Trains is now ClearML
This documentation applies to the legacy Trains versions. For the latest documentation, see ClearML.
Integrate Trains into code which uses autokeras. Initialize a Trains Task in your code, and Trains automatically logs scalars, plots, and images reported to TensorBoard, Matplotlib, Plotly, and Seaborn, as well as all other automatic logging, and explicit reporting you add to your code (see Logging).
Trains allows you to:
- Visualize experiment results in the Trains Web-App (UI).
- Track and upload models.
- Track model performance and create tracking leaderboards.
- Rerun experiments, reproduce experiments on any target machine, and tune experiments.
- Compare experiments.
See the AutoKeras example, which shows Trains automatically logging of scalars, hyperparameters, the log, and models, including visualizations in the Trains Web (UI).
To install Trains:
pip install trains
By default, Trains works with our demo Trains Server (https://demoapp.trains.allegro.ai/dashboard). You can deploy a self-hosted Trains Server, see the Deploying Trains Overview, and configure Trains to meet your requirements, see the Trains Configuration Reference page.
Adding Trains to code
Add two lines of code:
from trains import Task task = Task.init(project_name="myProject", task_name="myExperiment")
When the code runs, it initializes a Task in Trains Server. A hyperlink to the experiment's log is output to console.
TRAINS Task: created new task id=c1f1dc6cf2ee4ec88cd1f6184344ca4e TRAINS results page: https://app.trains-master.hosted.allegro.ai/projects/1c7a45633c554b8294fa6dcc3b1f2d4d/experiments/c1f1dc6cf2ee4ec88cd1f6184344ca4e/output/log
Later in the code, for example, define callbacks using TensorBoard, and Trains logs TensorBoard scalars, histograms, and images.