I am trying to make a neural net with kaggle’s house price prediction competition.And I have
created the model with fastai’s tabular data functionality.And this is a regressive predictions because house prices are continuous but my model is considering it as categorical and making categories for each prices from training set.So that is my problem that I don’t know how to create a regression model with tabular data.
Please tell me if you know.Thank you…!!
To create a regression task, I believe the following two conditions must be met:
1 The dependent variable is of type float
label_cls parameter is set to
FloatList as shown in the code snippet below:
data = (TabularList.from_df(df, path=path, cat_names=cat_vars, cont_names=cont_vars, procs=procs) .split_by_idx(valid_idx) .label_from_df(cols=dep_var, label_cls=FloatList, log=False) .databunch())
Thanks @Antoine I’ll try this for my model
@Antoine I tried this but it still creates catagories !
could you share the a snippet of your code where you set up your databunch? It is difficult to help without knowing what you did!
@marcmuc Thank for responding ! But I got it working infect it had already started working but I was misunderstanding…
I am Encountering the same issue… could you please shear what you have done?