Welcome to the kluster.ai cookbook! This repository is a collection of guides and examples designed to help developers unlock the full potential of the kluster.ai tool.
To run the notebooks in this cookbook, you must have a kluster.ai API key. You can get one here.
| Notebook | Description | Start here |
|---|---|---|
| Getting started | Quick start guide to getting started from scratch, submitting your first Batch job, and exploring the basics. |  |
| Text classification | Tutorial on using the kluster.ai Batch API for text classification, featuring an example with a movie dataset categorized into predefined genres. | |
| Text classification with Curator | Demonstration of how to use bespokelabs-curator to handle batch text classification requests to the kluster.ai Batch API. This example applies movie genre categorization using the IMDb dataset with minimal setup. | |
| Sentiment analysis | Application of the kluster.ai Batch API for sentiment analysis, demonstrated with data from Amazon musical instrument reviews | |
| Keyword extraction | Tutorial on keyword extraction using the kluster.ai Batch API, demonstrated with examples from the AG News dataset. | |
| Multiple inference requests | Simple guide to process large text datasets, including summarization, translation, classification, and keyword extraction, with minimal setup. | |
| Evaluating LLMs with labeled data | Hands-on tutorial on evaluating Language Models (LLMs) with the kluster.ai Batch API, featuring a comparison of state-of-the-art Llama models for a text classification task using annotated data. | |
| LLM as a judge | Explore how to use a Language Model as a judge to validate predictions made by another model, demonstrated with the IMDb Top 1000 dataset for movie genre classification. | |
