hello I am getting this error when trying to perform crossva

Question

hello I am getting this error when trying to perform crossvalidation but the dataframe I am providing isnt empty ``` ValueError Traceback most recent call last Cell In 112 line 2 1 Hago el crossvalidation para los modelos de ML con los mismos parametros gt 2 crossvalidation ml colombia = ml2 cross validation 3 df=colombia df 4 h=6 5 step size = 12 6 n windows = 3 7 Asigno unique id como el indice del dataframe 8 mlcrossvalidation df set index unique id inplace=True File ~ conda envs Merck lib site packages mlforecast utils py 164 in old kw to pos lt locals gt decorator lt locals gt inner args kwargs 162 new args append kwargs pop arg names i 163 new args append kwargs pop old name gt 164 return f new args kwargs File ~ conda envs Merck lib site packages mlforecast forecast py 741 in MLForecast cross validation self df n windows h id col time col target col step size static features dropna keep last n refit max horizon before predict callback after predict callback prediction intervals level input size fitted data window size 739 should fit = i window == 0 or refit gt 0 and i window refit == 0 740 if should fit gt 741 self fit 742 train 743 id col=id col 744 time col=time col 745 target col=target col 746 static features=static features 747 dropna=dropna 748 keep last n=keep last n 749 max horizon=max horizon 750 prediction intervals=prediction intervals 751 fitted=fitted 752 753 self cv models append self models 754 if fitted File ~ conda envs Merck lib site packages mlforecast utils py 164 in old kw to pos lt locals gt decorator lt locals gt inner args kwargs 162 new args append kwargs pop arg names i 163 new args append kwargs pop old name gt 164 return f new args kwargs File ~ conda envs Merck lib site packages mlforecast forecast py 462 in MLForecast fit self df id col time col target col static features dropna keep last n max horizon prediction intervals fitted data 450 X with info y = self preprocess 451 df=df 452 id col=id col 459 return X y=True 460 461 X = X with info self ts features order gt 462 self fit models X y 463 if fitted 464 fitted values = self compute fitted values 465 X with info=X with info 466 y=y 470 max horizon=max horizon 471 File ~ conda envs Merck lib site packages mlforecast forecast py 278 in MLForecast fit models self X y 274 self models name append 275 clone model fit X loc keep y keep col 276 277 else gt 278 self models name = clone model fit X y 279 return self File ~ conda envs Merck lib site packages lightgbm sklearn py 895 in LGBMRegressor fit self X y sample weight init score eval set eval names eval sample weight eval init score eval metric early stopping rounds verbose feature name categorical feature callbacks init model 888 def fit self X y 889 sample weight=None init score=None 890 eval set=None eval names=None eval sample weight=None 891 eval init score=None eval metric=None early stopping rounds=None 892 verbose= warn feature name= auto categorical feature= auto 893 callbacks=None init model=None 894 Docstring is inherited from the LGBMModel gt 895 super fit X y sample weight=sample weight init score=init score 896 eval set=eval set eval names=eval names eval sample weight=eval sample weight 897 eval init score=eval init score eval metric=eval metric 898 early stopping rounds=early stopping rounds verbose=verbose feature name=feature name 899 categorical feature=categorical feature callbacks=callbacks init model=init model 900 return self File ~ conda envs Merck lib site packages lightgbm sklearn py 748 in LGBMModel fit self X y sample weight init score group eval set eval names eval sample weight eval class weight eval init score eval group eval metric early stopping rounds verbose feature name categorical feature callbacks init model 745 evals result = 746 callbacks append record evaluation evals result gt 748 self Booster = train 749 params=params 750 train set=train set 751 num boost round=self n estimators 752 valid sets=valid sets 753 valid names=eval names 754 fobj=self fobj 755 feval=eval metrics callable 756 init model=init model 757 feature name=feature name 758 callbacks=callbacks 759 761 if evals result 762 self evals result = evals result File ~ conda envs Merck lib site packages lightgbm engine py 271 in train params train set num boost round valid sets valid names fobj feval init model feature name categorical feature early stopping rounds evals result verbose eval learning rates keep training booster callbacks 269 construct booster 270 try gt 271 booster = Booster params=params train set=train set 272 if is valid contain train 273 booster set train data name train data name File ~ conda envs Merck lib site packages lightgbm basic py 2605 in Booster init self params train set model file model str silent 2598 self set network 2599 machines=machines 2600 local listen port=params local listen port 2601 listen time out=params get time out 120 2602 num machines=params num machines 2603 2604 construct booster object gt 2605 train set construct 2606 copy the parameters from train set 2607 params update train set get params File ~ conda envs Merck lib site packages lightgbm basic py 1815 in Dataset construct self 1812 self set init score by predictor self predictor self data used indices 1813 else 1814 create train gt 1815 self lazy init self data label=self label 1816 weight=self weight group=self group 1817 init score=self init score predictor=self predictor 1818 silent=self silent feature name=self feature name 1819 categorical feature=self categorical feature params=self params 1820 if self free raw data 1821 self data = None File ~ conda envs Merck lib site packages lightgbm basic py 1474 in Dataset lazy init self data label reference weight group init score predictor silent feature name categorical feature params 1472 self pandas categorical = reference pandas categorical 1473 categorical feature = reference categorical feature gt 1474 data feature name categorical feature self pandas categorical = data from pandas data 1475 feature name 1476 categorical feature 1477 self pandas categorical 1478 label = label from pandas label 1480 process for args File ~ conda envs Merck lib site packages lightgbm basic py 566 in data from pandas data feature name categorical feature pandas categorical 564 if isinstance data pd DataFrame 565 if len data shape = 2 or data shape 0 lt 1 gt 566 raise ValueError Input data must be 2 dimensional and non empty 567 if feature name == auto or feature name is None 568 data = data rename columns=str ValueError Input data must be 2 dimensional and non empty ```

José Morales · Answer

It probably becomes empty after dropping the null values produced by the transformations Can you try with `dropna=False`