TypeError: __init__() got an unexpected keyword argument 'batch_first' about casanovo HOT 8 CLOSED

Hi,

The first command starting with casanovo is correct but for --model_path you should specify the filename, e.g. .../pretrained_excl_bacillus.ckpt, instead the of the directory. Let me know if that helps!

from casanovo.

Melih, can you add to the code a line to check whether the given argument is a file or a directory and to issue an error statement in the latter case?

from casanovo.

I just added this check to both model files and data directories.

from casanovo.

Hi, I tried the new command, but still got this wrong.
casanovo --mode=eval --model_path='./model/pretrained_excl_human.ckpt' --test_data_path='./data/' --config_path='../../casanovo/config.py'

====

INFO: Evaluating Casanovo...
Traceback (most recent call last):
  File "/usr/local/bin/casanovo", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 1128, in __call__
    return self.main(*args, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 1053, in main
    rv = self.invoke(ctx)
  File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 1395, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/usr/local/lib/python3.8/dist-packages/click/core.py", line 754, in invoke
    return __callback(*args, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/casanovo/casanovo.py", line 37, in main
    test_evaluate(test_data_path, model_path, config_path)
  File "/usr/local/lib/python3.8/dist-packages/casanovo/denovo/train_test.py", line 146, in test_evaluate
    model_trained = Spec2Pep().load_from_checkpoint(
  File "/usr/local/lib/python3.8/dist-packages/casanovo/denovo/model.py", line 103, in __init__
    self.encoder = SpectrumEncoder(
  File "/usr/local/lib/python3.8/dist-packages/depthcharge/components/transformers.py", line 60, in __init__
    layer = torch.nn.TransformerEncoderLayer(
TypeError: __init__() got an unexpected keyword argument 'batch_first'

from casanovo.

Can you run the following and post the output?

pip show torch

from casanovo.

Name: torch
Version: 1.8.0+cu111
Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration
Home-page: https://pytorch.org/
Author: PyTorch Team
Author-email: [email protected]
License: BSD-3
Location: /usr/local/lib/python3.8/dist-packages
Requires: numpy, typing-extensions
Required-by: casanovo, depthcharge-ms, efficientnet-pytorch, pretrainedmodels, pytorch-lightning, timm, torchaudio, torchmetrics, torchvision

from casanovo.

Thanks @piaopu0120. You'll need to upgrade to PyTorch >= 1.9 to use Casanovo. Version 1.9 is when they added the batch_first parameter. If you installed PyTorch with pip, you can do this with:

pip install --upgrade torch

@melihyilmaz - can you update setup.cfg to specify torch>=1.9 as a dependency?

from casanovo.

torch>=1.9 added to setup.cfg, closing this issue.

from casanovo.