main_eval.sh无法运行，麻烦帮忙看看如何解决 #114
Description
使用main_eval.sh的时候，报错如下(ps:这里尝试在对应的参数前面添加‘+’无效)：
(/mnt/dolphinfs/hdd_pool/docker/user/hadoop-fsprisk/tools/env_bk/rl-factory) [hadoop-hmart-fsp-ml@set-zw04-mlp-codelab-pc1123 RL-Factory]$ bash /home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/function-call/RL-Factory/main_eval.sh
++ pwd
+ FILE=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/function-call/RL-Factory/verl/utils/reward_score/search.py
+ FUNCTION_NAME=compute_score
+ export MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
+ MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
+ export REWARD_MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
+ REWARD_MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
+ export TEST_DATA=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/data/nq_search/test.parquet
+ TEST_DATA=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/data/nq_search/test.parquet
+ tee grpo.log
+ python3 -m verl.trainer.main_evaluate data.val_files=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/data/nq_search/test.parquet data.val_batch_size=2048 data.max_prompt_length=4096 data.max_response_length=512 actor_rollout_ref.model.path=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100 actor_rollout_ref.model.use_remove_padding=True actor_rollout_ref.model.enable_gradient_checkpointing=True actor_rollout_ref.actor.ppo_mini_batch_size=256 actor_rollout_ref.actor.ppo_micro_batch_size_per_gpu=32 actor_rollout_ref.rollout.log_prob_micro_batch_size_per_gpu=32 actor_rollout_ref.rollout.tensor_model_parallel_size=1 actor_rollout_ref.rollout.name=vllm actor_rollout_ref.rollout.gpu_memory_utilization=0.9 actor_rollout_ref.rollout.max_turns=2 actor_rollout_ref.rollout.val_kwargs.temperature=0 actor_rollout_ref.rollout.val_kwargs.top_k=-1 actor_rollout_ref.rollout.val_kwargs.top_p=1 actor_rollout_ref.ref.log_prob_micro_batch_size_per_gpu=32 actor_rollout_ref.env.name=search actor_rollout_ref.env.mcp_mode=stdio actor_rollout_ref.env.tool_manager=null actor_rollout_ref.env.enable_thinking=False actor_rollout_ref.env.config_path=envs/configs/mcp_tools.pydata reward_rollout.if_use_reward_rollout=False reward_rollout.rollout.tensor_model_parallel_size=4 reward_rollout.rollout.gpu_memory_utilization=0.75 reward_rollout.rollout.model_name=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100 reward_rollout.rollout.free_cache_engine=False reward_rollout.rollout.response_length=2048 reward_model.reward_manager=parallel 'trainer.logger=[tensorboard]' trainer.project_name=GRPO_search trainer.experiment_name=search_with_thinking trainer.n_gpus_per_node=8 trainer.nnodes=1 trainer.val_only=True trainer.default_local_dir=ckpt trainer.default_hdfs_dir=null
Could not override 'actor_rollout_ref.rollout.max_turns'.
