Logo
Akida 1.8.13
  • Overview
  • Installation
    • Requirements
    • Quick installation
    • Running examples
  • User guide
    • Getting started
      • For beginners
      • For users familiar with deep-learning
    • Akida user guide
      • The Akida Execution Engine
        • 1. The Spiking Neural Network model
        • 2. Input data format
        • 3. Determine training mode
        • 4. Interpreting outputs
      • Neural Network model
        • Specifying the Neural Network model
        • Saving and loading
        • Input layer types
        • Data-Processing layer types
      • Using Akida Unsupervised Learning
        • Learning constraints
        • Compiling a layer
        • Learning parameters
    • CNN2SNN toolkit
      • Overview
        • Conversion workflow
        • Typical training scenario
        • Design compatibility constraints
        • Quantization compatibility constraints
        • Command-line interface
      • Layers Considerations
        • Supported layer types
        • CNN2SNN Quantization-aware layers
        • Training-Only Layers
        • First Layers
        • Final Layers
      • Tips and Tricks
    • Akida models zoo
      • Overview
      • Command-line interface for model creation
      • Command-line interface for model training
        • CIFAR10 training and tuning
        • UTK Face training
        • KWS training
        • YOLO training
      • Layer Blocks
        • conv_block
        • dense_block
        • separable_conv_block
    • Hardware constraints
      • Akida NSoC (Pre-production)
        • InputConvolutional
        • Convolutional
        • SeparableConvolutional
        • FullyConnected
      • Akida NSoC (Production)
        • InputConvolutional
        • Convolutional
        • SeparableConvolutional
        • FullyConnected
    • Akida versions compatibility
      • Upgrading models with legacy quantizers
  • API reference
    • Akida Execution Engine
      • Model
      • Layer
      • LayerStatistics
      • InputData
      • InputConvolutional
      • FullyConnected
      • Convolutional
      • SeparableConvolutional
      • Concat
      • Dense
      • Backend
      • ConvolutionMode
      • PoolingType
      • LearningType
      • Compatibility
    • CNN2SNN
      • Tool functions
        • quantize
        • quantize_layer
        • convert
        • check_model_compatibility
        • load_quantized_model
        • load_partial_weights
      • Quantizers
        • WeightQuantizer
        • LinearWeightQuantizer
        • StdWeightQuantizer
        • TrainableStdWeightQuantizer
        • MaxQuantizer
        • MaxPerAxisQuantizer
      • Quantized layers
        • QuantizedConv2D
        • QuantizedDepthwiseConv2D
        • QuantizedDense
        • QuantizedSeparableConv2D
        • QuantizedActivation
        • ActivationDiscreteRelu
        • QuantizedReLU
    • Akida models
      • Layer blocks
        • conv_block
        • separable_conv_block
        • dense_block
      • Model zoo
        • Mobilenet
        • DS-CNN
        • VGG
        • YOLO
  • Examples
    • General examples
      • GXNOR/MNIST inference
        • 1. Dataset preparation
        • 2. Load the pre-trained Akida model
        • 3. Show predictions for a single image
        • 4. Check performance
      • DS-CNN CIFAR10 inference
        • 1. Dataset preparation
        • 2. Create a Keras DS-CNN model
        • 3. Quantized model
        • 4. Pretrained quantized model
        • 5. Conversion to Akida
      • MobileNet/ImageNet inference
        • 1. Dataset preparation
        • 2. Create a Keras MobileNet model
        • 3. Quantized model
        • 4. Pretrained quantized model
        • 5. Conversion to Akida
      • DS-CNN/KWS inference
        • 1. Load the preprocessed dataset
        • 2. Load a pre-trained native Keras model
        • 3. Load a pre-trained quantized Keras model satisfying Akida NSoC requirements
        • 4. Conversion to Akida
      • Regression tutorial
        • 1. Load the dataset
        • 2. Load a pre-trained native Keras model
        • 3. Load a pre-trained quantized Keras model satisfying Akida NSoC requirements
        • 4. Conversion to Akida
        • 5. Estimate age on a single image
      • Transfer learning with MobileNet for cats vs. dogs
        • Transfer learning process
        • 1. Load and preprocess data
        • 2. Modify a pre-trained base Keras model
        • 3. Train the transferred model for the new task
        • 4 Quantize the top layer
        • 5. Convert to Akida
        • 6. Plot confusion matrix
      • YOLO/PASCAL-VOC detection tutorial
        • 1. Introduction
        • 2. Preprocessing tools
        • 3. Model architecture
        • 4. Training
        • 5. Performance
        • 6. Conversion to Akida
    • CNN2SNN tutorials
      • CNN conversion flow tutorial
        • 1. Load and reshape MNIST dataset
        • 2. Model definition
        • 4. Model quantization
        • 5. Model fine tuning (quantization-aware training)
        • 6. Model conversion
      • Advanced CNN2SNN tutorial
        • 1. Design a CNN2SNN quantized model
        • 2. Weight Quantizer Details
        • 3. Quantized Activation Layer Details
    • Edge examples
      • Akida vision edge learning
        • 1. Dataset preparation
        • 2. Prepare Akida model for learning
        • 3. Edge learning with Akida
      • Akida edge learning for keyword spotting
        • 1. Edge learning process
        • 2. Dataset preparation
        • 3. Prepare Akida model for learning
        • 4. Learn with Akida using the training set
        • 4. Edge learning
      • Tips to set Akida learning parameters
        • 1. Akida learning parameters
        • 2. Create Akida model
        • 3. Estimate the required number of weights of the trainable layer
        • 4. Estimate the number of neurons per class
  • Changelog
  • Support
  • License
Akida Examples
  • »
  • User guide

User guideΒΆ

  • Getting started
    • For beginners
    • For users familiar with deep-learning
  • Akida user guide
    • The Akida Execution Engine
      • 1. The Spiking Neural Network model
      • 2. Input data format
        • Data format
        • Input events buffering
        • Data frames parallelization
      • 3. Determine training mode
        • Akida Unsupervised learning
        • CNN to SNN conversion
      • 4. Interpreting outputs
    • Neural Network model
      • Specifying the Neural Network model
      • Saving and loading
      • Input layer types
      • Data-Processing layer types
        • Activation parameters
        • Pooling parameters
    • Using Akida Unsupervised Learning
      • Learning constraints
      • Compiling a layer
      • Learning parameters
  • CNN2SNN toolkit
    • Overview
      • Conversion workflow
      • Typical training scenario
      • Design compatibility constraints
      • Quantization compatibility constraints
      • Command-line interface
    • Layers Considerations
      • Supported layer types
      • CNN2SNN Quantization-aware layers
      • Training-Only Layers
      • First Layers
        • Input Scaling
      • Final Layers
    • Tips and Tricks
  • Akida models zoo
    • Overview
    • Command-line interface for model creation
    • Command-line interface for model training
      • CIFAR10 training and tuning
      • UTK Face training
      • KWS training
      • YOLO training
    • Layer Blocks
      • conv_block
      • dense_block
      • separable_conv_block
  • Hardware constraints
    • Akida NSoC (Pre-production)
      • InputConvolutional
      • Convolutional
      • SeparableConvolutional
      • FullyConnected
    • Akida NSoC (Production)
      • InputConvolutional
      • Convolutional
      • SeparableConvolutional
      • FullyConnected
  • Akida versions compatibility
    • Upgrading models with legacy quantizers
Next Previous

© Copyright 2020, BrainChip Holdings Ltd. All Rights Reserved.