Quantize - 3.5 English

1. Launch the Docker image:

   
   [Host]$ cd Vitis-AI
   [Host]$ ./docker_run.sh xilinx/vitis-ai-cpu:latest

2. Download the model source code package tf2_custom_op_demo.tar.gz:

   
   [Docker]$ wget https://www.xilinx.com/bin/public/openDownload?filename=tf2_custom_op_demo.tar.gz -O tf2_custom_op_demo.tar.gz
   [Docker]$ tar -xzvf tf2_custom_op_demo.tar.gz
   [Docker]$ cd tf2_custom_op_demo

3. Quantize:

   
   [Docker]$ conda activate vitis-ai-tensorflow2
   [Docker]$ bash 1_run_train.sh
   [Docker]$ bash 3_run_quantize.sh

After quantizing, the quantized model named quantized.h5 is generated in the ./quantized/ directory.