Issue with temperature (float) values as extra_features

[fsGORM]

Hi

I'm trying to add temperature averages as extra_features in my MMM model, however when I fit the model it returns an error that says 'tuple index out of range'. My extra_features set contains two dataframe columns (with float point values) and I've saved my extra_features variable to be a list containing said columns, but when I try to fit a model with the scaled version of extra_features, it tells me the tuple index is out of range.

Anyone know a fix to this?

[statm3n]

extra_features should be in jnp.ndarray for fit. Maybe check if your extra_features are in the right format?

lightweight_mmm/lightweight_mmm/utils.py

Line 195 in 8152c43

def dataframe_to_jax(

can help convert pandas dataframe to jax arrays.

[fsGORM]

I've tried doing that with
extra_features_col = jnp.array(data["gns_temp"].values)
LightweightMMM expects extra_features_plates_shape = (extra_features.shape[1],)
Which is where it throws the error.
The shape of my array: (103,)