From ef4d4548a530360d4d1330b3392c0f0d024f3f2b Mon Sep 17 00:00:00 2001 From: Zack Date: Tue, 30 Apr 2024 21:31:11 -0500 Subject: [PATCH] fix: set profile in backgroud in setup --- servers/qwen/sky_serve.yaml | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/servers/qwen/sky_serve.yaml b/servers/qwen/sky_serve.yaml index afccc29..090f98c 100644 --- a/servers/qwen/sky_serve.yaml +++ b/servers/qwen/sky_serve.yaml @@ -41,10 +41,10 @@ setup: | git clone --depth=1 https://github.com/ZackBradshaw/lmdeploy && \ cd lmdeploy && \ pip install -r requirements.txt && \ - cd benchmark - python3 profile_generation.py $MODEL_NAME && \ + cd benchmark && \ + nohup python3 profile_generation.py $MODEL_NAME > profiling.log 2>&1 & \ wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json && \ - python3 profile_restful_api.py http://0.0.0.0:8000 internlm/internlm-7b ./ShareGPT_V3_unfiltered_cleaned_split.json + nohup python3 profile_restful_api.py http://0.0.0.0:8000 internlm/internlm-7b ./ShareGPT_V3_unfiltered_cleaned_split.json > restful_api.log 2>&1 & run: | #### LM_Deploy with lmdeploy https://lmdeploy.readthedocs.io/en/latest/get_started.html#installation @@ -52,9 +52,9 @@ run: | #### LM_Deploy with lmdeploy https://lmdeploy.readthedocs.io/en/latest/ge docker run --runtime nvidia --gpus all \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HUGGING_FACE_HUB_TOKEN={$HUGGING_FACE_HUB_TOKEN}"\ - -p 8080:8080 \ + -p 8080:8080 -p 8000:8000 \ --ipc=host \ openmmlab/lmdeploy:latest \ - pip install lmdeploy[vision] \ + pip install lmdeploy[vision] && \ lmdeploy serve api_server $MODEL_NAME --server-port 8080