search

HHousen / DocSum

A tool to automatically summarize documents abstractively using the BART or PreSumm Machine Learning Model.
https://haydenhousen.com/projects/docsum/
GNU General Public License v3.0
66 stars 13 forks source link

S3 bucket model not found error #2

Closed venkatapavithra closed 3 years ago

venkatapavithra commented 4 years ago

Traceback (most recent call last): File "run_summarization.py", line 325, in main() File "run_summarization.py", line 310, in main evaluate(args) File "run_summarization.py", line 34, in evaluate model = BertAbs.from_pretrained("bertabs-finetuned-cnndm") File "D:\anaconda3\lib\site-packages\transformers\modeling_utils.py", line 546, in from_pretrained kwargs, File "D:\anaconda3\lib\site-packages\transformers\configuration_utils.py", line 202, in from_pretrained config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, kwargs) File "D:\anaconda3\lib\site-packages\transformers\configuration_utils.py", line 272, in get_config_dict raise EnvironmentError(msg) OSError: Couldn't reach server at 'https://s3.amazonaws.com/models.huggingface.co/bert/remi/bertabs-finetuned-cnndm-extractive-abstractive-summarization-config.json' to download pretrained model configuration file.

I'm trying to run the code but facing the above error. The code fails to download the config file saying the file not found. Could you please help me to resolve this issue? i tried to access the model directly using the provided link but it raises the error "no such key"

HHousen commented 4 years ago

Hello @venkatapavithra. This happened because the huggingface/transformers repository changed their CDN, and thus their model URLs are different. I've updated the URL in commit 6579d79, but I may need to look into a different solution so ensure that this does not happen again. Try pulling the repository again. Does the code work now?