To append to your config use +actor_rollout_ref.rollout.max_turns=2
Key 'max_turns' is not in struct
full_key: actor_rollout_ref.rollout.max_turns
object_type=dict
Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
✅ MCP 警告屏蔽已启用
✅ MCP 警告屏蔽已启用对应的main_eval.sh如下：
source /opt/rh/devtoolset-9/enable
set -e -x
FILE="$(pwd)/verl/utils/reward_score/search.py"
FUNCTION_NAME="compute_score"
export MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
export REWARD_MODEL_PATH=/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/ckpt/global_step_100
export TEST_DATA='/home/hadoop-hmart-fsp-ml/hadoop-fsprisk/zhukeyu04/RL-Factory/data/nq_search/test.parquet'
# export VLLM_ATTENTION_BACKEND=XFORMERS
python3 -m verl.trainer.main_evaluate\
data.val_files=$TEST_DATA\
data.val_batch_size=2048\
data.max_prompt_length=4096\
data.max_response_length=512\
actor_rollout_ref.model.path=$MODEL_PATH\
actor_rollout_ref.model.use_remove_padding=True\
actor_rollout_ref.model.enable_gradient_checkpointing=True\
actor_rollout_ref.actor.ppo_mini_batch_size=256\
actor_rollout_ref.actor.ppo_micro_batch_size_per_gpu=32\
actor_rollout_ref.rollout.log_prob_micro_batch_size_per_gpu=32\
actor_rollout_ref.rollout.tensor_model_parallel_size=1\
actor_rollout_ref.rollout.name=vllm\
actor_rollout_ref.rollout.gpu_memory_utilization=0.9\
actor_rollout_ref.rollout.max_turns=2\
actor_rollout_ref.rollout.val_kwargs.temperature=0\
actor_rollout_ref.rollout.val_kwargs.top_k=-1\
actor_rollout_ref.rollout.val_kwargs.top_p=1\
actor_rollout_ref.ref.log_prob_micro_batch_size_per_gpu=32\
actor_rollout_ref.env.name=search\
actor_rollout_ref.env.mcp_mode=stdio\
actor_rollout_ref.env.tool_manager=null\
actor_rollout_ref.env.enable_thinking=False\
actor_rollout_ref.env.config_path=envs/configs/mcp_tools.pydata\
reward_rollout.if_use_reward_rollout=False\
reward_rollout.rollout.tensor_model_parallel_size=4\
reward_rollout.rollout.gpu_memory_utilization=0.75\
reward_rollout.rollout.model_name=$REWARD_MODEL_PATH\
reward_rollout.rollout.free_cache_engine=False\
reward_rollout.rollout.response_length=2048\
reward_model.reward_manager=parallel\
trainer.logger=['tensorboard']\
trainer.project_name='GRPO_search'\
trainer.experiment_name='search_with_thinking'\
trainer.n_gpus_per_node=8\
trainer.nnodes=1\
trainer.val_only=True\
trainer.default_local_dir=ckpt\
trainer.default_hdfs_dir=null $@ 2>&1 | tee grpo.log我的verl等版本如下：
(/mnt/dolphinfs/hdd_pool/docker/user/hadoop-fsprisk/tools/env_bk/rl-factory) [hadoop-hmart-fsp-ml@set-zw04-mlp-codelab-pc1123 RL-Factory]$ pip list
Package Version Editable project location
