Test it on the Peltarion Platform
A platform to build and deploy deep learning projects.
Even if you’re not an AI superstar.
An experiment can have many different states. The state is displayed just below the experiment’s name in the Experiments navigation section.
The experiment has been created. You can now build your model in the Modeling canvas.
When you’ve built your model, click Run. Then navigate to the Evaluation view the see how your experiment performs.
An experiment can be queued while waiting for it to start to run. This happens if you try to run more experiments at the same time than your number of allowed concurrent experiments. This number is shown in the Quota view.
The experiment is now locked and you have to duplicate the experiment to make any changes.
Click Dequeue to park your experiment.
The Dequeue button is disabled if you don’t own the experiment.
The experiment is running and is on it’s way to completion.
Click Pause to pause the experiment. This is a smart thing to do if you see that your experiment isn’t performing as it should. Then build a new experiment, try some new ideas.
When you have clicked Pause, the experiment will have Pausing state until the current epoch has finished.
The experiment is paused.
Click Resume to run the experiment from where it was paused.
The Resume button is disabled if you don’t own the experiment or if you try to run more experiments at the same time than your number of allowed concurrent experiments
Success! The experiment is done. Evaluate the experiment in the Evaluation view. + Duplicate the experiment if you are satisfied with the result. Otherwise, create a new experiment.
The experiment has failed. Check the error code in the Information pop-up and check for cause and remedy in our Error message section.
The experiment is locked and removed from the experiment run queue. If you want to make any changes, you need to duplicate the experiment.
Click Resume to put it back into the queue again.
The Resume button is disabled if you don’t own the experiment or if you try to run more experiments at the same time than your number of allowed concurrent experiments.