snap-stanford / GEARS

GEARS is a geometric deep learning model that predicts outcomes of novel multi-gene perturbations
MIT License
189 stars 38 forks source link

custom GO graph location #60

Closed Jiahao-Gao closed 2 weeks ago

Jiahao-Gao commented 5 months ago

line212 in utils.py fname under make_GO should use data_path variable, not hardcoded ./data/

yanwuanxin commented 5 months ago

I encountered an error at model_initialize > get_similarity_network > make_GO,after run pert_data = PertData(data_path = outdir, default_pert_graph = False). I made the following modifications, error message did not appear. image