Thibault de Varax

Results 42 issues of Thibault de Varax

Training config classes: - Changes `BaseTaskConfig` to a `TrainingConfig` class and puts it in `trainer` folder. - Splits `TrainingConfig` in several sub config classes. - Makes classification, regression and reconstruction...

refactoring

- adds `monai` as a dependency (in `poetry.lock` and `poetry.toml`), as well as the extras `einops` and `huggingface_hub` which are required for some networks. - creates a config class for...

refactoring

Ass mentioned in #662, there is an issue with the expression `any(["_run-" not in f.name for f in files])` line 668 of `clinicadl/utils/clinica_utils.py`: https://github.com/aramis-lab/clinicadl/blob/41824d6e608c761d9c7385819975443795e6e50a/clinicadl/utils/clinica_utils.py#L668 It seems that `files` does not...

Fixes #8085. ### Description I created: - `R2Metric` and `compute_r2_score` in `monai.metrics`, - a handler called `R2Score` in `monai.handlers`, - three unittest modules: `test_compute_r2_score.py`, `test_handler_r2_score.py` and `test_handler_r2_score_dist.py`. I also modified...

Fixes #8085. ### Description Average Precision is very similar to ROCAUC, so I was very much inspired by the ROCAUC implementation. More precisely, I created: - `AveragePrecisionMetric` and `compute_average_precision` in...

**Problem** I would like to use Maximum Mean Discrepancy metric with a kernel (e.g. gaussian kernel). According to the [reference paper](https://jmlr.org/papers/volume13/gretton12a/gretton12a.pdf), we have (eq. 3): $${{MMD}}^2 = \frac{1}{m(m-1)} \sum_{i=1}^m \sum_{j\neq...

Add example gallery and change doc structure.

refactoring

Currently, only int values are expected in masks.

refactoring

documentation
refactoring