Skip to content

Installation of the tinyBIG Library

Prerequisites

Python

It is recommended that you use Python 3.10+. You can download and install the latest Python from the python official website.

Package Manager

To install the tinybig binaries, you will need to use pip.

pip

If you installed Python via Homebrew or the Python website, pip (or pip3) was installed with it.

To install pip, you can refer to the pip official website.

To upgrade your pip, you can use the following command: 

python -m pip install --upgrade pip

Installation

The tinybig library has been published at both PyPI and the project github repository.

Install from PyPI

To install tinybig from PyPI, use the following command:

pip install tinybig

Install from Source Code

You can also install tinybig from the source code, which has been released at the project github repository.

You can download the public repository either from the project github webpage or via the following command: 

git clone https://github.com/jwzhanggy/tinyBIG.git

After entering the downloaded source code directory, tinybig can be installed with the following command:

python setup.py install

If you don't have setuptools installed locally, please consider to first install setuptools: 

pip install setuptools

Dependency Packages

The tinybig library is developed based on several dependency packages. The updated dependency requirement.txt of tinybig can be downloaded from the project github repository.

After downloading the requirement.txt, you can install all these dependencies with the pip command:

pip install -r requirements.txt

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
torch==2.2.2
torchdata==0.7.1
torchvision==0.17.2
torchtext==0.17.2
portalocker==2.10.0

numpy==1.26.3
scipy==1.13.1
scikit-learn==1.5.1

pyyaml==6.0.1
tqdm==4.66.4
matplotlib==3.9.1
pytest==8.3.3

Verification

If you have successfully installed both tinybig and the dependency packages, now you can use tinybig in your projects.

To ensure that tinybig was installed correctly, we can verify the installation by running the sample python code as follows:

>>> import torch
>>> import tinybig as tb
>>> expansion_func = tb.expansion.taylor_expansion()
>>> expansion_func(torch.Tensor([[1, 2]]))
The output should be something like: 
tensor([[1., 2., 1., 2., 2., 4.]])