Machine Learning Environment Configuration

Before using the machine learning feature, you need to configure the Python environment in advance. At the same time, the machine learning feature has certain requirements on the software and hardware environment of the computer. For environment configuration under Linux systems, please refer to the Machine Learning Environment Configuration_Linux page.

Recommended Basic Hardware Configuration

  • NVIDIA graphics card;
  • Video memory ≥ 10G, minimum requirement is 6G, object extraction and AI-assisted drawing should not be lower than 8G;
  • Latest graphics card driver (if it cannot be automatically updated, you can download it from the NVIDIA official website. Download link: https://www.nvidia.cn/Download/index.aspx?lang=cn

Configure Resource Environment

Download the Windows version of the AI extension package. Download link: https://pan.baidu.com/s/10Ec2T7HcHG0UAXR2D59UAw, extraction code: rdpd.

Notes:

After V12.0.1, the AI extension package no longer supports the PaddleSegmodel algorithm of the sample library AI Label. If needed, please download the V12.0.0 or V12.0.1 version of the extension package. Download links:

After downloading, extract it to the root directory of the SuperMap iDesktopX product package.

The SuperMap iDesktopX AI extension package provides materials such as the Python environment for AI analysis, machine learning resources, and video effects resources. Detailed descriptions are as follows:

  • resources_ml: Machine learning resource package, contains resources such as sample data and model configuration files for machine learning features;
  • support:
    • MiniConda: GPU runtime environment for AI analysis;
  • templates: Video effects resources;

Configure Python Environment

  1. Open SuperMap iDesktopX, click the Start -> Browse -> Python button, and in the left toolbar of the Python window, click the Python Environment Management button.
  2. In the Python Environment Management dialog, select the SuperMap iDesktopX product package/support/MiniConda/conda/python.exe environment, click the OK button, and restart the Python environment.

Related Content

Machine Learning Environment Configuration_Linux

Imagery Analysis