models.md

Models Module

The models.py file defines Mixture Density Network (MDN) network with a 4D Multivariate Normal Distribution neural network architecture using quantized layers. The implementation uses QKeras to quantize the weights and activations of the network.

Functions

CreateModel(shape, n_filters, pool_size)

Creates a quantized neural network model for regression task with quantized layers and activations as in Model. The model has 14 output nodes with 4 being the target variables and the rest 10 being the co-variances.

• Arguments:
  • shape (tuple): Input shape (e.g., (13, 21, 2)/ (13, 21, 20)).
  • n_filters (int): Number of filters for the convolutional layers.
  • pool_size (int): Size of the pool for the pooling layer.
• Returns:
  • keras.Model: A compiled Keras model instance.
• Example:

 from models import CreateModel

 model = CreateModel((13, 21, 2), n_filters=5, pool_size=3)
 model.summary()

---

### Additional Helper Functions

## `conv_network(var, n_filters=5, kernel_size=3)`
Defines the convolutional network block, with quantized layers and activations.

- **Arguments**:
 - `var (InputLayer: tf.Tensor)`: Input tensor.
 - `n_filters (int)`: Number of filters.
 - `kernel_size (int)`: Kernel size.

- **Returns**:
 - `tf.Tensor`: Output tensor.

## `var_network(var, hidden=10, output=2)`
Defines the dense network block, with quantized layers and activations.

- **Arguments**:
 - `var (InputLayer: tf.Tensor)`: Input tensor.
 - `hidden (int)`: Number of hidden units.
 - `output (int)`: Number of output units.

- **Returns**:
 - `tf.Tensor`: Output tensor.