FileNotFoundError

[SidneyRey]

Thank you for your open-source data!
I encountered an issue while testing my self-developed document parsing. The configuration is as follows:
end2end_eval:
metrics:
text_block:
metric:
- Edit_dist
- BLEU
- METEOR
display_formula:
metric:
- Edit_dist
- CDM
table:
metric:
- TEDS
- Edit_dist
reading_order:
metric:
- Edit_dist
dataset:
dataset_name: end2end_dataset
ground_truth:
# data_path: ./demo_data/omnidocbench_demo/OmniDocBench_demo.json
data_path: ./demo_data/OmniDocBench/OmniDocBench.json
prediction:
data_path: ./demo_data/hh_OmniDoc_res
match_method: quick_match
# filter:
# language: english

The error message is as follows:
raise FileNotFoundError(
FileNotFoundError: Couldn't find a module script at /data/source/ocr/OmniDocBench/bleu/bleu.py. Module 'bleu' doesn't exist on the Hugging Face Hub either.

Could you please advise on how to resolve this?site-packages/evaluate/loading.py", line 681, in evaluation_module_factory

[ouyanglinke]

Hi, this issue might be caused by a network proxy. Make sure you can connect to HuggingFace. Here is an issue that might be similar to yours: huggingface/evaluate#456

[SidneyRey]

thank you!

[Rezen1997]

thank you!
Hello, have you solved it yet? I still haven't solved the problem by implementing the following method. Can you explain how you did it：
git clone https://github.com/huggingface/evaluate.git
bleu = evaluate.load("evaluate/metrics/bleu/bleu.py", keep_in_memory=True, experiment_id=random.randint(1,1e8))

It will still report an error:
Traceback (most recent call last):
File "E:\OminiBeceh\OmniDocBench-main\pdf_validation.py", line 54, in
val_task(val_dataset, metrics_list, cfg[task]['dataset']['ground_truth']['data_path'], save_name) # 按页面区分
File "E:\OminiBeceh\OmniDocBench-main\task\end2end_run_eval.py", line 32, in init
samples, result_s = metric_val(samples).evaluate(group_info, f"{save_name}_{element}")
File "E:\OminiBeceh\OmniDocBench-main\metrics\cal_metric.py", line 100, in evaluate
bleu = evaluate.load("evaluate/metrics/bleu/bleu.py", keep_in_memory=True, experiment_id=random.randint(1,1e8))
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\loading.py", line 748, in load
evaluation_module = evaluation_module_factory(
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\loading.py", line 616, in evaluation_module_factory
return LocalEvaluationModuleFactory(
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\loading.py", line 421, in get_module
local_imports = _download_additional_modules(
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\loading.py", line 248, in _download_additional_modules
local_import_path = cached_path(
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\utils\file_utils.py", line 175, in cached_path
output_path = get_from_cache(
File "E:\OminiBeceh\OmniDocBench-main\venv\lib\site-packages\evaluate\utils\file_utils.py", line 514, in get_from_cache
raise ConnectionError(f"Couldn't reach {url} ({repr(head_error)})")
ConnectionError: Couldn't reach https://github.com/tensorflow/nmt/raw/master/nmt/scripts/bleu.py (ConnectionError(MaxRetryError('HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /tens
orflow/nmt/master/nmt/scripts/bleu.py (Caused by NameResolutionError("<urllib3.connection.HTTPSConnection object at 0x00000135F3AE7700>: Failed to resolve 'raw.githubusercontent.com' ([Errno 11004] getaddrinfo failed)"))')))

[ouyanglinke]

It's a network connection issue. Calculating BLEU and METEOR requires a connection to Huggingface. You can try fixing your network proxy.

Alternatively, you can evaluate only the edit distance without evaluating BLEU and METEOR. You can comment out these two metrics in your config file (e.g., end2end.yaml):

...
metric:
        - Edit_dist
        # - BLEU
        # - METEOR
...

[viviansmlie]

如何在不连接hunggingface的情况下加载本地的包进行bleu的评测呢

[ouyanglinke]

如何在不连接hunggingface的情况下加载本地的包进行bleu的评测呢

#26 (comment)