diff options
| author | aria1th <35677394+aria1th@users.noreply.github.com> | 2023-01-10 14:56:57 +0900 |
|---|---|---|
| committer | aria1th <35677394+aria1th@users.noreply.github.com> | 2023-01-10 14:56:57 +0900 |
| commit | a4a5475cfa3c68af6cb046081002a72f862ce4be (patch) | |
| tree | 54deee50926938b7be198b608bcfbdae7e7cb370 /models/deepbooru/Put your deepbooru release project folder here.txt | |
| parent | bd4587d2f5b70ed951d2c17f25a4622fc1cb31c2 (diff) | |
Variable dropout rate
Implements variable dropout rate from #4549
Fixes hypernetwork multiplier being able to modified during training, also fixes user-errors by setting multiplier value to lower values for training.
Changes function name to match torch.nn.module standard
Fixes RNG reset issue when generating previews by restoring RNG state
Diffstat (limited to 'models/deepbooru/Put your deepbooru release project folder here.txt')
0 files changed, 0 insertions, 0 deletions