PyTorch-Ignite 0.4.12 - Release Notes

New Features

Engine and Events

• Added model_transform to create_supervised_evaluator so that user be able to transform model output into actual prediction (y_pred) (#2896)

Metrics and handlers

• Updated the NeptuneLogger (#2881)
• Improved ClearMLLogger. Accessing attributes of the logger, retrieves those of the underlying clearml task. get_task method is also added (#2898)
• Added score_sign to add_early_stopping_by_val_score and gen_save_best_models_by_val_score to support both error-like and accuracy-like scores (#2898)

Bug Fixes

• Fixed error on importing Events in Python3.11 (#2907)
• Fixed an inefficiency in SSIM metric (#2914)
• Fixed NeptuneSaver (#2900, #2902)

Housekeeping (docs, CI, examples, tests, etc)

• #2871, #2872, #2875, #2877, #2879, #2886, #2887, #2897, #2899, #2905, #2918, #2922, #2924
• Added CUDA 11.7 support (#2873)
• Added an example on Siamese network used for image similarity estimation task (#2882)
• Added an example on sub-pixel convolution layer which is used in vision tasks such as super-resolution (#2885)
• Added PyTorch 1.13 support (#2890)

Acknowledgments

🎉 Thanks to our community and all our contributors for the issues, PRs and 🌟 ⭐️ 🌟 !
💯 We really appreciate your implication into the project (in alphabetical order):

@AlexanderChaptykov, @DeepC004, @Hummer12007, @divij-pawar, @guptaaryan16, @kshitij12345, @moienr, @normandy7, @sadra-barikbin, @sallycaoyu, @twolodzko, @vfdev-5

New Contributors

• @twolodzko made their first contribution in #2881
• @divij-pawar made their first contribution in #2887
• @normandy7 made their first contribution in #2886
• @kshitij12345 made their first contribution in #2900
• @moienr made their first contribution in #2914
• @Hummer12007 made their first contribution in #2929

Full Changelog: v0.4.11...v0.4.12