output file

[TheodoreZhao]

Hi Edward,
I have run your code on my dataset and got the .npz files. I find it contains 6 numpy.array variables W_emb b_output b_hidden b_emb W_output W_hidden. But on the GitHub Repo I can’t find further description about these output variables. Can you give some detailed instruction about the output? How can I get the code and visit representation?

[mp2893]

Hi Theodore,

First of all, let me explain how to obtain vector representations for each code.
If you look at Fig.2 of the paper, ReLU(W_emb x_t + b_emb) is the representation of multiple codes (since x_t will have multiple 1s). If you make x_t such that it has only a single 1, then ReLU(W_emb x_t + b_emb) will be a representation of a single code. For example, if the coordinate 0 represents "Lung cancer" and you make x_t such that <1,0,0,0,0,....,0,0>, then ReLU(W_emb x_t + b_emb) will be the vector representation of "Lung cancer"

For visit representation, you can derive the code-level representation using u_t = ReLU(W_emb x_t + b_emb), possibly with a multi-hot vector, then use v_t = ReLU(W_hidden u_t + b_hidden) to derive the visit representation.

You can also read section 3.1 of my paper. Figure 2 will also help you.

Thanks,
Ed

[mp2893]

Yeah that's correct. My mistake.