Target group
- Data Scientists in research and industry
Your requirements
You have data and have identified an appropriate prediction task for AI, then AutoML can help them either improve existing models or propose initial models.Our offer
The use of AI can present even experienced developers with a number of difficult design decisions, e.g., how to choose the right model class, neural network architecture, data preprocessing, or hyperparameters. In addition to the data provided, the success of AI applications depends heavily on such design decisions.
To make the development of new AI applications more efficient, we support Data Scientists (and similar professionals) with approaches, to provide suggestions for these design decisions. This dramatically speeds up development time and developers can focus more on what is important, instead of many failures due to suboptimal design decisions.
We offer a number of different AutoML tools optimized for different tasks. These include Auto-Sklearn for classical ML, Auto-PyTorch for Deep Learning, Auto-Pytorch for time series and SMAC for Hyperparameter Optimization.
Requirements
Basic knowledge in statistics and programming
Check out the documenation for our largest AutoML-package SMAC3: https://automl.github.io/SMAC3/main/