---------------------------------------- -------------- -------------------------------------------------------------------------------------
absl-py 2.3.0
accelerate 1.8.1
aiofiles 23.2.1
aiohappyeyeballs 2.6.1
aiohttp 3.12.13
aiohttp-cors 0.8.1
aiosignal 1.3.2
aiosqlite 0.21.0
airportsdata 20250622
annotated-types 0.7.0
anthropic 0.55.0
antlr4-python3-runtime 4.9.3
anyio 4.9.0
argon2-cffi 25.1.0
argon2-cffi-bindings 25.1.0
arrow 1.3.0
astor 0.8.1
asttokens 3.0.0
async-lru 2.0.5
async-timeout 5.0.1
attrs 25.3.0
av 14.4.0
babel 2.17.0
banks 2.2.0
beautifulsoup4 4.13.4
blake3 1.0.5
bleach 6.2.0
blessed 1.21.0
blinker 1.9.0
blobfile 3.0.0
boto3 1.38.21
botocore 1.38.46
bs4 0.0.2
cachebox 5.0.1
cachetools 5.5.2
certifi 2025.6.15
cffi 1.17.1
cfgv 3.4.0
charset-normalizer 3.4.2
click 8.2.1
cloudpickle 3.1.1
codetiming 1.4.0
colorama 0.4.6
colorful 0.5.6
comm 0.2.2
compressed-tensors 0.9.3
contourpy 1.3.2
cryptography 45.0.7
cuda-bindings 12.9.0
cuda-python 12.9.0
cupy-cuda12x 13.4.1
cycler 0.12.1
dashscope 1.24.4
dataclasses-json 0.6.7
datasets 3.6.0
debugpy 1.8.14
decorator 5.2.1
decord 0.6.0
defusedxml 0.7.1
Deprecated 1.2.18
depyf 0.18.0
dill 0.3.8
dirtyjson 1.0.8
diskcache 5.6.3
distlib 0.3.9
distro 1.9.0
dnspython 2.7.0
docker-pycreds 0.4.0
docopt 0.6.2
easydict 1.13
einops 0.8.1
email_validator 2.2.0
et_xmlfile 2.0.0
eval_type_backport 0.2.2
exceptiongroup 1.3.0
executing 2.2.0
faiss-gpu-cu12 1.12.0
fastapi 0.115.14
fastapi-cli 0.0.7
fastjsonschema 2.21.1
fastrlock 0.8.3
ffmpy 0.6.1
filelock 3.18.0
filetype 1.2.0
flash_attn 2.8.0.post2
flashinfer-python 0.2.5
Flask 3.1.2
fonttools 4.59.0
fqdn 1.5.1
frozenlist 1.7.0
fsspec 2025.3.0
gguf 0.17.1
gitdb 4.0.12
GitPython 3.1.44
google-api-core 2.25.1
google-auth 2.40.3
googleapis-common-protos 1.70.0
gpustat 1.1.1
gradio 5.23.1
gradio_client 1.8.0
greenlet 3.2.4
griffe 1.14.0
groovy 0.1.2
grpcio 1.73.1
h11 0.16.0
hf_transfer 0.1.9
hf-xet 1.1.5
hope 3.6.6
httpcore 1.0.9
httptools 0.6.4
httpx 0.27.2
httpx-sse 0.4.1
huggingface-hub 0.33.1
hydra-core 1.3.2
identify 2.6.12
idna 3.10
importlib_metadata 8.0.0
infinity_client 0.0.77
iniconfig 2.1.0
interegular 0.3.3
ipykernel 6.29.5
ipython 8.37.0
ipywidgets 8.1.7
isoduration 20.11.0
itsdangerous 2.2.0
jedi 0.19.2
jieba 0.42.1
Jinja2 3.1.6
jiter 0.10.0
jmespath 1.0.1
joblib 1.5.1
json5 0.12.1
jsonlines 4.0.0
jsonpointer 3.0.0
jsonschema 4.24.0
jsonschema-specifications 2025.4.1
jupyter 1.1.1
jupyter_client 8.6.3
jupyter-console 6.6.3
jupyter_core 5.8.1
jupyter-events 0.12.0
jupyter-lsp 2.2.6
jupyter_server 2.16.0
jupyter_server_terminals 0.5.3
jupyterlab 4.4.5
jupyterlab_pygments 0.3.0
