Overview
TensorFlow is an open source library to help you develop and train machine learning models.
RocketCE includes several packages from the TensorFlow ecosystem and these packages are available in channel rocketce of anaconda.org.
First release of RocketCE is based on OpenCE Release v1.4.1. In this version of RocketCE, there is TensorFlow 2.6.2.
These packages are supported for both Power 9 and Power 10 Systems of IBM with linux-ppc64le.
CPU Support
RocketCE includes a version of TensorFlow built without GPU support. This inclusion allows for training and inferencing to be done on Power 9 and Power 10 systems that do not have GPUs, or on systems where you want to train and inference without using the GPUs.
GPU Support
This version of Tensorflow will have support for GPU based operations. This can be used in Power 9 systems.
Installation instructions
CPU Only Variant
To install Tensorflow built for CPU support run the following command:
- conda install -c rocketce tensorflow-cpu=2.6.2
Once the installation is success, validate the installation by running the following command which lists the package tensorflow which is pulled from channel rocketce
- Conda list tensorflow-cpu=2.6.2
GPU Variant
To install Tensorflow built for CPU support run the following command:
- conda install -c rocketce tensorflow=2.6.2
Once the installation is success, validate the installation by running the following command which lists the package tensorflow which is pulled from channel rocketce
- Conda list tensorflow=2.6.2
TensorFlow conda packages
Following packages are available in rocketce channel and gets installed with tensorflow-cpu / tensorflow package
- tensorflow-base
- _tensorflow_select
- tensorboard
- tensorboard-plugin-wit
- tensorflow-estimator
Following packages are available in rocketce channel and which you can install to support tensorflow operations
1. tensorflow-text
2. tensorflow-hub
More information
The Tensorflow has various information, including tutorials, how to documents, and a getting started guide.
Additional tutorials and examples are available from the community, for example:
------------------------------
Uvaise Ahamed
Rocket Internal - All Brands
------------------------------