details about binarization

As far as I am understanding the paper and the implementation, the input image is not binarized, the weight is binarized by bnn binary operation, while the outputs are still real number, am I correct?

The issue asks for clarification on binarization in the model. Start by reading the paper linked in the repository description and the model implementation in the codebase. Look for binarization operations in the network layers. Determine if the question is about input preprocessing, weight binarization, or activation outputs. The answer should be documented in the README or code comments.

索引モデルが issue の本文から書いたものです。