mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2026-04-23 00:17:25 +08:00
@@ -1381,7 +1381,7 @@ class GPUModelRunner(ModelRunnerBase):
|
||||
# Set forward_meta.is_dummy_or_profile_run to True to skip init_kv_signal_per_query for attention backends
|
||||
self.forward_meta.is_dummy_or_profile_run = is_dummy_or_profile_run
|
||||
|
||||
# Initialzie attention meta data
|
||||
# Initialize attention meta data
|
||||
for attn_backend in self.attn_backends:
|
||||
attn_backend.init_attention_metadata(self.forward_meta)
|
||||
|
||||
@@ -2924,7 +2924,7 @@ class GPUModelRunner(ModelRunnerBase):
|
||||
|
||||
def extract_vision_features_ernie(self, vision_inputs: dict[str, list[paddle.Tensor]]) -> paddle.Tensor:
|
||||
"""
|
||||
vision feature extactor for ernie-vl
|
||||
vision feature extractor for ernie-vl
|
||||
"""
|
||||
assert len(vision_inputs["images_lst"]) > 0, "at least one image needed"
|
||||
|
||||
|
||||
Reference in New Issue
Block a user