ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

update minicpm to meet latest refactor (#11937)

Browse source
This commit is contained in:
SONG Ge 2024-08-27 15:08:01 +08:00 committed by GitHub
parent a81a329a5f
commit e211a5b076
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 6 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
python/llm/src/ipex_llm/transformers/npu_models/convert_mp.py
View file

@ -95,6 +95,12 @@ def optimize_llm(
from ipex_llm.transformers.npu_models.qwen2_mp import qwen2_casullm_forward from ipex_llm.transformers.npu_models.qwen2_mp import qwen2_casullm_forward
convert_forward(model, Qwen2ForCausalLM, qwen2_casullm_forward) convert_forward(model, Qwen2ForCausalLM, qwen2_casullm_forward)
elif model.config.model_type == "minicpm": elif model.config.model_type == "minicpm":
# for minicpm-1b
if intra_pp is None:
intra_pp = 2
if inter_pp is None:
inter_pp = 2
from ipex_llm.transformers.npu_models.minicpm_mp import gen_minicpm_fused_model_forward from ipex_llm.transformers.npu_models.minicpm_mp import gen_minicpm_fused_model_forward
from ipex_llm.transformers.npu_models.minicpm_mp import DecodeRunner, PrefillRunner from ipex_llm.transformers.npu_models.minicpm_mp import DecodeRunner, PrefillRunner