To help you learn Trains and some of its most useful features, we provide tutorials with step-by-step instructions for the following:
- Explicit Reporting - Go beyond our automagical features by extending Trains using the classes and methods in the Trains Python Client Package. This tutorial teaches you how to add explicit reporting (scalar metrics, plotting any other data, and logging messages, errors, warnings, and debugging), artifacts, and model snapshots to Python experiment scripts.
- Tuning Experiments - Use Trains and the Trains Web-App to tune your experiment and perfect your deep learning solution. This tutorial teaches you how to run an experiment, tune a copy of it, and compare the results of the original and tuned copy.
- Tracking Leaderboards - Tracking leaderboards let you monitor your experiments. This tutorial teaches you how to quickly setup a tracking leaderboard using the Trains Web-App.
- AutoML - Trains supports the easy implementation of autoML (for example, hyperparameter searching). This tutorial teaches you how to use several methods which are implemented in the Trains Python Client Package, Task class for autoML to perform a hyperparameter search on an experiment.