BFP

Server side:

In config.yaml change the 'docker' variable to True and the 'simulation' variable to False. (lines 4 and 5)

center_number = 1
python server.py -c $center_number

Where center_number is an integer representing the centers (clients) expected by the script.

Requirements

• Docker
• Host has NVIDIA GPUs
• Data structure:
  • DATA_PATH/
    • [images_info].csv
    • images/

Installation

Preparing the set-up

docker pull registry.gitlab.bsc.es/bfp/fl_breast_mg_classification
git clone https://github.com/Linardos/BFP.git 
cd BFP

Run FD client (OPTIMAM)

Defining the environmental variables and running the client.

DATA_PATH=/home/lidia-garrucho/datasets/OPTIMAM/png_screening_cropped_fixed
CSV_FILENAME=client_images_screening.csv
IMAGES_FOLDER=images
DATA_LOADER_TYPE=optimam

docker run -it -v $DATA_PATH:/BFP/dataset -v $PWD/src:/BFP/src -e csv_path=/BFP/dataset/$CSV_FILENAME -e data_loader_type=$DATA_LOADER_TYPE -e dataset_path=/BFP/dataset/images -e server=84.88.186.195:8080 -e client_log_path=/BFP/src/client_logs -e NVIDIA_VISIBLE_DEVICES=2 --runtime=nvidia -e NVIDIA_DRIVER_CAPABILITIES=compute,utility registry.gitlab.bsc.es/bfp/fl_breast_mg_classification

OPTIMAM stge / jarv center split:

replace CSV_FILENAME=client_images_screening.csv with: CSV_FILENAME=jarv_info.csv or CSV_FILENAME=stge_info.csv

For clients using BCDR, InBreast or CMMD datasets:

InBreast:

DATA_PATH=/home/lidia-garrucho/datasets/INBREAST
CSV_FILENAME=INbreast_updated_cropped_breast.csv
IMAGES_FOLDER=AllPNG_cropped
DATA_LOADER_TYPE=inbreast

docker run -it -v $DATA_PATH:/BFP/dataset -v $PWD/src:/BFP/src -e csv_path=/BFP/dataset/$CSV_FILENAME -e data_loader_type=$DATA_LOADER_TYPE -e dataset_path=/BFP/dataset/AllPNG_cropped -e server=84.88.186.195:8080 -e client_log_path=/BFP/src/client_logs -e NVIDIA_VISIBLE_DEVICES=2 --runtime=nvidia -e NVIDIA_DRIVER_CAPABILITIES=compute,utility registry.gitlab.bsc.es/bfp/fl_breast_mg_classification

BCDR:

DATA_PATH=/home/lidia-garrucho/datasets/BCDR
DATA_LOADER_TYPE=bcdr

docker run -it -v $DATA_PATH:/BFP/dataset -v $PWD/src:/BFP/src -e csv_path=/BFP/dataset/ -e data_loader_type=$DATA_LOADER_TYPE -e dataset_path=/BFP/dataset/ -e server=84.88.186.195:8080 -e client_log_path=/BFP/src/client_logs -e NVIDIA_VISIBLE_DEVICES=2 --runtime=nvidia -e NVIDIA_DRIVER_CAPABILITIES=compute,utility registry.gitlab.bsc.es/bfp/fl_breast_mg_classification

CMMD:

DATA_PATH=/home/akis-linardos/Datasets/CMMD
CSV_FILENAME=info.csv
DATA_LOADER_TYPE=cmmd

docker run -it -v $DATA_PATH:/BFP/dataset -v $PWD/src:/BFP/src -e csv_path=/BFP/dataset/$CSV_FILENAME -e data_loader_type=$DATA_LOADER_TYPE -e dataset_path=/BFP/dataset/ -e server=84.88.186.195:8080 -e client_log_path=/BFP/src/client_logs -e NVIDIA_VISIBLE_DEVICES=2 --runtime=nvidia -e NVIDIA_DRIVER_CAPABILITIES=compute,utility registry.gitlab.bsc.es/bfp/fl_breast_mg_classification

CMMD SUBSET:

DATA_PATH=/home/akis-linardos/Datasets/CMMD_subset
CSV_FILENAME=info.csv
DATA_LOADER_TYPE=cmmd

docker run -it -v $DATA_PATH:/BFP/dataset -v $PWD/src:/BFP/src -e csv_path=/BFP/dataset/$CSV_FILENAME -e data_loader_type=$DATA_LOADER_TYPE -e dataset_path=/BFP/dataset/ -e server=84.88.186.195:8080 -e client_log_path=/BFP/src/client_logs -e NVIDIA_VISIBLE_DEVICES=2 --runtime=nvidia -e NVIDIA_DRIVER_CAPABILITIES=compute,utility registry.gitlab.bsc.es/bfp/fl_breast_mg_classification