You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I am trying to test your model in google colab. I put a dummy image a dummy folder:
'/content/colorization-pytorch/dataset/ilsvrc2012/val/MyImages'
I tested these two following commands:
#!python test.py --name siggraph_caffemodel --mask_cent 0
!python test.py --name siggraph_retrained
Both led to the same error:
Traceback (most recent call last):
File "test.py", line 53, in
data_raw[0] = util.crop_mult(data_raw[0], mult=8)
File "/content/colorization-pytorch/util/util.py", line 277, in crop_mult
return data[:,:,h:h+Hnew,w:w+Wnew]
TypeError: slice indices must be integers or None or have an index method
I fixed it by casting indices to integers. Then I got another error in test.py line 57:
img_path = [string.replace('%08d_%.3f' % (i, sample_p), '.', 'p')]
Hi, I am trying to test your model in google colab. I put a dummy image a dummy folder:
'/content/colorization-pytorch/dataset/ilsvrc2012/val/MyImages'
I tested these two following commands:
#!python test.py --name siggraph_caffemodel --mask_cent 0
!python test.py --name siggraph_retrained
Both led to the same error:
Traceback (most recent call last):
File "test.py", line 53, in
data_raw[0] = util.crop_mult(data_raw[0], mult=8)
File "/content/colorization-pytorch/util/util.py", line 277, in crop_mult
return data[:,:,h:h+Hnew,w:w+Wnew]
TypeError: slice indices must be integers or None or have an index method
I fixed it by casting indices to integers. Then I got another error in test.py line 57:
img_path = [string.replace('%08d_%.3f' % (i, sample_p), '.', 'p')]
string.replace is deprecated in python 2.7, and does not exists in python 3, see here: https://docs.python.org/2.7/library/string.html?highlight=string%20replace#string.replace
However, when I change my google colab execution to python 2, test.py works like a charm.
The text was updated successfully, but these errors were encountered: