1) bash preprocess.sh with dataset_name=ZINC and random=canonical
2) bash preprocess.sh with dataset_name=PROTAC and random=random
3) cp data/ZINC/canonical/.vocab.pt data/PROTAC/random/
4) bash fine-tune-training.sh
I made a small change to fine-tune-training.sh, I replaced the mkdir command with mkdir -p $pathsave.
Here is the error in full:
Traceback (most recent call last):
File "train.py", line 118, in <module>
main(opt)
File "train.py", line 51, in main
single_main(opt, 0)
File "/gpfs/gibbs/pi/krishnaswamy_smita/dbhaskar/projects/PROTAC-RL/onmt/train_single.py", line 87, in main
init_logger(opt.log_file)
File "/gpfs/gibbs/pi/krishnaswamy_smita/dbhaskar/projects/PROTAC-RL/onmt/utils/logging.py", line 19, in init_logger
file_handler = logging.FileHandler(log_file)
File "/home/db2454/gibbs/dbhaskar/conda_envs/PROTAC-RL/lib/python3.6/logging/__init__.py", line 1032, in __init__
StreamHandler.__init__(self, self._open())
File "/home/db2454/gibbs/dbhaskar/conda_envs/PROTAC-RL/lib/python3.6/logging/__init__.py", line 1061, in _open
return open(self.baseFilename, self.mode, encoding=self.encoding)
FileNotFoundError: [Errno 2] No such file or directory: '/gpfs/gibbs/dbhaskar/projects/PROTAC-RL/log/transformer_PROTAC_random_protac7000_zinc300000.txt'
It's our mistake to forget pre-training step in the example of README. Please follow new example process to train PROTAC-RL.
It seems that github ignored origin empty log folder in our code. Therefore your program couldn't find log/transformer_PROTAC_random_protac7000_zinc300000.txt. You can mkdir a new log folder to solve this problem.
Here are the steps to reproduce this error:
1)
bash preprocess.sh
withdataset_name=ZINC
andrandom=canonical
2)bash preprocess.sh
withdataset_name=PROTAC
andrandom=random
3)cp data/ZINC/canonical/.vocab.pt data/PROTAC/random/
4)bash fine-tune-training.sh
I made a small change to
fine-tune-training.sh
, I replaced the mkdir command withmkdir -p $pathsave
.Here is the error in full: