Unverified Commit fafbe11a authored by Michael Yao's avatar Michael Yao Committed by GitHub
Browse files

[Docs] Fix griffe warnings in vllm/lora/ops (#25369)


Signed-off-by: default avatarwindsonsea <haifeng.yao@daocloud.io>
parent 78237e43
......@@ -83,8 +83,8 @@ class LoRAKernelMeta:
Prepare kernel metadata tensors for the current forward pass.
Args:
token_lora_tensor (torch.Tensor): Tensor containing lora indices
for each input token.
token_lora_mapping (torch.Tensor): Tensor containing lora indices
for each input token.
"""
self._reset()
......@@ -136,7 +136,7 @@ class LoRAKernelMeta:
Args:
token_nums (int): Number of input tokens in the current forward
pass.
pass of the kernel.
"""
return (
self.token_lora_mapping[:token_nums],
......
......@@ -93,7 +93,6 @@ def bgmv_shrink(
inputs (torch.Tensor): Input tensor of shape [num_tokens, hidden_size].
lora_b_weights (torch.Tensor): LoRA weights of shape
[num_loras, lora_rank, hidden_size].
output_tensor (torch.Tensor): (Unused) output tensor (placeholder).
lora_indices_tensor (torch.Tensor): Tensor of shape [num_tokens]
indicating which LoRA matrix to use for each token.
scaling (float, optional): Scalar multiplier applied to the output.
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment