Platform For AI:Manage third-party libraries

DSW provides the following development environments: Python2, Python3, PyTorch, and TensorFlow2.0. By default, third-party libraries are installed in Python3. If you want to install a third-party library in another development environment, you must manually switch to the environment where you want to install the third-party library. Format:

# Install a third-party library in Python3. 
pip install --user <yourLibraryName>
# Install a third-party library in Python2. 
source activate python2
pip install --user <yourLibraryName>
# Install a third-party library in TensorFlow2.0. 
source activate tf2
pip install --user <yourLibraryName>

You must replace <yourLibraryName> with the name of the third-party library that you want to install. For example, you can run the pip install --user sklearn command to install the sklearn library.

Run the following command to view installed third-party libraries:

pip list

Run the following command to uninstall a third-party library:

pip uninstall <yourLibraryName>

You must replace <yourLibraryName> with the name of the third-party library that you want to uninstall.

Specific third-party libraries cannot be uninstalled. For example, tensorflow-gpu cannot be uninstalled. You can only run the update command to install a specified version of tensorflow-gpu. The specified version must be compatible with the Compute Unified Device Architecture (CUDA) version of the DSW instance that you are using. Subscription DSW instances use CUDA 10. Pay-as-you-go DSW instances use CUDA 9. Run the following command to update an installed third-party library:

pip install --upgrade --user tensorflow-gpu=<versionNumber>

You must replace <versionNumber> with the version number of tensorflow-gpu that you want to install.