Accelerate training of machine learning algorithms
Achieve faster training of machine learning algorithms using Google TensorFlow on IBM PowerAI
This code pattern teaches developers to quickly train a machine learning algorithm using PowerAI virtualization software through Nimbix. You can increase speeds over a non-Power architecture when running unsupervised learning iterations using NVIDIA GPUs and the CUDA parallel computing platform.
This code pattern is designed for anyone who wants to increase their machine learning speed, showing you how to leverage IBM’s new PowerAI for machine learning. You’ll use a Jupyter Notebook to showcase an example of machine learning with a time series on IBM Power8® systems. The notebook focuses on evaluating the predictability of future financial market values in the renewable energy sector by examining related markets and sentiment detected in The New York Times news articles.
When you’ve completed this pattern, you will understand how to:
- Extract and format structured data from various external sources
- Extract and format unstructured data and use IBM Watson™ cognitive services to analyze data sentiment
- Build and train neural networks
- Display and share results in Jupyter Notebooks
This pattern will assist application developers who need to efficiently build powerful deep learning applications and improve their machine learning speeds quickly. It’s also ideal for developers who do not have extensive data science experience.
- The developer loads the provided notebook, which is run on a PowerAI system.
- As the notebook is run, it uses data from The New York Times and market data.
- The notebook uses the IBM Watson Natural Language Understanding service to analyze the text.
- The notebook uses TensorFlow and machine learning to develop models and predictions.