Hi, I'm trying to fit a StemGNN model to a dataset of 25 additional series so I have set the n_series parameter to 26 (target + additional series) with a forecasting horizon of 2 and input size of 20, however I am getting the following error when fitting the model error: RuntimeError: mat1 and mat2 shapes cannot be multiplied (832x1 and 26x1) the dataset used for fitting is of shape (433, 28) #25 features + y + ds + unique_id minimum example

StemGNN(n_series=26, h=horizon, max_steps=750, input_size=20)
nf = NeuralForecast(models=models, freq='15min')
nf.fit(df=dataset)

Do I have to name the features something in particular for it to work? I am not sure where the 832x1 comes from either TIA!