Releases: StijnvLierop/DeepfakeDetection
v0.24.1
v0.24.1 (2025-12-24)
This release is published under the MIT License.
Detailed Changes: v0.24.0...v0.24.1
v0.24.0
v0.24.0 (2025-12-24)
This release is published under the MIT License.
Bug Fixes
-
Cnndetect transform function now transforms inputs instead of returning the transform function (
881ddda) -
Sigma variable naming in randomgaussianblur augmentation (
0a07380)
Refactoring
-
Change default augmentation params to be same as in cnndetect paper (
1853c3d) -
Transform inputs function now returns function instead of transformed inputs (
ab8c545)
Detailed Changes: v0.23.0...v0.24.0
v0.23.0
v0.23.0 (2025-12-21)
This release is published under the MIT License.
Features
- Add evaluator class to quickly calculate metrics (
40d6203)
Detailed Changes: v0.22.0...v0.23.0
v0.22.0
v0.22.0 (2025-12-20)
This release is published under the MIT License.
Features
- Add cnndetect dataset (
b8f4794)
Detailed Changes: v0.21.0...v0.22.0
v0.21.0
v0.21.0 (2025-12-20)
This release is published under the MIT License.
Features
- Add support for pytorch training (
d1654ae)
Refactoring
- Fix formatting (
04953b9)
Testing
- Fix unit tests (
1a1407e)
Detailed Changes: v0.20.0...v0.21.0
v0.20.0
v0.20.0 (2025-12-13)
Features
- Move predict function to model class so models only have to implement predict_batch function (
d44a078)
Detailed Changes: v0.19.0...v0.20.0
v0.19.0
v0.19.0 (2025-12-11)
Features
Refactoring
- Add missing dependency and fix module import (
ecd30d3)
Detailed Changes: v0.18.0...v0.19.0
v0.18.0
v0.17.0
v0.16.0
v0.16.0 (2025-12-07)
Bug Fixes
- test: Add save method to MockInstance class (
1e7c504)
Refactoring
-
Remove dataset saving functionality (
d381be6) -
Remove dnn-cnn denoising model (
8ba8042) -
Remove specific models and configs (
2cfe027) -
Write_predictions_to_file function now only takes predictions and filepath as parameters (
86dc7dd)
Testing
- Remove tests that are not needed anymore (
fd6ea70)
Detailed Changes: v0.15.1...v0.16.0