## Model description

This model demonstrates real-valued non-volume preserving (real NVP) transformations, a tractable yet expressive approach to modeling high-dimensional data.

This model is used to map a simple distribution - which is easy to sample and whose density is simple to estimate - to a more complex one learned from the data. This kind of generative model is also known as "normalizing flow". The latent distribution we wish to map to in this example is Gaussian.

## Training and evaluation data

This model is trained on a toy dataset, the make_moons from sklearn.datasets.

## Requirements

This model requires

- Tensorflow 2.9.1
- Tensorflow probability 0.17.0

## Loading this Model

Only the model weights are stored. To load this model, use `load_model`

function from load_model.py.

## Original Data Distribution and Latent Distribution

### Training hyperparameters

The following hyperparameters were used during training:

name | learning_rate | decay | beta_1 | beta_2 | epsilon | amsgrad | training_precision |
---|---|---|---|---|---|---|---|

Adam | 9.999999747378752e-05 | 0.0 | 0.8999999761581421 | 0.9990000128746033 | 1e-07 | False | float32 |

- Downloads last month
- 2