Reproduce the paper

The FairMedFM paper's experiments ran in the conda environment from the repository (Python 3.8, PyTorch 2.3, transformers 4.40). Use it to reproduce the published numbers:

git clone https://github.com/FairMedFM/FairMedFM.git
cd FairMedFM
conda env create -f environment.yml
conda activate fairmedfm

In this checkout, python main.py runs the same experiments as fairmedfm run with the same arguments, for example:

python main.py --task cls --usage lp --dataset CXP --sensitive_name Sex --method erm --total_epochs 100 \
  --warmup_epochs 5 --blr 2.5e-4 --batch_size 128 --optimizer adamw --min_lr 1e-5 --weight_decay 0.05

Prepare the data as described in Datasets and the checkpoints as described in Run experiments.

Notebook tutorials

Notebook	Colab
Linear probing	Open in Colab
CLIP zero-shot and adaptation	Open in Colab
Segmentation	Open in Colab

Metric consistency

Fairness metrics are computed by fairmedfm.metrics, which reproduces the paper's original implementation to within 1e-6 (see Fairness metrics). Results from newer environments with the benchmark runner installed from GitHub can differ slightly from the paper because of newer model and library versions.