jupyterlab_server 2.27.3
jupyterlab_widgets 3.0.15
kiwisolver 1.4.8
lark 1.2.2
latex2sympy2_extended 1.10.2
liger_kernel 0.5.10
litellm 1.73.2
llama-cloud 0.1.35
llama-cloud-services 0.6.54
llama-index 0.12.42
llama-index-agent-openai 0.4.12
llama-index-cli 0.4.4
llama-index-core 0.12.42
llama-index-embeddings-openai 0.3.1
llama-index-indices-managed-llama-cloud 0.8.0
llama-index-llms-openai 0.4.7
llama-index-multi-modal-llms-openai 0.5.1
llama-index-program-openai 0.3.2
llama-index-question-gen-openai 0.3.1
llama-index-readers-file 0.4.11
llama-index-readers-llama-parse 0.4.0
llama-parse 0.6.54
llguidance 0.7.30
llvmlite 0.44.0
lm-format-enforcer 0.10.11
lxml 6.0.0
Markdown 3.8.2
markdown-it-py 3.0.0
MarkupSafe 3.0.2
marshmallow 3.26.1
math-verify 0.8.0
mathruler 0.1.0
matplotlib 3.10.3
matplotlib-inline 0.1.7
mcp 1.9.3
mdurl 0.1.2
mistral_common 1.6.2
mistune 3.1.3
modelscope 1.27.1
modelscope_studio 1.1.7
mpmath 1.3.0
msgpack 1.1.1
msgspec 0.19.0
multidict 6.6.2
multiprocess 0.70.16
mypy_extensions 1.1.0
nanobind 2.7.0
nbclient 0.10.2
nbconvert 7.16.6
nbformat 5.10.4
nest-asyncio 1.6.0
networkx 3.4.2
ninja 1.11.1.4
nltk 3.9.1
nodeenv 1.9.1
notebook 7.4.5
notebook_shim 0.2.4
numba 0.61.2
numpy 1.26.4
nvidia-cublas-cu12 12.4.5.8
nvidia-cuda-cupti-cu12 12.4.127
nvidia-cuda-nvrtc-cu12 12.4.127
nvidia-cuda-runtime-cu12 12.4.127
nvidia-cudnn-cu12 9.1.0.70
nvidia-cufft-cu12 11.2.1.3
nvidia-curand-cu12 10.3.5.147
nvidia-cusolver-cu12 11.6.1.9
nvidia-cusparse-cu12 12.3.1.170
nvidia-cusparselt-cu12 0.6.2
nvidia-ml-py 12.575.51
nvidia-nccl-cu12 2.21.5
nvidia-nvjitlink-cu12 12.4.127
nvidia-nvtx-cu12 12.4.127
nvitop 1.5.3
omegaconf 2.3.0
openai 1.93.0
opencensus 0.11.4
opencensus-context 0.1.3
opencv-fixer 0.2.5
opencv-python 4.11.0.86
opencv-python-headless 4.11.0.86
openpyxl 3.1.5
opentelemetry-api 1.26.0
opentelemetry-exporter-otlp 1.26.0
opentelemetry-exporter-otlp-proto-common 1.26.0
opentelemetry-exporter-otlp-proto-grpc 1.26.0
opentelemetry-exporter-otlp-proto-http 1.26.0
opentelemetry-proto 1.26.0
opentelemetry-sdk 1.26.0
opentelemetry-semantic-conventions 0.47b0
opentelemetry-semantic-conventions-ai 0.4.9
optree 0.16.0
orjson 3.10.18
outlines 0.1.11
outlines_core 0.1.26
overrides 7.7.0
packaging 25.0
pandas 2.2.3
pandocfilters 1.5.1
parso 0.8.4
partial-json-parser 0.2.1.1.post6
pdfminer.six 20250506
pdfplumber 0.11.7
Pebble 5.1.3
peewee 3.18.2
peft 0.15.2
pexpect 4.9.0
pillow 11.2.1
pip 24.0
platformdirs 4.3.8
pluggy 1.6.0
pre_commit 4.2.0
prettytable 3.16.0
prometheus_client 0.22.1
prometheus-fastapi-instrumentator 7.1.0
prompt_toolkit 3.0.51
propcache 0.3.2
