We provide a config file and a pretrained model to reproduce the experiment on pMTnet reported in our paper.
-
In the provided config file at
<PATH_TO_REPO>/sample_configs/pmtnet.json
, substitute all the occurences of<PATH_TO_REPO>
to the actual path where you cloned this repository. -
In the provided config file at
<PATH_TO_REPO>/sample_configs/pmtnet.json
, change theroot
entry to a folder where you want to store the results. -
(optional) If you installed
DECODE
in a conda environment, activate the environment, e.g. (if your environment is calleddecode
)
conda activate decode
- Run
DECODE
run_interpretability_pipeline --config <PATH_TO_REPO>/sample_configs/pmtnet.json