search

PaddlePaddle / PaddleHelix

Bio-Computing Platform Featuring Large-Scale Representation Learning and Multi-Task Deep Learning “螺旋桨”生物计算工具集
Other
1.02k stars 225 forks source link

Usage problem #299

Open ActionBeam opened 5 months ago

ActionBeam commented 5 months ago

There are two files in pdbbind_core.json: label_file and complex_id_file, but they do not seem to be available in your tutorials. Additionally, posebusters.json contains the same content as pdbbind_core.json.

Noisyntrain commented 5 months ago

Hi ActionBeam, the 'label_file' and 'complex_id_file' is used for the data generating, as the processed npzs are provided(in 'cache_dir'), they are not needed during inference. The posebsuter.json's issue has been fixed, you can pull to latest version to check. Hope this can be helpful to you.

ActionBeam commented 5 months ago

I noticed that latest version only updated 'reproduce_posebuster.sh', I mean the content of 'posebuster.json' is: { "test": { "posebuster" : { "data_dir": "../data/PDBbind_v2020/complex/", "label_file": "../data_info/INDEX_general_PL_data.2020", "complex_id_file": "../data/PDBbind_v2020/complex_ids/casf_test.txt", "cache_dir": "data/processed/posebuster_processed", which "data_dir" and "label_file" are same as "pdbbind_core.json"