Simple Feed Forward - MXNet

Simple Feed Forward is a simple neural network that forecasts probability distributions for the next forecast horizon values, given the preceding context length values. This implementation relies on MXNet.

Training parameters

Batch size

The size of the batches to be used for training and prediction.

Epochs

Use default nb. of batches per epoch

Default value is the number of batches for which on average each time step appears in one sample per epoch.

Number of batches per epoch

Random state

Using a fixed random seed allows for reproducible result.

Model parameters

Use default context length

Set the context length to the model's default value (context length = forecast horizon).

Output distribution

Try Student’s t distribution Try Gaussian distribution Try Negative binomial distribution

Distribution to fit.

Batch normalization

Try with batch normalization Try without batch normalization

Normalize the layers' inputs by re-centering and re-scaling. This can make deep neural networks faster and more stable.

Mean scaling

Try with mean scaling Try without mean scaling

Scale the network input by the data mean and the network output by its inverse.

Hidden layer sizes

Number of nodes in each hidden layer. Specify one value per layer.

Number of parallel samples

Number of evaluation samples per time series, to increase parallelism during inference.