Just a heads up for anyone who might run into this issue.
When I tried training a network from a pre-trained model (squeezesegV2), I got the following error code.
RuntimeError: CUDA out of memory. Tried to allocate 128.00 MiB (GPU 0; 7.93 GiB total capacity; 7.21 GiB already allocated; 9.00 MiB free; 45.45 MiB cached)
Solution : in the config file, change the batch size (default is 8 for the model I was using). After changing it to 4 to see if it would help, the training process seemed to have started up fine.
Just a heads up for anyone who might run into this issue.
When I tried training a network from a pre-trained model (squeezesegV2), I got the following error code.
RuntimeError: CUDA out of memory. Tried to allocate 128.00 MiB (GPU 0; 7.93 GiB total capacity; 7.21 GiB already allocated; 9.00 MiB free; 45.45 MiB cached)Solution : in the config file, change the batch size (default is 8 for the model I was using). After changing it to 4 to see if it would help, the training process seemed to have started up fine.