search

eurecom-asp / RawGAT-ST-antispoofing

This repository includes the code to reproduce our paper "End-to-End Spectro-Temporal Graph Attention Networks for Speaker Verification Anti-Spoofing and Speech Deepfake Detection" (https://arxiv.org/abs/2107.12710) published in the ASVspoof 2021 workshop.
MIT License
74 stars 18 forks source link

pretrained model error #7

Closed pradeepkc11 closed 9 months ago

pradeepkc11 commented 10 months ago

cache_fname cache_eval_LA_Raw_GAT.npy Traceback (most recent call last): File "/home/pradeep/.local/lib/python3.10/site-packages/torch/serialization.py", line 619, in save _save(obj, opened_zipfile, pickle_module, pickle_protocol, _disable_byteorder_record) File "/home/pradeep/.local/lib/python3.10/site-packages/torch/serialization.py", line 853, in _save zip_file.write_record(name, storage.data_ptr(), num_bytes) RuntimeError: [enforce fail at inline_container.cc:588] . PytorchStreamWriter failed writing file data/51697: file write failed

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/home/pradeep/RawGAT-ST-antispoofing/main.py", line 228, in dev_set = data_utils.ASVDataset(database_path=args.database_path,protocols_path=args.protocols_path,is_train=False, is_logical=is_logical, File "/home/pradeep/RawGAT-ST-antispoofing/data_utils.py", line 106, in init torch.save((self.data_x, self.data_y, self.data_sysid, self.files_meta), self.cache_fname) File "/home/pradeep/.local/lib/python3.10/site-packages/torch/serialization.py", line 618, in save with _open_zipfile_writer(f) as opened_zipfile: File "/home/pradeep/.local/lib/python3.10/site-packages/torch/serialization.py", line 466, in exit self.file_like.write_end_of_file() RuntimeError: [enforce fail at inline_container.cc:424] . unexpected pos 11996555136 vs 11996555000

TakHemlata commented 10 months ago

Hi @pradeepkc11 I think this error is related to Pytorch new version. Regarding second error you have to use correct path for development protocol so that will save correct file name for feature extraction.

I suggest that you can use data-loader part and main script from this repo (https://github.com/asvspoof-challenge/2021/tree/main/LA/Baseline-RawNet2) with RawGAT-ST model script. Much more simple data-loader script.

Please let me know if it works for you.

Thanks