p12475398
/
LLaMA-Factory-Mirror
forked from p04798526/LLaMA-Factory-Mirror
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix #4242

This commit is contained in:
hiyouga 2024-06-12 16:50:11 +08:00
parent 656b2bbdaf
commit 577de2fa07
2 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Dockerfile
View File

@ -32,7 +32,7 @@ RUN EXTRA_PACKAGES="metrics"; \
EXTRA_PACKAGES="${EXTRA_PACKAGES},deepspeed"; \ EXTRA_PACKAGES="${EXTRA_PACKAGES},deepspeed"; \
fi; \ fi; \
pip install -e .[$EXTRA_PACKAGES] && \ pip install -e .[$EXTRA_PACKAGES] && \
pip uninstall -y transformer-engine pip uninstall -y transformer-engine flash-attn
# Set up volumes # Set up volumes
VOLUME [ "/root/.cache/huggingface/", "/app/data", "/app/output" ] VOLUME [ "/root/.cache/huggingface/", "/app/data", "/app/output" ]

5
src/llamafactory/chat/vllm_engine.py
View File

@ -13,10 +13,11 @@ from .base_engine import BaseEngine, Response
if is_vllm_available(): if is_vllm_available():
from vllm import AsyncEngineArgs, AsyncLLMEngine, RequestOutput, SamplingParams from vllm import AsyncEngineArgs, AsyncLLMEngine, RequestOutput, SamplingParams
from vllm.lora.request import LoRARequest from vllm.lora.request import LoRARequest
try: try:
from vllm.multimodal import MultiModalData # vllm==0.5.0 from vllm.multimodal import MultiModalData # type: ignore (for vllm>=0.5.0)
except ImportError: except ImportError:
from vllm.sequence import MultiModalData # vllm<0.5.0 from vllm.sequence import MultiModalData # for vllm<0.5.0
if TYPE_CHECKING: if TYPE_CHECKING: