Session: Cooking with Granite : Granite AI Models Workshop
The Granite family of foundation models span an increasing variety of modalities, including language, code, time series, and science (e.g., materials) – with much more to come. This walk-through of the Granite Community’s Cookbooks for the IBM Granite models will give you hands-on experience using the Granite models (including those specialized for code and time-series) in short recipes. The session will start with simple use cases that demonstrate zero-shot inference with the code and time-series models. The recipes will grow more complex to cover data preparation, RAG, tool use, and agentic workflows. By the end of this workshop, you will be able to:
- Summarize a text document using text summarization
- Generate specific information from a large document using the RAG technique
- Predict future trends using time series forecasting
- Generate programming code (Bash) by prompting a code model
The workshop https://ibm.github.io/granite-workshop/
The prerequisites
a device with Python 3.11 and Jupyter Notebook installed, or
a device with a browser and GitHub and Google accounts
To set-up environment with Python 3.11:
python3.11 -m venv –upgrade-deps .venv
source .venv/bin/activate
python3 -m pip install –require-virtualenv notebook ipywidgets
