the output images' format should be changed from CHW to HWC

[nightmaredimple]

Tensors in pytorch are formatted in CHW(BCHW) by default, so if you wanna output the results of depth,flow and mask, you should change them into HWC format.
such as:
test_flow.py line 180

row1_viz_im = Image.fromarray((255*row1_viz).astype('uint8'))
row2_viz_im = Image.fromarray((row2_viz).astype('uint8'))

this will raise TypeError("Cannot handle this data type")
you should transpose/permute the format into HWC like this below:

row1_viz_im = Image.fromarray((255*row1_viz).astype('uint8').transpose((1,2,0)))
row2_viz_im = Image.fromarray((row2_viz).astype('uint8').transpose((1,2,0)))

[nightmaredimple]

Due to the problem above, when testing masks' output, the results of flow are totally black. Besides, the colormap of hot needs to be defined in utils/COLORMAPS.

[anuragranj]

okay. this is due to version changes in torchvision. Can you send a pull request for the fix? Thanks so much.

[nightmaredimple]

I'm getting rid of some features in old versions of pytorch . so maybe a few days later~

[anuragranj]

Okay. Thank you so much :)

[ChenjieWang]

Due to the problem above, when testing masks' output, the results of flow are totally black. Besides, the colormap of hot needs to be defined in utils/COLORMAPS.

Do you know how to define the colormap of hot in utils/COLORMAPS please?

[anuragranj]

hot can be found here: https://matplotlib.org/3.1.1/gallery/color/colormap_reference.html
and can be used by modifying this line:
https://github.com/anuragranj/cc/blob/master/utils.py#L34

[swrdZWJ]

I have the same problem with you. Do you solve the problem correctly and how to modify the code in 'test_mask.py' and 'utils.py'?

viz3_im = Image.fromarray(viz3.astype('uint8'))
TypeError: Cannot handle this data type

Thank you (^_^) @anuragranj @ChenjieWang

[lotus0903]

Due to the problem above, when testing masks' output, the results of flow are totally black. Besides, the colormap of hot needs to be defined in utils/COLORMAPS.

I wonder how to solve the black problem. Thanks a lot~