About environment issue

[mikelmh025]

Hi there thanks for the great work!
When I was trying to run
python pretrain_STREAM.py --cfg cfg/STREAM/bird.yaml --gpu 0
on colab, I got this error:

TypeError: Cannot handle this data type: (1, 1, 48), |u1

I guess that might be an environmental issue, would you mind share your environment requirement for this implementation?

[Bila12]

Hello @mikelmh025 did you manage to solve the problem ? I'm facing the same issue and would like to avoid spending days trying to find the right environment settings.
Thanks !

[mikelmh025]

It's been a while. I don't remember actually. But I believe I did managed to get this model running some how.

Pretty sure the issue is related to environment issue. Check tensorboard, pytorch version, things related to you gpu. Oh maybe try running with CPU first. Just to see what might related to this.

[Bila12]

Ok thanks anyway. I might have solved it but not sure, so I'm giving my solution here. The skimage.transform.pyramid_expand line needs multichannel=True, otherwise it's gonna expand the 17x17x3 map into 272x272x48 rather than 272x272x3, which is now usable by PIL

[mikelmh025]

So you just changed you PIL version? That sounds familiar :P

[Bila12]

Haha not even that, although it might have been a solution. I just added the multichannel=True option in the code

[mikelmh025]

Oh wow