daquang / YAMDA

Yet Another Motif Discovery Algorithm
MIT License
51 stars 7 forks source link

max_log_likelihoods, max_log_likelihoods_index = log_likelihoods.max(dim=0) IndexError: max(): Expected reduction dim 0 to have non-zero size. #12

Open Jean497 opened 2 years ago

Jean497 commented 2 years ago

python3 ./YAMDA/YAMDA-0.1/run_em.py -f 0.1 -r -e -w 20 -maxs 20000 -i ./raw_data/positive_sequence/R5_LUNG_M_2_positive_seq_uniq_name.fa -j ./raw_data/negtive_sequence/R5_LUNG_M_2_negtive_seq_uniq_name_paired.fa -oc ./motif_find/R5_output/LUNG Loading sequences from FASTA Searching positive sequences: 100%|███████████████████████████████████████████████████████████████████| 9/9 [00:02<00:00, 3.70it/s] Searching negative sequences: 100%|███████████████████████████████████████████████████████████████████| 9/9 [00:02<00:00, 3.79it/s] Converting letter sequences to tensors Pass 1/1: 100%|██████████████████████████████████████████████████████████████████████████████████| 230/230 [00:00<00:00, 297.42it/s] On-line EM: 100%|█████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1.29it/s] Pass 1/1: 100%|█████████████████████████████████████████████████████████████████████████████████| 230/230 [00:00<00:00, 1058.46it/s] Batch EM: 100%|███████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 4.58it/s] Computing log likelihood: 100%|█████████████████████████████████████████████████████████████████| 230/230

[00:00<00:00, 1831.33it/s] Traceback (most recent call last): File "./YAMDA/YAMDA-0.1/run_em.py", line 139, in main() File "./YAMDA/YAMDA-0.1/run_em.py", line 124, in main seqs, seqs_neg = model.fit(seqs, seqs_neg) File "/public1/yaoj/SELEX/YAMDA/YAMDA-0.1/yamda/mixture.py", line 126, in fit max_log_likelihoods, max_log_likelihoods_index = log_likelihoods.max(dim=0) IndexError: max(): Expected reduction dim 0 to have non-zero size.

Jean497 commented 2 years ago

Do you know how to deal with it?