#Pre-installation Preparation

#Development Environment Preparation

#Development Machine Preparation

In order to use the toolchain smoothly, we recommends that the development machine you choose should meet the following requirements:

For more information about CUDA compatibility with graphics cards, refer to NVIDIA website information.

#Docker Container Preparation

Horizon requires the following Docker base environment, please complete the installation on your host computer in advance.

• Docker (20.10.10 and above, 20.10.10 is recommended), refer to the Docker Installation Manual.
• NVIDIA Container Toolkit (1.16.2 and above, 1.17.8 is recommended), refer to installation manual NVIDIA Container Toolkit installation manual.

After completing the installation of the Docker environment, remember to add non-root users into Docker users group by running below command:

sudo groupadd docker
sudo gpasswd -a ${USER} docker
sudo service docker restart

#Local Manual Installation Preparation

This section introduces the environment-related dependencies and descriptions for each of the two quantization schemes and Horizon open-source efficient model training.

#PTQ Quantization Environment Dependence

The PTQ scheme has the following software dependencies on the base software of the development machine operating environment:

• Operating system: Ubuntu22.04
• Python3.10
• libpython3.10
• python3-devel
• python3-pip
• gcc&g++: 12.2.1
• graphviz

#QAT Quantization Environment Dependence

The QAT quantization environment is installed in the local environment and you need to ensure that the following basic environmental conditions are met.

The environmental dependencies required for the quantitative training tool to be trained are listed below:

After completing the training of the QAT model, you can install the relevant toolkits in the current training environment and complete the subsequent model conversion directly through the interface call.

#Efficient Model Floating-point Training Environment Instruction

Horizon provides the source code of several open-source efficient models in samples/ai_toolchain/horizon_model_train_sample. For information on the floating-point and QAT base environment, refer to section QAT Quantization Environment Deployment.

#Runtime Environment Preparation

Once the model has been quantized, the compiled model can be deployed on the dev board environment for inference and execution.

To deploy the runtime environment, you need to prepare a dev board with the system image programmed, and then copy the relevant supplementary files to the dev board.

Before this, you need to verify the usability of the dev board, and if there are no issues, you can program the available system image to the dev board. Please refer to the description in the Release Notes section for the corresponding version range of the system image.

For more relevant information such as the acquisition and upgrade guidance of the system image BSP version, please contact the Horizon System Software Technical Support Staff for assistance.