AMCT (TensorFlow)

Use the AMCT installation user to install the dependent software. If the installation user is a non-root user, ensure that the user has the sudo permission. Run the su - username command to switch to the non-root user and run the following commands. If the installation user is root, delete --user from the command for installing the dependent software. .

**Table 1** Dependency list
Dependency	Version	Installation Command
Python	3.9.2 or 3.7.5	For Ubuntu, Kylin V10 SP1, openEuler, and UOS, see Python 3.9.2 Installation on Ubuntu.
TensorFlow	2.6.5 or 1.15.0	Install the CPU or GPU version as needed. CPU version of TensorFlow 2.6.5 python3 -m pip install tensorflow-cpu==2.6.5 --user GPU version of TensorFlow 2.6.5 python3 -m pip install tensorflow-gpu==2.6.5 --user CPU version of TensorFlow 1.15 python3 -m pip install tensorflow-cpu==1.15.0 --user GPU version of TensorFlow 1.15 python3 -m pip install tensorflow-gpu==1.15.0 --user NOTICE: If different TensorFlow versions are used, ensure that the Python version in your installation command is correct. (Change the environment variables to match the Python version in use.) Currently, quantization is supported only in TensorFlow 1.15.0, whereas inferences during quantization can be run in both TensorFlow 1.15.0 and 2.6.5.
numpy	1.16.0–1.23.5 or 1.20.0–1.23.5	pip3 install numpy==1.16.0 --user Or pip3 install numpy==1.20.0 --user
protobuf	3.13.0+	pip3 install protobuf==3.13.0 --user

Parent topic: Dependency Installation