page-brochureware.php

IBM PowerAI developer portal

Learn about deep learning and PowerAI. Create something amazing.

Cognitive code patterns

Get your hands dirty by checking out these real world examples and tutorials that highlight PowerAI and PowerAI Vision in action.

Model for counting cars

This Code Pattern demonstrates how to create a video car counter using PowerAI Vision Video Data Platform, OpenCV and a Jupyter Notebook. You’ll use a little manual labeling and a lot of automatic labeling to train an object classifier to recognize cars on a highway. You’ll load another car video into a Jupyter Notebook where you’ll process the individual frames and annotate the video.

Integrate inference on mobile phones

This tutorial will show you how to build and deploy a PowerAI Vision model and use it in an iOS app. You’ll learn how to:

  1. Create a data set
  2. Train a model for image classification
  3. Deploy it to a Web API
  4. Integrate the API into an iOS App

Locate and count items with object detection

Imagine that you’re a supplier of an item (such as Coca-Cola) and you want to know how many bottles are on a store’s shelf. You can build an app that helps you do just that. PowerAI Vision uses deep learning to create trained models based on images that you upload and label. You don’t need to write any code to train, deploy, and test a new object detection model. You simply upload the images, use your mouse to label the objects in your images, and then let PowerAI Vision do the learning.

Accelerate training of machine learning algorithms

Quickly train a machine learning algorithm using PowerAI through Nimbix Cloud. Start with a Jupyter notebook and use machine learning to investigate the predictability of future financial market shifts in the “renewable energy” sector by examining related markets and sentiment detected in The New York Times articles.

Leverage POWER8 with NVLink to dramatically decrease your training time vs. non-POWER architectures.

Image recognition training with PowerAI notebooks

Use transfer learning to create your own image classifier. In this pattern, you’ll use a model designed to recognize houses to become a classifier of houses with swimming pools.

  • Load and run a Jupyter Notebook with PowerAI on Nimbix Cloud
  • Use transfer learning to leverage the TensorFlow Inception model to create a custom classifier from a set of images
  • Test and demonstrate the resulting classifier

SETI Signal Classification on PowerAI with Multi GPU

Each night, using the Allen Telescope Array (ATA) in northern California, the Search for Extra Terrestrial Intelligence (SETI) Institute scans the sky at various radio frequencies, observing star systems with known exoplanets, searching for faint but persistent signals. Framing the radio signal data into spectrogram (a 2D visual representation), we can convert the problem into something akin to image classification. This code pattern demonstrates the use of multi-GPUs and Spark to accelerate the data preparation and training process.