[Question]: ERNIE-ViL 2.0微调报错

[zwjwhxz]

请提出你的问题

按照实例方式，微调报错
环境：ubuntu20.04 python3.9
aiohttp 3.9.5
aiosignal 1.3.1
aistudio-sdk 0.2.4
annotated-types 0.7.0
anyio 3.7.1
asgiref 3.8.1
astor 0.8.1
async-timeout 4.0.3
attrs 23.2.0
Babel 2.15.0
bce-python-sdk 0.9.17
blinker 1.8.2
cachetools 6.0.0
certifi 2024.8.30
chardet 5.2.0
charset-normalizer 3.3.2
click 8.1.7
colorama 0.4.6
coloredlogs 15.0.1
colorlog 6.8.2
contourpy 1.2.1
cssselect 1.3.0
cssutils 2.11.1
cycler 0.12.1
dataclasses-json 0.6.7
datasets 2.20.0
decorator 5.1.1
dill 0.3.4
distro 1.9.0
dnspython 2.6.1
easydict 1.13
einops 0.8.1
email_validator 2.2.0
et_xmlfile 2.0.0
exceptiongroup 1.2.1
faiss-cpu 1.8.0.post1
fastapi 0.111.0
fastapi-cli 0.0.4
fastdeploy-gpu-python 1.0.7
fastdeploy-tools 0.0.5
filelock 3.15.4
Flask 3.0.3
flask-babel 4.0.0
flatbuffers 24.3.25
fonttools 4.53.0
frozenlist 1.4.1
fsspec 2024.5.0
ftfy 6.3.1
future 1.0.0
gast 0.3.3
gevent 24.2.1
GPUtil 1.4.0
greenlet 3.0.3
gunicorn 22.0.0
h11 0.14.0
hnswlib 0.8.0
httpcore 1.0.5
httptools 0.3.0
httpx 0.27.0
huggingface-hub 0.23.4
humanfriendly 10.0
idna 3.7
imagesize 1.4.1
importlib_metadata 8.0.0
importlib_resources 6.4.0
itsdangerous 2.2.0
jieba 0.42.1
Jinja2 3.1.4
jiter 0.10.0
joblib 1.4.2
jsonlines 4.0.0
jsonpatch 1.33
jsonpointer 3.0.0
kiwisolver 1.4.5
langchain 0.2.17
langchain-community 0.2.17
langchain-core 0.2.43
langchain-openai 0.1.25
langchain-text-splitters 0.2.4
langsmith 0.1.147
lxml 5.4.0
markdown-it-py 3.0.0
MarkupSafe 2.1.5
marshmallow 3.26.1
matplotlib 3.9.0
mdurl 0.1.2
more-itertools 10.7.0
mpmath 1.3.0
multidict 6.0.5
multiprocess 0.70.12.2
mypy_extensions 1.1.0
numpy 1.24.4
onnx 1.16.1
onnxruntime 1.18.1
openai 1.63.2
opencv-contrib-python 4.10.0.84
opencv-python 4.6.0.66
openpyxl 3.1.5
opt-einsum 3.3.0
orjson 3.10.6
packaging 24.1
paddle2onnx 1.3.1
paddleclas 2.6.0
paddlefsl 1.1.0
paddlenlp 2.8.1
paddleocr 3.0.0
paddlepaddle-gpu 2.6.1
paddleslim 2.6.0
paddlex 3.0.0
pandas 1.5.3
pillow 10.4.0
pip 24.2
premailer 3.10.0
prettytable 3.10.0
protobuf 5.27.2
psutil 6.0.0
py-cpuinfo 9.0.0
pyarrow 16.1.0
pyarrow-hotfix 0.6
pybind11 2.13.1
pyclipper 1.3.0.post6
pycryptodome 3.20.0
pydantic 2.8.0
pydantic_core 2.20.0
Pygments 2.18.0
pyparsing 3.1.2
pypdfium2 4.30.1
python-dateutil 2.9.0.post0
python-dotenv 1.0.1
python-multipart 0.0.9
pytz 2024.1
PyYAML 6.0.2
pyzmq 26.2.0
rarfile 4.2
regex 2024.5.15
requests 2.32.3
requests-toolbelt 1.0.0
rich 13.7.1
ruamel.yaml 0.18.11
ruamel.yaml.clib 0.2.12
safetensors 0.4.3
scikit-learn 1.5.0
scipy 1.13.1
sentencepiece 0.2.0
seqeval 1.2.2
setuptools 75.1.0
shapely 2.0.7
shellingham 1.5.4
six 1.16.0
sniffio 1.3.1
SQLAlchemy 2.0.41
starlette 0.37.2
swig 4.3.0
sympy 1.12.1
tenacity 8.5.0
threadpoolctl 3.5.0
tiktoken 0.9.0
timeout-decorator 0.5.0
tokenizers 0.19.1
tool-helpers 0.1.1
tqdm 4.66.4
typer 0.12.3
typing_extensions 4.12.2
typing-inspect 0.9.0
tzdata 2024.1
ujson 5.10.0
urllib3 2.2.2
uvicorn 0.16.0
uvloop 0.19.0
visualdl 2.5.3
watchfiles 0.22.0
watchgod 0.8.2
wcwidth 0.2.13
websockets 12.0
Werkzeug 3.0.3
wheel 0.44.0
xxhash 3.4.1
yarl 1.9.4
zipp 3.19.2
zope.event 5.0
zope.interface 6.4.post2

DATAPATH=..//traindata

data options

train_data=${DATAPATH}/arrow
val_data=${DATAPATH}/arrow
python run_finetune.py
--device gpu
--output_dir ./output/output_0731
--model_name_or_path ../baseModel/ernie_vil-2.0-base-zh
--train_data=${train_data}
--val_data=${val_data}
--do_train
--learning_rate 5e-5
--warmup_steps 100
--logging_steps 50
--per_device_train_batch_size 128
--dataloader_num_workers 8
--save_steps 50
--num_train_epochs 5
--weight_decay 0.001
--save_total_limit 50
--seed 1
--label_names index
--data_root $DATAPATH
--lr_scheduler_type cosine
--recompute

报错信息如下：
Traceback (most recent call last):
File "/appslog/homezwj/pycharmProject/clsLab/ERNIE_VIL/trainScript/run_finetune.py", line 117, in
do_train()
File "/appslog/homezwj/pycharmProject/clsLab/ERNIE_VIL/trainScript/run_finetune.py", line 107, in do_train
train_result = trainer.train(resume_from_checkpoint=checkpoint)
File "/appslog/miniconda3/envs/pEnv39/lib/python3.9/site-packages/paddlenlp/trainer/trainer.py", line 668, in train
train_dataloader = self.get_train_dataloader()
File "/appslog/miniconda3/envs/pEnv39/lib/python3.9/site-packages/paddlenlp/trainer/trainer.py", line 1387, in get_train_dataloader
train_sampler = self._get_train_sampler()
File "/appslog/miniconda3/envs/pEnv39/lib/python3.9/site-packages/paddlenlp/trainer/trainer.py", line 1206, in _get_train_sampler
return paddle.io.BatchSampler(
File "/appslog/miniconda3/envs/pEnv39/lib/python3.9/site-packages/paddle/io/dataloader/batch_sampler.py", line 128, in init
self.sampler = RandomSampler(dataset)
File "/appslog/miniconda3/envs/pEnv39/lib/python3.9/site-packages/paddle/io/dataloader/sampler.py", line 223, in init
raise ValueError(
ValueError: num_samples should be a positive integer, but got num_samples=0

微调数据共16类图片，共计3600多个文件，每个文件一行，请问这个报错如何解决。