5,337
edits
Line 6: | Line 6: | ||
** If it doesn't overfit, there may be an issue with your code. | ** If it doesn't overfit, there may be an issue with your code. | ||
** You can try increasing the capacity (e.g number of filters or number of nodes in FC) 2-4x. | ** You can try increasing the capacity (e.g number of filters or number of nodes in FC) 2-4x. | ||
*** For CNNs, these days, you can fit O(10 million) parameters on a single GPU. | |||
**: If the input is 3 channels, then the first conv layer should have more than 3 channels. | **: If the input is 3 channels, then the first conv layer should have more than 3 channels. | ||
** Check that your loss is implemented correctly and taken against the correct ground truth image. | ** Check that your loss is implemented correctly and taken against the correct ground truth image. |