behavython 0.8.0

Automated behavioral analysis software and pipeline for neuroscience.

To be used in conjunction with Deeplabcut to extract behavior from video
· Report Bug · Request Feature

Table of Contents

1. Getting Started
   • Prerequisites
   • Installation
2. Usage
3. Contributing
4. License
5. Contact
6. Acknowledgments

About

• This software was developed to be used in conjunction with Deeplabcut to extract behavior from video. It is a interface that allows the user to select the data files that were generated by DLC and then run the analysis, allowwing users to run the analysis using pretrained models on their data. The results are saved in a csv file that can be used for further analysis.

Built With

• Python
• Qt Designer

Getting Started

Prerequisites

Behavython relies on a scientific Python stack and machine learning frameworks. All required dependencies are installed automatically via pip, but note the following:

• Python 3.10.x is required
• GPU support depends on your local CUDA setup
• Installation may be heavy due to ML dependencies (TensorFlow, PyTorch, DeepLabCut)

---

Installation (pip)

Due to PyTorch CUDA wheels, you must include the PyTorch index during installation:

pip install behavython --extra-index-url https://download.pytorch.org/whl/cu118

If this flag is omitted, PyTorch may install incorrectly (e.g., CPU-only or incompatible builds).

---

Alternative: Windows automated setup

You can also use the provided installer:

run_behavython.bat

This script sets up the environment and installs all dependencies automatically.

---

Notes

• Behavython has been primarily tested on Windows
• Installation time may be significant depending on your system
• For reproducibility, dependency versions are strictly pinned in the package

---

Usage

• Windows

1. Open the interface typing "behavython" on the command line
   • If you installed it as a pip package you can just type "behavython" on the command line
   • If you downloaded the source code you need to go to the folder where you downloaded it and type "python Bbehavython_front.py" on the command line

2. Select all the photo-data pairs that you want to analyze
   • In this step is important that you don't forget to verify that you got all the bonsai files, including the data and a image of the arena that you are analyzing

3. Wait for the program to finish the analysis
   • Currently the program looks like it freezed when running. It is expected behavior but we are looking into it. Right now you only need to wait a little bit.

4. When finished the progress bar will show 100% and a preview of the results will be available on the right

See the open issues for a full list of proposed features (and known issues).

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated. If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue explaining what is the problem. Also, you can reach us by mail - listed at the end :)

License

Distributed under the GNU General Public License v3.0. See LICENSE.txt for more information.

Contact

João Pedro Carvalho Moreira - [e-mail]
Matheus Costa - [e-mail]

Acknowledgments

• Flávio Mourão:
  Github: Flávio Mourão
• Lab:
  Núcleo de Neurociências

Developed at

Nucleo de Neurociencias - NNC
Universidade Federal de Minas Gerais - UFMG
Brazil