proto-plus 1.26.1
protobuf 6.32.1
psutil 7.0.0
ptyprocess 0.7.0
pure_eval 0.2.3
py-cpuinfo 9.0.0
py-spy 0.4.0
pyarrow 20.0.0
pyasn1 0.6.1
pyasn1_modules 0.4.2
pybind11 2.13.6
pycountry 24.6.1
pycparser 2.22
pycryptodomex 3.23.0
pydantic 2.9.2
pydantic_core 2.23.4
pydantic-extra-types 2.10.5
pydantic-settings 2.10.1
pydub 0.25.1
pyecharts 2.0.8
pyext 0.7
Pygments 2.19.2
pylatexenc 2.10
pymilvus 2.6.2
pynvml 12.0.0
pyparsing 3.2.3
pypdf 5.9.0
pypdfium2 4.30.0
pytest 8.3.5
python-dateutil 2.9.0.post0
python-docx 1.1.2
python-dotenv 1.1.0
python-json-logger 3.3.0
python-multipart 0.0.20
python-pptx 1.0.2
pytz 2025.2
PyYAML 6.0.2
pyzmq 26.4.0
qwen-agent 0.0.26 /mnt/dolphinfs/hdd_pool/docker/user/hadoop-fsprisk/zhukeyu04/tmp/Qwen-Agent-0.0.26
qwen-vl-utils 0.0.11
rank-bm25 0.2.2
ray 2.46.0
referencing 0.36.2
regex 2024.11.6
requests 2.32.3
rfc3339-validator 0.1.4
rfc3986-validator 0.1.1
rich 13.9.4
rich-toolkit 0.14.5
rl_factory 0.5.0 /mnt/dolphinfs/hdd_pool/docker/user/hadoop-fsprisk/zhukeyu04/function-call/RL-Factory
rpds-py 0.24.0
rsa 4.9.1
ruff 0.11.9
s3transfer 0.12.0
safehttpx 0.1.6
safetensors 0.5.3
sandbox-fusion 0.3.7
scikit-learn 1.6.1
scipy 1.15.3
seaborn 0.13.2
semantic-version 2.10.0
Send2Trash 1.8.3
sentencepiece 0.2.0
sentry-sdk 2.27.0
setproctitle 1.3.6
setuptools 69.5.1
sgl-kernel 0.1.1
sglang 0.4.6.post3
shellingham 1.5.4
simplejson 3.20.1
six 1.17.0
smart-open 7.1.0
smmap 5.0.2
sniffio 1.3.1
snowballstemmer 3.0.1
soundfile 0.13.1
soupsieve 2.7
SQLAlchemy 2.0.43
sse-starlette 2.3.4
stack-data 0.6.3
starlette 0.46.2
striprtf 0.0.26
swanboard 0.1.8b1
swankit 0.2.4
swanlab 0.6.8
symeval 0.0.1
sympy 1.13.1
tabulate 0.9.0
tenacity 9.1.2
tensorboard 2.19.0
tensorboard-data-server 0.7.2
tensordict 0.6.0
terminado 0.18.1
threadpoolctl 3.6.0
thrift 0.21.0
tiktoken 0.9.0
timeout-decorator 0.5.0
tinycss2 1.4.0
tokenizers 0.21.1
tomli 2.2.1
tomlkit 0.13.3
torch 2.6.0
torch_memory_saver 0.0.5
torchao 0.11.0
torchaudio 2.6.0
torchdata 0.10.0
torchvision 0.21.0
tornado 6.4.2
tqdm 4.67.1
traitlets 5.14.3
transformers 4.51.1
triton 3.2.0
typer 0.15.3
types-python-dateutil 2.9.0.20250809
typing_extensions 4.13.2
typing-inspect 0.9.0
typing-inspection 0.4.0
tzdata 2025.2
ujson 5.10.0
uri-template 1.3.0
urllib3 2.4.0
uvicorn 0.34.2
uvloop 0.21.0
virtualenv 20.31.2
vllm 0.8.5.post1
wandb 0.19.11
watchfiles 1.0.5
wcwidth 0.2.13
webcolors 24.11.1
webencodings 0.5.1
websocket-client 1.8.0
websockets 15.0.1
Werkzeug 3.1.3
wheel 0.43.0
widgetsnbextension 4.0.14
word2number 1.1
wrapt 1.17.2
xformers 0.0.29.post2
xgrammar 0.1.18
xlsxwriter 3.2.8
xxhash 3.5.0
yarl 1.20.0
zipp 3.21.0