I m trying to train an AutoNHITS and AutoRNN But I got this

Question

I m trying to train an AutoNHITS and AutoRNN But I got this error on the fit def config nhits trial return max steps 50 1000 Number of SGD steps input size 26 Size of input window lags learning rate trial suggest loguniform learning rate 1e 3 1e 1 Initial Learning rate n pool kernel size trial suggest categorical n pool kernel size 2 2 2 16 8 1 MaxPool s Kernelsize n freq downsample trial suggest categorical n freq downsample 168 24 1 24 12 1 1 1 1 Interpolation expressivity ratios val check steps 52 Compute validation every 50 steps random seed 5 scaler type trial suggest categorical scaler type standard robust stat exog list static list Static exogenous variables futr exog list future list Future exogenous variables activation trial suggest categorical activation ReLU LeakyReLU random seed tune randint 1 10 reconciliation tune choice BottomUp MinTraceOLS MinTraceWLS def config rnn trial return max steps 50 1000 Number of SGD steps input size 26 Size of input window lags learning rate trial suggest loguniform learning rate 1e 3 1e 1 Initial Learning rate n pool kernel size trial suggest categorical n pool kernel size 2 2 2 16 8 1 MaxPool s Kernelsize n freq downsample trial suggest categorical n freq downsample 168 24 1 24 12 1 1 1 1 Interpolation expressivity ratios val check steps 52 Compute validation every 50 steps random seed 5 scaler type trial suggest categorical scaler type standard robust stat exog list static list Static exogenous variables futr exog list future list Future exogenous variables encoder activation trial suggest categorical encoder activation relu tanh random seed tune randint 1 10 reconciliation tune choice BottomUp MinTraceOLS MinTraceWLS models = AutoNHITS h=horizon config=config nhits search alg=optuna samplers TPESampler backend= optuna num samples = 5 AutoRNN h=horizon config=config rnn search alg=optuna samplers TPESampler backend= optuna num samples = 5 nf = NeuralForecast models=models freq= W SUN nf fit df 1 static df=static df

José Morales · Answer

Hey What is the minimum size of your series