namisan / mt-dnn

Multi-Task Deep Neural Networks for Natural Language Understanding
MIT License
2.24k stars 412 forks source link

Pretrained weights for transfer learning STS benmark #255

Open viethoang303 opened 1 year ago

viethoang303 commented 1 year ago

Hello author,

This is interesting work. I am wanting to transfer learning on STS-benmark. My scripts is: bash scripts/run_stsb.sh 16 16 0

and I got the error: Traceback (most recent call last): File "../train.py", line 476, in main() File "../train.py", line 365, in main state_dict = torch.load(init_model, map_location=device) File "/root/miniconda3/envs/ML/lib/python3.6/site-packages/torch/serialization.py", line 593, in load return _legacy_load(opened_file, map_location, pickle_module, pickle_load_args) File "/root/miniconda3/envs/ML/lib/python3.6/site-packages/torch/serialization.py", line 763, in _legacy_load magic_number = pickle_module.load(f, pickle_load_args) EOFError: Ran out of input

And I detected that the weight file is empty. Can you provide it again?

Thank you.