/ThirdAILabs-Demos

Notebooks for ThirdAI demos

Primary LanguageJupyter NotebookOtherNOASSERTION

Forks Stargazers Issues License


Logo

Demos

Interactive notebooks for exploring ThirdAI's library.

[Website] · [Docs] · [Report Bug] · [We're Hiring]


Table of Contents
  1. Welcome
  2. Quickstart
    • Downloading a License
    • Installation
  3. Usage
  4. License
  5. Contact

👋 Welcome

ThirdAI's library is a deep-learning framework that leverages sparsity to make training and inference computationally feasible on CPUs. Our demo repo will help get you familiar with our BOLT API through interactive notebooks.

(back to top)

🚀 Quickstart

Step 1: Downloading a License

To use our library, you will first need to register for a free license here. We will email you a unique download link for your license.

To download your license, run:

wget your_download_link ~/license.serialized

This will download the license file to your home directory. Please note: a valid, unexpired license must be placed in your home directory. If a valid license is not found, these demos will fail.

Step 2: Installation

You can download the ThirdAI library with your package manager of choice.

For example:

pip3 install thirdai

(back to top)

🎮 Usage

First, clone the demo repo:

git clone https://github.com/ThirdAILabs/Demos.git

Each of the demo notebooks will walk you through different aspects of the BOLT engine that you can run on any CPUs (ARM, AMD, Intel) and even desktops and laptops

  • TextClassifier.ipynb will show you how to get near SOTA accuracy on most text classification via a plug and play classifier at any given budget (everything autotuned).
  • TabularClassifier.ipynb will show you how to get near SOTA accuracy on most tabular dataset classification via a plug and play classifier at any given budget (everything autotuned).
  • DocSearch.ipynb will show you how to use your own dataset (or one of the provided datasets) to create a simple document + query embedding model.
  • SentimentAnalysis.ipynb will take you throught the process of creating a network to use during sparse training and sparse inference with the goal of predicting positive/negative sentiment.

(back to top)

📄 License

See LICENSE.txt for more information.

(back to top)

🎙 Contact

ThirdAILabs - @ThirdAILab - contact@thirdai.com

Project Link: https://github.com/ThirdAILabs/Demos

(back to top)