# NN_GWTC

**Repository Path**: zhoub86/NN_GWTC

## Basic Information

- **Project Name**: NN_GWTC
- **Description**: Simulations for the paper "Deep Learning for the Gaussian Wiretap Channel" with Tensorflow 2
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2021-03-18
- **Last Updated**: 2021-03-18

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# NN_GWTC
Simulations for the paper "Deep Learning for the Gaussian Wiretap Channel" with Tensorflow 2

Requirements: The code needs a package for an equal cluster-size version of K-Means.
I am using the code from https://github.com/ndanielsen/Same-Size-K-Means for that.
Note that newer versions of K-Means/Sci-kit learn require that the fit function of the original k-means
gets a sample_weight vector. This leads to breaking changes within the same-size kmeans code. 
A quick fix is to introduce a uniform sample weighting via:

        n_samples_var = X.shape[0]
        sample_weight = np.ones(n_samples_var, dtype=X.dtype)