Prepare Data for Quantizing Networks

Quantizing a deep neural network in MATLAB^® requires calibration by exercising the network to determine the dynamic ranges of its weights, biases, and activations. After calibrating, you can validate the performance of the quantized network and inspect the reduction in size and performance (if any) of the network. However, not all data formats supported for training and prediction are supported for quantization workflows.

Datastores

The calibrate and validate functions require the calibration data and validation data to be passed as a datastore object. Many built-in datastores are supported for quantization workflows. The supported datastores for particular networks are detailed below. Passing data as a numeric array or table is not supported.

For more information on using datastore objects for training and prediction, see Datastores for Deep Learning.

Choose a Built-In Datastore

The following table describes which built-in datastores are supported for quantization of different network types.

Network Type	Supported Datastores
Image classification or regression	`imageDatastore` `augmentedImageDatastore` `TransformedDatastore`
Object detection	`CombinedDatastore` `TransformedDatastore`
Semantic segmentation	`pixelLabelDatastore` (Computer Vision Toolbox) `CombinedDatastore`
Sequence and numeric feature classification and regression	`CombinedDatastore` `TransformedDatastore`

Calibration and Validation

To be a valid input for validation, the read function of a datastore must return data either as a cell array or a table with the responses as the second column.

As calibration exercises the network and collects the dynamic range statistics, the calibration data does not require responses.

The format of the predictors used for calibration and validation depends on the type of input.

Data	Format of Predictors
2-D image	h-by-w-by-c numeric array, where h, w, and c are the height, width, and number of channels of the image, respectively.
Vector sequence	c-by-s matrix, where c is the number of features of the sequence and s is the sequence length. Each sequence must have the same length.
2-D image sequence	h-by-w-by-c-by-s array, where h, w, and c correspond to the height, width, and number of channels of the image, respectively, and s is the sequence length. Each sequence in the mini-batch must have the same sequence length.
Features	1-by-c column vector, where c is the number of features.

For data returned in tables, the elements must contain a 1-by-1 cell array containing a numeric array.

For validation, the datastore must return responses. The format of the responses depend on the type of task.

Task	Format of Responses
Classification	Categorical scalar
Regression	Scalar Numeric vector 3-D numeric array representing an image
Segmentation	The default format returned by reading from a `pixelLabelDatastore` (Computer Vision Toolbox)
Object detection	Categorical array

Transform and Combine Datastores

Deep learning frequently requires data to be preprocessed and augmented before data is in an appropriate form to input to a network. The transform and combine functions of datastores are useful in preparing data to be fed into a network.

Transform Datastores

A transformed datastore applies a particular data transformation to an underlying datastore when reading data. To create a transformed datastore, use the transform function and specify the underlying datastore and the transformation.

For validation, reading from a transformed datastore must return a cell array with the predictors as the first column and the responses as the third column.

Combine Datastores

The combine function combines data from multiple datastores. Operating on the resulting CombinedDatastore, such as resetting the datastore, performs the same operation on all of the underlying datastores.

The calibrate and validate functions support only CombinedDatastore objects with two underlying datastores. The first datastore must be an imageDatastore, augmentedImageDatastore, or arrayDatastore and the second datastore must be a pixelLabelDatastore (Computer Vision Toolbox) or arrayDatastore.