Multiple Inputs and Mixed Data

Part 1 (2019)
1

I am looking for ways to combine multiple inputs including numerical data, categorical data and image data for training a model. Numerical and categorical data include features like:

  1. Number of bedrooms
  2. Number of bathrooms
  3. Area (i.e., square footage)
  4. Zip code

Images are of four different sections of a house like the following:

keras_regression_cnns_houses.jpg600Ɨ600
.
Refer here for more on the dataset. The image example shown above is also a montage which does not come by default.

The aim is to combine this image data and the other numerical and categorical data mentioned above and train a model to predict house price. Any direction to for preparing a fastai compatible data for this purpose would be highly helpful.

5 Likes
2

iā€™m looking to do something similar. have you found a way to do this with fastai ?

2 Likes