Jian Zhou

these are saved xgboost models and you can load models like these (.save and .dump are just two different formats, you don't need both) ``` modelList = pd.read_csv(args.modelList,sep='\t',header=0) models =...

1. Conv2D is equivalent to Conv1D when used in this way 2. Forward and reverse strand predictions are usually averaged to provide better predictions, in training we use both too....

Hi, Thanks! You can get sequence class assignments from bed file here https://zenodo.org/record/7113989. Yes the most affected sequence class does not have to be the assigned sequence class, even though...

Got it. Then an option can be training a classifier model (e.g. with https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html, which supports multiclass='multinomial' ) to predict sequence class as assigned in the bed files from the...