양자화, 사영, 가지치기

양자화, 사영 또는 가지치기를 수행하여 심층 신경망 압축

Deep Learning Toolbox™와 함께 Deep Learning Toolbox Model Quantization Library 지원 패키지를 사용하여 심층 신경망의 메모리 사용량과 계산 요구 사항을 줄이려면 다음과 같이 하십시오.

1차 테일러 근사를 사용하여 컨벌루션 계층에서 필터를 가지치기합니다. 그런 다음 가지치기된 신경망에서 C/C++ 또는 CUDA^® 코드를 생성할 수 있습니다.
훈련 데이터를 대표하는 데이터 세트를 사용하여 계층 활성화 부분에 주성분 분석(PCA)을 수행하여 계층을 사영하고, 계층의 학습 가능 파라미터에 선형 사영을 적용합니다. 라이브러리가 없는 C/C++ 코드 생성을 사용하여 임베디드 하드웨어에 신경망을 배포하는 경우에는 사영된 심층 신경망을 순방향 통과시키는 것이 일반적으로 더 빠릅니다.
계층의 가중치, 편향 및 활성화를 정수 데이터형으로 스케일링한 낮은 정밀도로 양자화합니다. 그런 다음 양자화된 신경망에서 C/C++, CUDA 또는 HDL 코드를 생성할 수 있습니다.
C/C++ 및 CUDA 코드 생성 시, 컨벌루션 계층의 가중치, 편향 및 활성화를 8비트로 스케일링된 정수 데이터형으로 양자화하여 컨벌루션 심층 신경망용 코드를 생성합니다. 양자화는 calibrate 함수에 의해 생성된 보정 결과 파일을 codegen (MATLAB Coder) 명령에 제공하여 수행됩니다.
quantize 함수에 의해 생성된 양자화된 심층 신경망은 코드 생성에 지원되지 않습니다.

함수

`taylorPrunableNetwork`	Network that can be pruned by using first-order Taylor approximation (R2022a 이후)
`forward`	Compute deep learning network output for training (R2019b 이후)
`predict`	Compute deep learning network output for inference (R2019b 이후)
`updatePrunables`	Remove filters from prunable layers based on importance scores (R2022a 이후)
`updateScore`	Compute and accumulate Taylor-based importance scores for pruning (R2022a 이후)
`dlnetwork`	Deep learning neural network (R2019b 이후)

`compressNetworkUsingProjection`	Compress neural network using projection (R2022b 이후)
`neuronPCA`	Principal component analysis of neuron activations (R2022b 이후)
`unpackProjectedLayers`	Unpack projected layers of neural network (R2023b 이후)
`ProjectedLayer`	Compressed neural network layer using projection (R2023b 이후)
`gruProjectedLayer`	Gated recurrent unit (GRU) projected layer for recurrent neural network (RNN) (R2023b 이후)
`lstmProjectedLayer`	Long short-term memory (LSTM) projected layer for recurrent neural network (RNN) (R2022b 이후)

`dlquantizer`	Quantize a deep neural network to 8-bit scaled integer data types (R2020a 이후)
`dlquantizationOptions`	Options for quantizing a trained deep neural network (R2020a 이후)
`calibrate`	Simulate and collect ranges of a deep neural network (R2020a 이후)
`quantize`	Quantize deep neural network (R2022a 이후)
`validate`	Quantize and validate a deep neural network (R2020a 이후)
`quantizationDetails`	Display quantization details for a neural network (R2022a 이후)
`estimateNetworkMetrics`	Estimate network metrics for specific layers of a neural network (R2022a 이후)
`equalizeLayers`	Equalize layer parameters of deep neural network (R2022b 이후)

Quantize deep neural network to 8-bit scaled integer data types (R2020a 이후)

Parameter Pruning and Quantization of Image Classification Network
Use parameter pruning and quantization to reduce network size.
Prune Image Classification Network Using Taylor Scores
This example shows how to reduce the size of a deep neural network using Taylor pruning.
Prune Filters in a Detection Network Using Taylor Scores
This example shows how to reduce network size and increase inference speed by pruning convolutional filters in a you only look once (YOLO) v3 object detection network.
Prune and Quantize Convolutional Neural Network for Speech Recognition
This example shows how to compress a convolutional neural nework (CNN) to prepare it for deployment on an embedded system.

Compress Neural Network Using Projection
This example shows how to compress a neural network using projection and principal component analysis.
Compress Network for Estimating Battery State of Charge
This example shows how to compress a neural network for predicting the state of charge of a battery using projection and principal component analysis. (R2023b 이후)
Train Smaller Neural Network Using Knowledge Distillation
This example shows how to reduce the memory footprint of a deep learning network by using knowledge distillation. (R2023b 이후)

Quantization of Deep Neural Networks
Understand effects of quantization and how to visualize dynamic ranges of network convolution layers.
Quantization Workflow Prerequisites
Products required for the quantization of deep learning networks.
Prepare Data for Quantizing Networks
Supported datastores for quantization workflows.
Quantize Multiple-Input Network Using Image and Feature Data
Quantize Multiple Input Network Using Image and Feature Data

Generate INT8 Code for Deep Learning Networks (GPU Coder)
Quantize and generate code for a pretrained convolutional neural network.
Quantize Residual Network Trained for Image Classification and Generate CUDA Code
This example shows how to quantize the learnable parameters in the convolution layers of a deep learning neural network that has residual connections and has been trained for image classification with CIFAR-10 data.
Quantize Layers in Object Detectors and Generate CUDA Code
This example shows how to generate CUDA® code for an SSD vehicle detector and a YOLO v2 vehicle detector that performs inference computations in 8-bit integers for the convolutional layers.
Quantize Semantic Segmentation Network and Generate CUDA Code
Quantize Convolutional Neural Network Trained for Semantic Segmentation and Generate CUDA Code

Quantize Network for FPGA Deployment (Deep Learning HDL Toolbox)
Reduce the memory footprint of a deep neural network by quantizing the weights, biases, and activations of convolution layers to 8-bit scaled integer data types.
Classify Images on FPGA Using Quantized Neural Network (Deep Learning HDL Toolbox)
This example shows how to use Deep Learning HDL Toolbox™ to deploy a quantized deep convolutional neural network (CNN) to an FPGA.
Classify Images on FPGA by Using Quantized GoogLeNet Network (Deep Learning HDL Toolbox)
This example show how to use the Deep Learning HDL Toolbox™ to deploy a quantized GoogleNet network to classify an image.

Generate int8 Code for Deep Learning Networks (MATLAB Coder)
Quantize and generate code for a pretrained convolutional neural network.
Generate INT8 Code for Deep Learning Network on Raspberry Pi (MATLAB Coder)
Generate code for deep learning network that performs inference computations in 8-bit integers.