search

elnino9ykl / OmniSupervised-ConvNet

Omni-Supervised ConvNet for Robust Semantic Segmentation
0 stars 0 forks source link

Dimension errors #1

Closed anamce closed 3 months ago

anamce commented 3 months ago

Hi , I just picked ERFPSPNEt to train on my custom dataset and I am getting this error RuntimeError: Given input size: (128x32x64). Calculated output size: (128x0x0). Output size is too small Could you please help me ?

elnino9ykl commented 3 months ago

The input size is a bit small, you could try first enlarging the input size and check whether it works.

anamce commented 3 months ago

Hi I changed input resolution from (512,512) to (512,1024) still it doesn’t work .

Sent from Outlook for iOShttps://aka.ms/o0ukef

From: Kailun Yang @.> Sent: Wednesday, September 4, 2024 1:54:46 PM To: elnino9ykl/OmniSupervised-ConvNet @.> Cc: anamce @.>; Author @.> Subject: Re: [elnino9ykl/OmniSupervised-ConvNet] Dimension errors (Issue #1)

Closed #1https://github.com/elnino9ykl/OmniSupervised-ConvNet/issues/1 as completed.

— Reply to this email directly, view it on GitHubhttps://github.com/elnino9ykl/OmniSupervised-ConvNet/issues/1#event-14130514701, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BCVTWXK7HDQVVILJK6MYRRLZU37JNAVCNFSM6AAAAABNUCDJUOVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJUGEZTANJRGQ3TAMI. You are receiving this because you authored the thread.Message ID: @.***>

anamce commented 3 months ago

I also run this github repo over the cityscapes but it throws assersion errors m using 1 GPU. Could you please guide me on this :)

Sent from Outlook for iOShttps://aka.ms/o0ukef

From: Anam Manzoor @.> Sent: Wednesday, September 4, 2024 2:57:57 PM To: elnino9ykl/OmniSupervised-ConvNet @.>; elnino9ykl/OmniSupervised-ConvNet @.> Cc: Author @.> Subject: Re: [elnino9ykl/OmniSupervised-ConvNet] Dimension errors (Issue #1)

Hi I changed input resolution from (512,512) to (512,1024) still it doesn’t work .

Sent from Outlook for iOShttps://aka.ms/o0ukef

From: Kailun Yang @.> Sent: Wednesday, September 4, 2024 1:54:46 PM To: elnino9ykl/OmniSupervised-ConvNet @.> Cc: anamce @.>; Author @.> Subject: Re: [elnino9ykl/OmniSupervised-ConvNet] Dimension errors (Issue #1)

Closed #1https://github.com/elnino9ykl/OmniSupervised-ConvNet/issues/1 as completed.

— Reply to this email directly, view it on GitHubhttps://github.com/elnino9ykl/OmniSupervised-ConvNet/issues/1#event-14130514701, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BCVTWXK7HDQVVILJK6MYRRLZU37JNAVCNFSM6AAAAABNUCDJUOVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJUGEZTANJRGQ3TAMI. You are receiving this because you authored the thread.Message ID: @.***>

elnino9ykl commented 2 months ago

Hi, you could try printing the size of the tensors in the network for debugging. Then, you will have a better understanding of what happens there.

anamce commented 2 months ago

Alright! I might get back to you 👍

Sent from Outlook for iOShttps://aka.ms/o0ukef

From: Kailun Yang @.> Sent: Wednesday, September 4, 2024 5:52:54 PM To: elnino9ykl/OmniSupervised-ConvNet @.> Cc: anamce @.>; Author @.> Subject: Re: [elnino9ykl/OmniSupervised-ConvNet] Dimension errors (Issue #1)

Hi, you could try printing the size of the tensors in the network for debugging. Then, you will have a better understanding of what happens there.

— Reply to this email directly, view it on GitHubhttps://github.com/elnino9ykl/OmniSupervised-ConvNet/issues/1#issuecomment-2329556167, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BCVTWXNK3P3WSEBW442TDWLZU43GNAVCNFSM6AAAAABNUCDJUOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRZGU2TMMJWG4. You are receiving this because you authored the thread.Message ID: @.***>