About SmallTrain

  A wrapper able to run on both TensorFlow and PyTorch
A library for adopting the latest algorithms

SmallTrain is a wrapper able to use TensorFlow and Pytorch.
It’s designed as a hi-accuracy model.


SmallTrain is a deep learning library.
adopted the-state-of-the-art algorithms from the newest paper and our own algorithms


You can easily build your own trained model.
Additional learning with your small data can create your high-accuracy model.
Only a few lines of programming is required.

See the different sections of the documentation and site for more ideas.

How to Use SmallTrain




1. A library of deep learning models, and a wrapper
There are three main ways to create AI models:
- Understand algorithms and mathematics and build your own model
- Algorithms and mathematical functions are self-made using calculation libraries such as TensorFlow
- Easily create models using wrappers to call calculation libraries (most man-hours can be reduced)

SmallTrain aims to be a wrapper equivalent to Keras.
- Difference from Keras … Keras is suitable for PoC, SmallTrain is for both PoC and commercial use.
- Similarities to Keras ... the same is easy to use for beginners.

2. Wrapping TensorFlow mathematical functions

3. You can call mathematical functions of both PyTorch and TensorFlow.

4. You can also call SmallTrain's own calculation library.
We read the state-of-the-art papers and implement the cutting edge of algorithms.

5. AI model of SmallTrain
Using TensorFlow, PyTorch, and unique mathematical functions, we have built a neural network* with more than 60 layers.
*It is a deep neural network that holds Pyramid Network and can produce highly accurate results. It incorporates CNN and other techniques.

6. Trained model
SmallTrain has been learned.
It is highly versatile and supports all kinds of data, and learns with various data such as image data and time series data.

7. Providing your own trained model
By inputting user's data, you can easily build your own trained model.
Getting Started describes how to recognize images as an introduction.
It supports various data.

Great features of SmallTrain

- Easy to use
- Almost no programming to building your own pre-trained model.
- Both of TensorFlow and Pytorch are available to use.
- Always adapting algorisms from the State-of-the-Art paper
- Small data and small traing of model are enogh to build
- License free for commercial use

SmallTrain Case Study

- Image recognition
- Time series data forcasting
- Object detection
- Anomaly detection
- Voice recognition