Skip to content

NeuroSyd/randomness

Repository files navigation

Recurrent Neural Network using LSTM to Evaluate Random Number Generator’s Effectiveness

A simple RNN model is designed with LSTM (Long Short Term Memory) is used to predict the next hex number with a given array of consecutive random hex numbers.

Two simple RNN models are presented as follows:

  1. Input -> LSTM -> Dense;
  2. Input -> Conv1d -> Relu -> Pooling -> LSTM -> Dense

The dataset is acquired from https://qrng.anu.edu.au/RainHex.php. There are 327399 (317399 for training, 10000 for testing) hex numbers in this dataset.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published