Skip to main content
Version: 4.2

Uploading Algorithm Dependency

Environment dependencies for running algorithm models can be uploaded and managed on X-BD.

  1. Log in to supOS, and then select Data Intelligence > Data Science > Env Resources > Packages.
  2. Search for the dependencies you need to run the model to see if it already exists.
  3. For non-existing dependencies, upload them to X-BD.
  • For single dependency file, make sure it is .whl or .tar.gz file downloaded from pypi source, corresponds to specific Python version, and can be run under Linux.
  • For multiple dependency files, pack them into a .zip file with a requirements.text stating the dependency package name and version number.
  • Check the dependencies for the package before upload, and pack them together with a requirements file.
info

For the following dependencies, the requirements file content should be: cvxopt==1.3.0 dpipdeptree==2.2.1

  1. Click Add, select the file from local, and then click OK.

tip

X-BD runs a test for the uploading package, and indicates necessary information in its error message for unsuccessful cases.

  1. Search for the dependencies after upload, check whether the file is uploaded and parsed successfully.