SmallTrain

Three advantages of SmallTrain

１．Free and easy creation of commercially available AI models equipped with cutting-edge AI algorithms

２．Use as a trained model that supports small data can significantly reduce man-hours

３．As a wrapper * for both TensorFlow and PyTorch, a library function to provide unique algorithms will be installed

• What is …. wrapping a library such as TensorFlow or PyTorch, and users can access the library via the wrapper so that even if the library is replaced or the interface of the library is changed, the changes are stopped only inside the wrapper and the effect of the change are stopped.

Deep Learning Framework SmallTrain

High-precision, high-performance SmallTrain that can be used as a wrapper for TensorFlow and PyTorch, and as a library that provides original algorithms You can build a variety of AI models using the versatile pre-trained model by SmallTrain.

Explanation of the figure (the numbers in the figure match the following)

１. A deep learning library and wrapper

There are three major ways to create an AI model：

• Understand algorithms and mathematical functions, make them yourself and program them to create an AI model yourself
• Create your own AI model using algorithms and mathematical functions of calculation libraries such as TensorFlow
• Easily create an AI model using a wrapper to call a calculation library (reduce development time)

SmallTrain aims to be a wrapper equivalent to Keras.

Differences from Keras …
Keras is suitable for PoC, SmallTrain is for both PoC and commercial use.

Similarities to Keras …
It is the same where the usage is easy even for beginners

２. Wrapping TensorFlow mathematical functions

３. Available to call not only TensorFlow functions but also PyTorch functions

４. Available to use SmallTrain’s own calculation library

SmallTrain is adopted the algorithms of the state of the art paper as early as possible and also owned mathematical functions of SmallTrain.

５. AI model of Small Train

Using TensorFlow, PyTorch, and original mathematical functions, AI model of SmallTrain has a neural network* with more than 60 layers.
*It is a deep neural network that uses a pyramid network and can produce highly accurate results.

Also SmallTrain is adopted various methods such as CNN and others.

６. Trained model

SmallTrain has been learned.
It is versatile and supports various data, and has been learned with various data such as image data and time series data.

７. Providing a trained model for users

You can easily build your own trained model by inputting your own data into SmallTrain.
The Getting Started describes the image recognition method as an introduction. It supports almost all data.

Uses of SmallTrain

SmallTrain provides a versatile trained model for a wide variety of data.
SmallTrain has been used for image recognition, time series data prediction, and other development.