https://github.com/nixtla logo
#neural-forecast
Title
# neural-forecast
m

Milim Kim

03/27/2023, 8:23 AM
Hi, I'm using nhits library well, but I have one simple question. I need predictions for the train and validation datasets. ————————————————— model = NHITS(futr_exog_list = cols, h = 12) nf = NeuralForecast(models = [model], freq = ‘M’) nf.fit(df = df_train, val_size = val_size) yhat_valid = nf.predict(futr_df = df_train) ————————————————— These values seem to be validation prediction values, but can i obtain the prediction values of train data?
c

Cristian (Nixtla)

03/27/2023, 3:03 PM
Hi @Milim Kim, you can use the
predict_rolled
method. We are additionally working on a
predict_insample
method to make it more simple
👍 1
5 Views