Expected shape error in AlphaFold step, advanced notebook about colabfold

[sokrypton]

The error should now be fixed! Can you refresh and try again on the same example?

[phylars]

Hi. I got exactly the same error as well. By the time I check this issue report, this problem is NOT fixed yet. I repeated so many times with different setting, the same problem always occur like above:

Running model_1_ptm_seed_0: 0%
0/5 [elapsed: 12:50 remaining: ?]

IndexError Traceback (most recent call last)
in ()
188
189 prediction_result, (r, t) = cf.to(model_runner.predict(processed_feature_dict, random_seed=seed),"cpu")
--> 190 outs[key] = parse_results(prediction_result, processed_feature_dict)
191
192 # report

3 frames
/usr/local/lib/python3.7/dist-packages/jax/_src/numpy/lax_numpy.py in _expand_bool_indices(idx, shape)
5400 expected_shape = shape[len(out): len(out) + _ndim(i)]
5401 if i_shape != expected_shape:
-> 5402 raise IndexError("boolean index did not match shape of indexed array in index "
5403 f"{dim_number}: got {i_shape}, expected {expected_shape}")
5404 out.extend(np.where(i))

IndexError: boolean index did not match shape of indexed array in index 2: got (63,), expected (64,)

[sokrypton]

@phylars can you refresh the notebook and try again? I wonder if you didn't get the latest notebook yet.

Can you send me the sequence you used (you can email me at [email protected] if it's private) I can try on my side.

[phylars]

@sokrypton Thank you. I just reopened a new one. It works now.

[sokrypton]

Yay! Thanks for the report.