I am working on a problem where the input contains both continuous and categorical features, and output is a numerical value.
I convert each categorical feature into a fixed-dimension vector by one-hot or embedding layer, and concatenate the generated vectors and continuous features. The concatenated vector is sent to fully connected layers. And the last layer outputs a vector with one dimension.
This approach may be intuitive and simple. Is there any more effective way to deal with input with both continuous and categorical features?
