Colabs throw error 'MyRandomPerspective' object has no attribute 'resample'

[richpav]

I tried Start_Here.ipynb and Swap_Model.ipynb

Same error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-3-d6266773d3a4> in <module>()
     12 settings = pixray.apply_settings()
     13 pixray.do_init(settings)
---> 14 pixray.do_run(settings)

11 frames
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py in __getattr__(self, name)
   1129                 return modules[name]
   1130         raise AttributeError("'{}' object has no attribute '{}'".format(
-> 1131             type(self).__name__, name))
   1132 
   1133     def __setattr__(self, name: str, value: Union[Tensor, 'Module']) -> None:

AttributeError: 'MyRandomPerspective' object has no attribute 'resample'

[dribnet]

This is due to a dependency change. Try the new Swap_Model notebook which now lives here.

[richpav]

Here's the error I get with Pixray Swap Model from where you said.

Using seed: 11543842745187050271
Working with z of shape (1, 256, 16, 16) = 65536 dimensions.
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to models/checkpoints/vgg16-397923af.pth
100%
528M/528M [00:03<00:00, 177MB/s]

Downloading vgg_lpips model from https://heibox.uni-heidelberg.de/f/607503859c864bc1b30b/?dl=1 to taming/modules/autoencoder/lpips/vgg.pth
8.19kB [00:00, 2.76MB/s]                   
loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth
VQLPIPSWithDiscriminator running with hinge loss.
Restored from models/vqgan_imagenet_f16_16384.ckpt
100%|███████████████████████████████████████| 402M/402M [00:06<00:00, 66.8MiB/s]
100%|████████████████████████████████████████| 338M/338M [00:01<00:00, 177MiB/s]
100%|████████████████████████████████████████| 335M/335M [00:02<00:00, 175MiB/s]
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-4-d6266773d3a4> in <module>()
     11 
     12 settings = pixray.apply_settings()
---> 13 pixray.do_init(settings)
     14 pixray.do_run(settings)

5 frames
/usr/local/lib/python3.7/dist-packages/PIL/TiffImagePlugin.py in ImageFileDirectory_v2()
    685                 (TiffTags.FLOAT, "f", "float"),
    686                 (TiffTags.DOUBLE, "d", "double"),
--> 687                 (TiffTags.IFD, "L", "long"),
    688             ],
    689         )

AttributeError: module 'PIL.TiffTags' has no attribute 'IFD'```

[dribnet]

i thin that might just be restarting the runtime after running setup

[richpav]

I just went down the page clicking every run button without changing any settings. Throws an error. I'm using the $10/month Google account.

[dribnet]

Right. But did you restart the notebook when it said to restart the notebook? That can't be done by the button and has to be done via the menu. You don't have to run setup twice - but you do have to restart the runtime after running setup so that the installations take effect in the runtime.

[richpav]

No, because I'm a goddamn idiot.

Thank you. And sorry.

[dribnet]

Not your fault at all - this is a very common problem! the colab does not provide a good way to signal that the runtime needs to be restarted which is unfortunate.

[richpav]

You could put a message in H1 after the cell that throws an error to remind people to restart the notebook. I usually don't pay attention to what messages come out of the setup.