apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-04 16:22:57 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Clear dead code And supplementary notes (#2757)
Some checks failed
Deploy GitHub Pages / deploy (push) Has been cancelled
Details

Browse Source 
* 1.supplementary notes 2.delete dead code

* fix bug of forward meta

* Global modification of forward meta

* fix vl model_runner bug
This commit is contained in:
RAM
2025-07-09 16:17:34 +08:00
committed by GitHub
parent b89180f1cd
commit 03a74995b8
12 changed files with 248 additions and 463 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
fastdeploy/worker/iluvatar_model_runner.py
View File

@@ -593,8 +593,23 @@ class IluvatarModelRunner(ModelRunnerBase):
Initialize forward meta and attention meta data
"""
# Initialize forward meta
self.forward_meta = ForwardMeta.init_forward_meta(
self.share_inputs, self.attn_backends[0])
self.forward_meta = ForwardMeta(
input_ids=self.share_inputs["input_ids"],
ids_remove_padding=self.share_inputs["ids_remove_padding"],
rotary_embs=self.share_inputs["rope_emb"],
attn_backend=self.attn_backends[0],
decoder_batch_ids=self.share_inputs["decoder_batch_ids"],
decoder_tile_ids_per_batch=self.share_inputs["decoder_tile_ids_per_batch"],
seq_lens_encoder=self.share_inputs["seq_lens_encoder"],
seq_lens_decoder=self.share_inputs["seq_lens_decoder"],
seq_lens_this_time=self.share_inputs["seq_lens_this_time"],
cum_offsets=self.share_inputs["cum_offsets"],
padding_offset=self.share_inputs["padding_offset"],
cu_seqlens_q=self.share_inputs["cu_seqlens_q"],
cu_seqlens_k=self.share_inputs["cu_seqlens_k"],
block_tables=self.share_inputs["block_tables"],
caches=self.share_inputs["caches"]
)
# Initialzie attention meta data
for attn_backend in self.attn_backends: