search

nf-core / deepmodeloptim

Stochastic Testing and Input Manipulation for Unbiased Learning Systems
https://nf-co.re/deepmodeloptim
MIT License
23 stars 9 forks source link

[feat] parallelize predict wrapper #178

Closed suzannejin closed 1 month ago

mathysgrapotte commented 2 months ago

@suzannejin what do you mean by this ?

suzannejin commented 2 months ago

@mathysgrapotte the PredictWrapper code is not parallelized. Meaning only one core is used to predict on new data using the trained model. Sometimes when data is big (like for example in ibis it takes longer predicting than training), we might need this.

mathysgrapotte commented 1 month ago

should be adressed in stimulus-py