issues
search
bunnech
/
cellot
Learning Single-Cell Perturbation Responses using Neural Optimal Transport
BSD 3-Clause "New" or "Revised" License
109
stars
9
forks
source link
Unable to reproduce the results of Sciplex3
#18
Closed
XiejcSYSU
closed
5 months ago
XiejcSYSU
commented
5 months ago
Hello, I trained and evaluated the performance of CellOT on the sciplex3 dataset using the following command (without changing any configuration):
python ./scripts/train.py --outdir ./results/sciplex3/drug-trametinib/model-cellot --config ./configs/tasks/sciplex3.yaml --config ./configs/models/cellot.yaml --config.data.target trametinib python ./scripts/evaluate.py --outdir ./results/sciplex3/drug-trametinib/model-cellot --setting iid --where data_space
However, I did not receive the results reported in the article. This is the result of my reproduction:
1000,all,l2-means,4.904272556304932 1000,all,l2-stds,4.1161394119262695 1000,all,r2-means,0.6526695314139299 1000,all,r2-stds,0.9440400841747668 1000,all,r2-pairwise_feat_corrs,0.43102186295905737 1000,all,l2-pairwise_feat_corrs
Hello, I trained and evaluated the performance of CellOT on the sciplex3 dataset using the following command (without changing any configuration):
python ./scripts/train.py --outdir ./results/sciplex3/drug-trametinib/model-cellot --config ./configs/tasks/sciplex3.yaml --config ./configs/models/cellot.yaml --config.data.target trametinib python ./scripts/evaluate.py --outdir ./results/sciplex3/drug-trametinib/model-cellot --setting iid --where data_space
However, I did not receive the results reported in the article. This is the result of my reproduction:
1000,all,l2-means,4.904272556304932 1000,all,l2-stds,4.1161394119262695 1000,all,r2-means,0.6526695314139299 1000,all,r2-stds,0.9440400841747668 1000,all,r2-pairwise_feat_corrs,0.43102186295905737 1000,all,l2-pairwise_feat_corrs