gcorso
python evaluate_files.py failed
I run the command:
python -m inference --protein_ligand_csv data/testset_csv.csv --out_dir results/user_predictions_testset --inference_steps 20 --samples_per_complex 40 --batch_size 10 --actual_steps 18 --no_final_step_noise
and it shows:
Failed on tensor([331]) linalg.svd: The algorithm failed to converge because the input matrix is ill-conditioned or has too many repeated singular values (error code: 2). 363it [9:12:19, 91.29s/it] Failed for 3 complexes Skipped 1 complexes Results are in results/user_predictions_testset
Then I run the evaluation command:
python evaluate_files.py --results_path results/user_predictions_testset --file_to_exclude rank1.sdf --num_predictions 40
and I got the result like this:
Did not find a directory for 6qrg . We are skipping that complex Did not find a directory for 6nxz . We are skipping that complex Did not find a directory for 6iby . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 5 6 7 8 14 15 16 17 33 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6gj7 . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 1 3 18 21 22 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6qr3 . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 1 2 15 20 21 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6qr1 . We are skipping that complex Did not find a directory for 6s9x . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 60 61 62 63 64 65 66 67 68 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6q4q . We are skipping that complex Did not find a directory for 6hbn . We are skipping that complex Did not find a directory for 6nw3 . We are skipping that complex Did not find a directory for 6tel . We are skipping that complex Did not find a directory for 6p8y . We are skipping that complex Did not find a directory for 6d5w . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 0 3 4 5 6 7 8 13 14 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6t6a . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 12 13 14 15 16 17 18 19 23 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6o5g . We are skipping that complex Did not find a directory for 6r7d . We are skipping that complex 99%|██████████████████████████████████████████████████████████▏| 358/363 [00:01<00:00, 253.46it/s]Can't kekulize mol. Unkekulized atoms: 0 1 2 3 4 5 6 7 18 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6pya . We are skipping that complex Did not find a directory for 6ffe . We are skipping that complex Did not find a directory for 6d3x . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 5 6 7 8 19 20 21 22 23 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6gj8 . We are skipping that complex Can't kekulize mol. Unkekulized atoms: 28 29 30 31 32 RDKit was unable to read the molecule. Using the .sdf file failed. We found a .mol2 file instead and are trying to use that. Did not find a directory for 6mo2 . We are skipping that complex 100%|███████████████████████████████████████████████████████████| 363/363 [00:01<00:00, 222.21it/s] evaluate_files.py:121: RuntimeWarning: invalid value encountered in scalar divide f'{overlap}steric_clash_fraction': (100 * (min_cross_distances < 0.4).sum() / len(min_cross_distances) / args.num_predictions).round(2), evaluate_files.py:122: RuntimeWarning: invalid value encountered in scalar divide f'{overlap}self_intersect_fraction': (100 * (min_self_distances < 0.4).sum() / len(min_self_distances) / args.num_predictions).round(2), Traceback (most recent call last): File "evaluate_files.py", line 123, in
f'{overlap}mean_rmsd': rmsds[:,0].mean(), IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed
I don't what caused the result like this, I saw the inference results in /results/user_predictions_testset, and it seems has a problem since it named with numbers instead of protein names like: 6qrg, 6nxz.... :
