,
        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,
    ) -> torch.Tensor:
        seq_length = hidden_states.shape[0]
        q, k, v = self.qkv(hidden_states).reshape(seq_length, 3, self.num_heads, -1).permute(1, 0, 2, 3).unbind(0)
        # print(111, 222, 333, 444, 555, 666, 777, 888, 999)
        if position_embeddings is None:
            logger.warning_once(
                "The attention layers in this model are transitioning from computing the RoPE embeddings internally "
                "through `rotary_pos_emb` (2D tensor of RoPE theta values), to using externally computed "
                "`position_embeddings` (Tuple of tensors, containing cos and sin). In v4.54 `rotary_pos_emb` will be "
                "removed and `position_embeddings` will be mandatory."
            )
            emb = torch.cat((rotary_pos_emb, rotary_pos_emb), dim=-1)
            cos = emb.cos().float()
            sin = emb.sin().float()
        else:
            cos, sin = position_embeddings
            # Add this
            cos = cos.to(torch.float)
            sin = sin.to(torch.float)
        q, k = apply_rotary_pos_emb_flashatt(q.unsqueeze(0), k.unsqueeze(0), cos, sin)
        q = q.squeeze(0)
        k = k.squeeze(0)

        max_seqlen = (cu_seqlens[1:] - cu_seqlens[:-1]).max().item()
        attn_output = flash_attn_varlen_func(q, k, v, cu_seqlens, cu_seqlens, max_seqlen, max_seqlen).reshape(
            seq_length, -1
        )
        attn_output = self.proj(attn_output)
        return attn_output