-
Notifications
You must be signed in to change notification settings - Fork 13
Description
Hi,
Thank you for sharing Promotech.
I am working with Whole-genome, and recurrently, it retrieves problems to access files in the models directory.
I could run the first part of the pipeline using the RF-HOT model with no worries
python ~/programs/Promotech-master/promotech.py -pg -m RF-HOT -f ~/genomes/Pseudomonas_pergaminensis/GCF_024112395.2_ASM2411239v2_genomic.fna
Yet, if I try to run for the GRU model or when running the second command line, I receive an error regarding the Tokenizer file:
LOADING TOKENIZER: ./models/tokenizer.data
Traceback (most recent call last):
File "/home/externo/jugalhardo/programs/Promotech-master/promotech.py", line 83, in
data_type = args.model,
File "/home/externo/jugalhardo/programs/Promotech-master/genome/process_genome.py", line 92, in parseGenome40NTSequences
X = dataConverter( seqs=cutted_seqs, data_type=data_type, tokenizer_path=tokenizer_path, print_fn=print_fn, log_file=log_file )
File "/home/externo/jugalhardo/programs/Promotech-master/sequences/../core/utils.py", line 241, in dataConverter
tokenizer = pickle.load(open( tokenizer_path, 'rb' ))
FileNotFoundError: [Errno 2] No such file or directory: './models/tokenizer.data'
I also tried providing the absolute path for the models' directory on the script and giving chmod 777 permission on the Promotech-master directory, but I still received the same error.
There was a similar issue in which the solution was to download/unzip the model, but it has already been done.
Would anyone know how to procede?