Conditional Sequence Generation example command throwing an error

[JDaksh]

I have the local environment for evodiff set up, got the uniref split from zenodo and I'm able to run all the unconditional generation example commands.

When I try running the following (example) conditional generation command,
python evodiff/generate_msa.py --model-type msa_oa_dm_maxsub --batch-size 1 --n-sequences 64 --n-sequences 256 --subsampling MaxHamming --start-msa

I get the error

Traceback (most recent call last):
File "/workspace/evodiff/evodiff/generate_msa.py", line 413, in
main()
File "/workspace/evodiff/evodiff/generate_msa.py", line 138, in main
sample, _string = generate_msa(model, tokenizer, args.batch_size, args.n_sequences, args.seq_length,
File "/workspace/evodiff/evodiff/generate_msa.py", line 186, in generate_msa
valid_msas, query_sequences, tokenizer = get_valid_data(data_top_dir, batch_size, 'autoreg',
File "/workspace/evodiff/evodiff/generate_msa.py", line 351, in get_valid_data
dataset = A3MMSADataset(selection_type, n_sequences, max_seq_len, data_dir=os.path.join(data_top_dir,data_dir), min_depth=64)
File "/opt/conda/lib/python3.10/site-packages/evodiff/data.py", line 353, in init
all_files = np.array(all_files)[keep_idx]
IndexError: index 0 is out of bounds for axis 0 with size 0

I tried the same command out in the docker cford38/evodiff and got the exact same error again.

How should I fix this?