AN 993: Using Custom Models with Intel® FPGA AI Suite

Download PDF
ID 777190
Date 5/01/2023
Public
Document Table of Contents
Document Table of Contents x
1. Introduction 2. Prerequisites 3. Intel® FPGA AI Suite Model Development Overview 4. Custom Model Examples 5. Common Errors When Using an Custom Model 6. AN 993 Using Custom Models with Intel® FPGA AI Suite Revision History
4. Custom Model Examples x
4.1. Example 1: Customized ResNet-18 Model 4.2. Example 2: Customized Multilayer Perceptron (MLP) Model
1. Introduction
2. Prerequisites
3. Intel® FPGA AI Suite Model Development Overview
4. Custom Model Examples
5. Common Errors When Using an Custom Model
6. AN 993 Using Custom Models with Intel® FPGA AI Suite Revision History

4.1. Example 1: Customized ResNet-18 Model

This example removes the last Fully Connected (FC) layer to test if a performance difference exists. The ResNet-18 model is supported by Intel® FPGA AI Suite, but removing the last layer has not been tested. This example does not show how to test the modified model.

The removal is shown as an example only. The performance of this customized model has not been tested or optimized.

Model information:
  • Model: ResNet-18
  • Framework: Caffe
Figure 2. ResNet-18 Fully Connected (FC) Layer 
layer {
    bottom: "pool5"
    top: "fc1000"
    name: "fc1000"
    type: "InnerProduct"
    param {
        lr_mult: 1
        decay_mult: 1
    }
    param {
        lr_mult: 2
        decay_mult: 1
    }
    inner_product_param {
        num_output: 1000
        weight_filler {
            type: "xavier"
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    bottom: "fc1000"
    name: "prob"
    type: "Softmax"
    top: "prob"
}

This code example shows that the Fully Connected layer was removed from the model. This removal is done by deleting the lines of code from the PROTOTXT (.prototxt) file.

The modified PROTOTXT file is then used to generate the OpenVINO™ intermediate representation (IR) form of the model. To generate the IR for the modified model, run the following command: 
mo_caffe.py --input_model <path to model>.caffemodel --input_proto <path to .prototxt>

This command runs the OpenVINO™ Model Optimizer and creates three files that are the IR of this customized model: deploy.xml, deploy.mapping, and deploy.bin.

Level Two Title