The enterprise platform for
modern deep learning

Explore all features
Neural network designer
Training scheduler and monitoring
Real-time collaboration
Automated hyperparameter optimization

Deep neural network

Model designer

You can quickly create a deep neural network model using our model designer. For example a regression or image classification, without deep understanding of neural networks.

You can train the model completely on your own server infrastructure.
Python code based on the Keras framework is automatically generated and versioned.


Datasets in AETROS Trainer allows you to quickly create small datasets and link your created model with your own data.

You data stays in your data center and never reaches AETROS.
Use built-in feature image augmentation to prevent overfitting of your model.


With activated insights you get automatically a snapshot of all layers of your model per epoch and other insights like a confusion matrix.

Simple deployment

You can access your trained model via our API. The only thing you need to do is to train your model and tag your job with a version number.

You can also download the state of your trained model and run it in your own infrastructure.

marc@osx ~ $ curl -F "input=@/path_to/local/image.jpg"
{ "prediction": [ { "class": 8, "prediction": 0.13521507382393 }, { "class": 9, "prediction": 0.1149984151125 }, { "class": 4, "prediction": 0.10820455849171 }, ], "times": { "prediction": 0.1549129486084, "prepare_fetch_input": 0.1016149520874 } }

your model

Supervise key performance indicators
and custom metrics in real time of your model
created in the model designer or your
custom python model (framework independent).

Interact live with
your python model

Register an action in your script and
trigger it while it trains within our interface

Team collaboration
in real time

Share models or training runs with your
colleagues and watch in real-time.

Custom metrics

Send logs and metrics generated by your model
to AETROS Trainer using our Python SDK to
see in real-time your performance indicators.
More information in our docs Python SDK Channels.

job = aetros.backend.start_job('model/name')
accuracy_channel = job.create_channel('accuracy')
for i in range(0, 100):
accuracy_channel.send(x=i, y=...)
job.progress(i, 100)


Supports any model written in python, no matter which framework.

Rich web

Monitor, manage and start training jobs everywhere using only your browser.


Connect Git to start jobs on several servers.

Manage models

Link your python model with AETROS or create own deep neural network models using our model designer. With our Git integration you next traning job scheduler for your own python models is only a few lines of code using Python SDK away.

Manage jobs and experiments

Compare current running training jobs or history to see which
model and hyper-parameters perform best for your problem. With custom channels you can add as many metrics as you need.

Own server

With our aetros-cli you can easily connect your server to AETROS Trainer allowing you to keep all your data in your own server infrastructure.
AETROS has absolutely no access to your training data (images, csv files, etc). More information in our docs External server.

user@machine ~ $ aetros server --secure-key=my-secure-key
Connected to as marcj/alpha

Automatic hyperparameter optimization

With our fully automated and scaleable hyperparameter optimization based on Hyperop you find easier and faster better hyperparameters for your model. More information in our documentation.

Are you ready?

Improve your current machine learning workflow
or start digging into machine learning using our neural network designer.

Register now