The fit function that calls _fit expects numpy arrays or dataframes. For the project I am working on, it would be preferred to provide the training data as a tensorflow dataset. Does it make sense to overwrite the _fit function to allow the usage of tensorflow dataset or would it break to many dependencies?
The fit function that calls _fit expects numpy arrays or dataframes. For the project I am working on, it would be preferred to provide the training data as a tensorflow dataset. Does it make sense to overwrite the _fit function to allow the usage of tensorflow dataset or would it break to many dependencies?