Create completely new applications using deep learning, a sub field of artificial intelligence. With AETROS Trainer you can quickly create, train and deploy your own deep neural network to build completely new ideas or improve your existing business with a very new technologie.
Example use case
Computer vision is an everyday use case.
A human has to identify constantly situations by classifying optical data: such as red or green traffic light, street signs, unknown persons or friends, or differentiate between a harmless fish and an attacking shark. Several years ago, these tasks were neither completely affordable nor accurate. Today we have the chance to use neural networks to help humans classify images or videos with affordable servers and hardware.
Since these tasks are so basic for humans, it is almost impossible to think about all possible use cases. Here we have some examples for you, and we are curious what problem you want to solve with it:
Find similarities based on images not on content.
Train a network to find objects, patterns, products, or even animals/plants in images.
Detect in video or image sequences whether a behavior is given or not.
Our application AETROS TRAINER gives you a tool to quickly train and test your own digital image recognition expert for all kinds of use cases. You only need to upload your test data (images) and wait until the system finishes the training session. Once trained, you can verify the results and deploy the network in your own infrastructure.
Researchers and data scientists can use AETROS TRAINER to build
different neural networks using a graphical interface. Train on AETROS high-performance or your own server,
measure the performance of each network, and share with your colleagues the current training process on an easy-to-use web application.
The application generates basically valid Python code based on Keras.io, so you can fine-tune your network or get a better insight into how the network is built.