error

Describe the bug hello, I installed the localcolabfold on my windows10/wsl2 following the instructions included into the github and i run the command as fellow:

colabfold_batch --amber --templates --num-recycle 3 fff.fasta outputdir/ and i got the following error:

WARNING: You are welcome to use the default MSA server, however keep in mind that it's a limited shared resource only capable of processing a few thousand MSAs per day. Please submit jobs only from a single IP address. We reserve the right to limit access to the server case-by-case when usage exceeds fair use.

If you require more MSAs, please host your own API and pass it to `--host-url`
2022-04-01 09:54:45,192 Running colabfold 1.2.0 (b532e910b15434f707f0b7460abc25c70fcb9b26)
2022-04-01 09:54:45,195 Found 8 citations for tools or databases
2022-04-01 09:54:53,540 Query 1/1: sp_O15552_FFAR2_HUMAN_Free_fatty_acid_receptor_2_OS_Homo_sapiens_OX_9606_GN_FFAR2_PE_1_SV_1 (length 330)
2022-04-01 09:55:03,288 Sequence 0 found templates: [b'5w0p_C' b'6i9k_A' b'6osa_R' b'5w0p_D' b'2ksb_A' b'6cmo_R' b'6c1q_B'
 b'5yhl_A' b'5ywy_A' b'6tod_A' b'6tod_B' b'2z73_A' b'3ayn_B' b'6wwz_R'
 b'6ko5_A' b'6c1r_B' b'5dhh_B' b'5unf_A' b'5dhg_B' b'6b73_A']
2022-04-01 09:55:03,557 Running model_3
Traceback (most recent call last):
  File "/home/najib/colabfold_batch/colabfold-conda/bin/colabfold_batch", line 8, in <module>
    sys.exit(main())
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 1467, in main
    save_pair_representations=args.save_pair_representations,
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 1169, in run
    prediction_callback=prediction_callback,
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 246, in predict_structure
    feature_dict, random_seed=random_seed
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/alphafold/model/model.py", line 135, in process_features
    random_seed=random_seed)
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/alphafold/model/features.py", line 94, in np_example_to_features
    np_example=np_example, features=feature_names)
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/alphafold/model/tf/proteins_dataset.py", line 165, in np_to_tensor_dict
    tensor_dict = parse_reshape_logic(tensor_dict, features_metadata)
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/alphafold/model/tf/proteins_dataset.py", line 104, in parse_reshape_logic
    parsed_features[k] = tf.reshape(v, new_shape, name="reshape_%s" % k)
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/tensorflow/python/util/traceback_utils.py", line 153, in error_handler
    raise e.with_traceback(filtered_tb) from None
  File "/home/najib/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1939, in _create_c_op
    raise ValueError(e.message)
ValueError: Cannot reshape a tensor with 1215380 elements to shape [3628,330,1] (1197240 elements) for '{{node reshape_msa}} = Reshape[T=DT_INT32, Tshape=DT_INT32](Const_6, reshape_msa/shape)' with input shapes: [3628,335], [3] and with input tensors computed as partial shapes: input[1] = [3628,330,1].

Desktop (please complete the following information):

• Windows10 & WSL2
• CUDA version 11.6

[najibveto]

yes thank you a lot, i removed the previous version and installed the 11.6 and it works fine. thank you a lot.

[YoshitakaMo]

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243

That’s the cause. The version must be 11.1 or later. Please upgrade it and try again!