Code patterns

Roadmaps for solving complex programming challenges.

Patterns give you a 360-degree view of the underlying code, including overviews, architecture diagrams, process flows, repo pointers, and additional reading.

Train and evaluate an audio classifier

Train a deep learning model to classify audio embeddings on Watson Machine Learning and perform inference/evaluation with Watson Studio.


Train a model on fashion dataset

Leverage Tensorflow and Fabric for Deep Learning to train and deploy Fashion MNIST model on Kubernetes.


Integrate adversarial attacks into a model training pipeline

Use a Jupyter Notebook to integrate the Fast Gradient Method from the Adversarial Robustness Toolbox into a model training pipeline leveraging FfDL.


Create a web app to interact with machine learning generated image captions

Use an open source image caption generator deep learning model to filter images based on their content in a web application.


Create a virtual help desk

Use Node.js, Watson Assistant, and Watson Discovery services to create a virtual help desk to interact with end users for simple FAQ answer discovery.


Create a smarter procurement system with Watson services

Use Watson Knowledge Studio (WKS) to build a custom model to better categorize and structure procurement-related data for Watson Discovery to process and enrich so you can make purchasing decisions more quickly and accurately.


Analyze bank marketing data using XGBoost to gain insights into client purchases

Use machine learning to predict a bank client's CD purchase with XGBoost, scikit-learn, and Python in IBM Watson Studio.


Analyze product reviews and generate a shopping guide

Create a Node.js app that takes Amazon reviews and feeds them into the Watson Natural Language Understanding service, which will determine the average review sentiment for you.


Build an iOS game powered by Core ML and Watson Visual Recognition

Use Watson Visual Recognition and Core ML to create a Kitura-based iOS game that has a user search for a predetermined list of objects.


Determine target audience and run marketing campaigns

Pattern demonstrates the methodology to determine target audience and run marketing campaigns using Watson Studio and Watson Campaign Automation.


Detect, track, and count cars in a video

Use automatic labeling to create a model from a video, then use the model to annotate a video.


Develop a web-based mobile health app that uses machine learning

Deploy and use a web-based health app on your smartphone using Watson services on IBM Cloud and IBM Watson Studio.


Classify ICD-10 data with Watson

A simple web app shows how Watson's Natural Language Classifier (NLC) can classify ICD-10 code.


Create an app to get stock information, prices, and sentiment

Use IBM Watson Discovery, CloudantDB, Node.js, and Alpha Vantage to create a web app to monitor sentiment, price, and news for individually listed stocks.


Optimize your visual recognition classification

Build training and prediction apps with Java, Watson IoT Platform, Node-RED, and Watson Visual Recognition service.