Spaces:

simonduerr

/

ProteinMPNN

like 100

Running on T4

Generating sequences...

ValueError Traceback (most recent call last)
in <cell line: 2>()
25 S_sample = sample_dict["S"]
26 else:
---> 27 sample_dict = model.tied_sample(X, randn_2, S, chain_M, chain_encoding_all, residue_idx, mask=mask, temperature=temp, omit_AAs_np=omit_AAs_np, bias_AAs_np=bias_AAs_np, chain_M_pos=chain_M_pos, omit_AA_mask=omit_AA_mask, pssm_coef=pssm_coef, pssm_bias=pssm_bias, pssm_multi=pssm_multi, pssm_log_odds_flag=bool(pssm_log_odds_flag), pssm_log_odds_mask=pssm_log_odds_mask, pssm_bias_flag=bool(pssm_bias_flag), tied_pos=tied_pos_list_of_lists_list[0], tied_beta=tied_beta, bias_by_res=bias_by_res_all)
28 # Compute scores
29 S_sample = sample_dict["S"]

2 frames
/usr/local/lib/python3.10/dist-packages/opt_einsum/contract.py in contract_path(*operands, **kwargs)
236 size_dict[char] = dim
237 elif dim not in (1, size_dict[char]):
--> 238 raise ValueError("Size of label '{}' for operand {} ({}) does not match previous "
239 "terms ({}).".format(char, tnum, size_dict[char], dim))
240 else:

ValueError: Size of label 'i' for operand 1 (348) does not match previous terms (462).

I pushed some fixes to the server. AlphaFold predictions are currently still broken.

all fixed