AI Rank
中文
← Open Source
glouw

tinn

A tiny neural network library.

Model DevelopmentDeep Learning FrameworksC
Open on GitHub
Momentum
+1stars in 24 hours+0.0%
2.14k
Stars
188
Forks
+3
This week
4
Contributors
Created 2018-03-27 · Updated 2026-10-16 · #5167 today
Top developers
glouwnicolas-sauzedesaschagrunert

glouw/tinn is ranked 5,167 with 2,139 stars.

Same-category comparison

glouw/tinn is ranked 5,167 among these 8 peers; marlbenchmark/on-policy is ranked 5,170.

Same-category comparison
NameRankMetrics
oxwhirl/pymarl5,129
Language:
Python
Stars:
2,223
Forks:
414
Description:
Python Multi-Agent Reinforcement Learning framework.
RubixML/ML5,139
Language:
PHP
Stars:
2,211
Forks:
192
Description:
A high-level machine learning and deep learning library for the PHP language.
MoonshotAI/MoBA5,148
Language:
Python
Stars:
2,194
Forks:
161
Description:
MoBA: Mixture of Block Attention for Long-Context LLMs.
open-mmlab/mmrotate5,158
Language:
Python
Stars:
2,176
Forks:
644
Description:
OpenMMLab Rotated Object Detection Toolbox and Benchmark.
marlbenchmark/on-policy5,170
Language:
Python
Stars:
2,121
Forks:
387
Description:
This is the official implementation of Multi-Agent PPO (MAPPO).
eloialonso/diamond5,172
Language:
Python
Stars:
2,120
Forks:
168
Description:
DIAMOND (DIffusion As a Model Of eNvironment Dreams) is a reinforcement learning agent trained in a diffusion world model. NeurIPS 2024 Spotlight.
facebookresearch/ai4animationpy5,173
Language:
Python
Stars:
2,114
Forks:
257
Description:
A Python framework for AI-driven character animation using neural networks.
opendatalab/OmniDocBench5,174
Language:
Python
Stars:
2,111
Forks:
204
Description:
[CVPR 2025] A Comprehensive Benchmark for Document Parsing and Evaluation.
Key parameters and specifications
Key parameters and specifications
ParameterValue
LanguageC
CategoryModel Development
SubcategoryDeep Learning Frameworks
KeywordsEducation
Stars2,139
Forks188
Contributors4
Created date2018-03-27
Updated date2026-10-16
Stars in 1 day1
Stars in 7 days3
Rank5,167
Previous rank8,990
Rank change3,823
7-day stars rank6,778
Previous 7-day stars rank7,867
7-day stars rank change1,089
Total stars rank6,007
Previous total stars rank5,998
Total stars rank change-9
CountryCanada
README summary

Tinn (Tiny Neural Network) is a 200 line dependency free neural network library written in C99. For a demo on how to learn hand written digits, get some training data: The training data consists of hand written digits written both slowly and quickly. Each line in the data set corresponds to one handwritten digit. Each digit is 16x16 pixels in size giving 256 inputs to the neural network. At the end of the line 10 digits signify the hand written

digit: This gives 10 outputs to the neural network. The test program will output the accuracy for each digit. Expect above 99% accuracy for the correct digit, and less that 0. Features Portable - Runs where a C99 or C++98 compiler is present. Tips Tinn will never use more than the C standard library. Train a model on your powerful desktop and load it onto a microcontroller and use the analog to digital converter to predict real time events. The

Tinn source code will always be less than 200 lines. Functions externed in the Tinn header are protected with the _xt_ namespace standing for _externed tinn_. Tinn can easily be multi-threaded with a bit of ingenuity but the master branch will remain single threaded to aid development for embedded systems. Tinn does not seed the random number generator. Shuffle again after every training iteration. Get greater training accuracy by annealing your learning rate. For instance, multiply your learning rate

by 0. This will zero in on a good learning minima. Disclaimer Tinn is a practice in minimalism. Tinn is not a fully featured neural network C library like Kann, or Genann: Ports Other A Tutorial using Tinn NN and CTypes Tiny Neural Network Library in 200 Lines of Code

Frequently asked questions

How many stars and forks are recorded?

glouw/tinn has 2,139 stars and 188 forks.

What is the repository rank?

glouw/tinn is ranked 5,167.

What language and category are recorded?

glouw/tinn has 2,139 stars, uses C and belongs to Model Development.

Which repository is closest in rank?

glouw/tinn is compared with marlbenchmark/on-policy, ranked 5,170 with 2,121 stars.

README

Tinn (Tiny Neural Network) is a 200 line dependency free neural network library written in C99.

For a demo on how to learn hand written digits, get some training data:

wget http://archive.ics.uci.edu/ml/machine-learning-databases/semeion/semeion.data

make; ./test

The training data consists of hand written digits written both slowly and quickly. Each line in the data set corresponds to one handwritten digit. Each digit is 16x16 pixels in size giving 256 inputs to the neural network.

At the end of the line 10 digits signify the hand written digit:

0: 1 0 0 0 0 0 0 0 0 0
1: 0 1 0 0 0 0 0 0 0 0
2: 0 0 1 0 0 0 0 0 0 0
3: 0 0 0 1 0 0 0 0 0 0
4: 0 0 0 0 1 0 0 0 0 0
...
9: 0 0 0 0 0 0 0 0 0 1

This gives 10 outputs to the neural network. The test program will output the accuracy for each digit. Expect above 99% accuracy for the correct digit, and less that 0.1% accuracy for the other digits.

Features

  • Portable - Runs where a C99 or C++98 compiler is present.

  • Sigmoidal activation.

  • One hidden layer.

Tips

  • Tinn will never use more than the C standard library.

  • Tinn is great for embedded systems. Train a model on your powerful desktop and load it onto a microcontroller and use the analog to digital converter to predict real time events.

  • The Tinn source code will always be less than 200 lines. Functions externed in the Tinn header are protected with the xt namespace standing for externed tinn.

  • Tinn can easily be multi-threaded with a bit of ingenuity but the master branch will remain single threaded to aid development for embedded systems.

  • Tinn does not seed the random number generator. Do not forget to do so yourself.

  • Always shuffle your input data. Shuffle again after every training iteration.

  • Get greater training accuracy by annealing your learning rate. For instance, multiply your learning rate by 0.99 every training iteration. This will zero in on a good learning minima.

Disclaimer

Tinn is a practice in minimalism.

Tinn is not a fully featured neural network C library like Kann, or Genann:

https://github.com/attractivechaos/kann

https://github.com/codeplea/genann

Ports

Rust: https://github.com/dvdplm/rustinn

Other

A Tutorial using Tinn NN and CTypes

Tiny Neural Network Library in 200 Lines of Code

Nearby in the ranking
Similar items