Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/rouge/rouge.py

[yananchen1989]

datasets version: 1.12.1

metric = datasets.load_metric('rouge')

The error:

ConnectionError Traceback (most recent call last)
in
----> 1 metric = datasets.load_metric('rouge')

/usr/local/lib/python3.6/dist-packages/datasets/load.py in load_metric(path, config_name, process_id, num_process, cache_dir, experiment_id, keep_in_memory, download_config, download_mode, script_version, **metric_init_kwargs)
613 download_config=download_config,
614 download_mode=download_mode,
--> 615 dataset=False,
616 )
617 metric_cls = import_main_class(module_path, dataset=False)

/usr/local/lib/python3.6/dist-packages/datasets/load.py in prepare_module(path, script_version, download_config, download_mode, dataset, force_local_path, dynamic_modules_path, return_resolved_file_path, **download_kwargs)
328 file_path = hf_github_url(path=path, name=name, dataset=dataset, version=script_version)
329 try:
--> 330 local_path = cached_path(file_path, download_config=download_config)
331 except FileNotFoundError:
332 if script_version is not None:

/usr/local/lib/python3.6/dist-packages/datasets/utils/file_utils.py in cached_path(url_or_filename, download_config, **download_kwargs)
296 use_etag=download_config.use_etag,
297 max_retries=download_config.max_retries,
--> 298 use_auth_token=download_config.use_auth_token,
299 )
300 elif os.path.exists(url_or_filename):

/usr/local/lib/python3.6/dist-packages/datasets/utils/file_utils.py in get_from_cache(url, cache_dir, force_download, proxies, etag_timeout, resume_download, user_agent, local_files_only, use_etag, max_retries, use_auth_token)
603 raise FileNotFoundError("Couldn't find file at {}".format(url))
604 _raise_if_offline_mode_is_enabled(f"Tried to reach {url}")
--> 605 raise ConnectionError("Couldn't reach {}".format(url))
606
607 # Try a second time

ConnectionError: Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/rouge/rouge.py

Is there any remedy to solve the connection issue ?

[mariosasko]

Hi,

Did you try to run the code multiple times (GitHub URLs can be down sometimes for various reasons)? I can access https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/rouge/rouge.py, so this code is working without an error on my side.

Additionally, can you please run the datasets-cli env command because it seems to me that you are using the datasets version different from 1.12.1?

[CharlieUnderwood]

Same issue when running metric = datasets.load_metric("accuracy").
Error info is:

metric = datasets.load_metric("accuracy")
Traceback (most recent call last):

  File "<ipython-input-2-d25db38b26c5>", line 1, in <module>
    metric = datasets.load_metric("accuracy")

  File "D:\anaconda3\lib\site-packages\datasets\load.py", line 610, in load_metric
    module_path, _ = prepare_module(

  File "D:\anaconda3\lib\site-packages\datasets\load.py", line 330, in prepare_module
    local_path = cached_path(file_path, download_config=download_config)

  File "D:\anaconda3\lib\site-packages\datasets\utils\file_utils.py", line 288, in cached_path
    output_path = get_from_cache(

  File "D:\anaconda3\lib\site-packages\datasets\utils\file_utils.py", line 605, in get_from_cache
    raise ConnectionError("Couldn't reach {}".format(url))
ConnectionError: Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/accuracy/accuracy.py

My datasets-cli env result is as follows:

• datasets version: 1.11.0
• Platform: Windows-10-10.0.19041-SP0
• Python version: 3.8.8
• PyArrow version: 6.0.0

@yananchen1989 did you find a way to solve this?

[CharlieUnderwood]

It seems to be able to solve this issue by adding the equivalent accuracy.py locally.
change metric = datasets.load_metric("accuracy") to metric = datasets.load_metric(path = "./accuracy.py").
Copy accuracy.py from browser at accuracy.py

[qiuganbin123]

It seems to be able to solve this issue by adding the equivalent accuracy.py locally. change metric = datasets.load_metric("accuracy") to metric = datasets.load_metric(path = "./accuracy.py"). Copy accuracy.py from browser at accuracy.py

This is really a good way