Taipy Demo: Job Monitoring

This demo shows how to do Taipy Job Monitoring from Taipy! Run jobs and see their progress using a simple Taipy table.

More information on Taipy here.

Demo Type

Level: Advanced Topic: Taipy-Core, Taipy-GUI

Components: TOML-based configuration, Job execution API

Demo Walkthough

When run, the demo shows a single page with a table displaying all the jobs run so far. If you started the application for the first time, no jobs would appear in the list:

You can click on the "Run Pipeline..." button to run a pipeline:

Select the train pipeline for example (the demo comes with a very simple LogisticRegression example), and press "run":

The table now shows two jobs:

• a "preprocess" task (it's JOB ID is JOB_preprocess_28d29a7e-aada-4f7c-82c2-65ae0e095f69 here-above, but may be different for you)
• a "train" task (it's JOB ID is JOB_train_9cbb7b4c-2a6d-43ba-9336-b4cd209d889e but can be different for you)

Those two jobs are fast and are already in status "COMPLETED".

You can open the details of the job by clicking on a row. This opens a side panel with details:

If a job has failed because of a code exception, then the side panel will display the related stack trace.

You can run two other pipelines: the "predict" pipeline which would use the output of the "train" pipeline to run a simple Logistic Regression model using Scikit-Learn, or a "long_running" pipeline which does nothing else than simulating a 20 seconds jobs. After 20 seconds, the job will automatically appears as "COMPLETED" in the user interface.

Directory Structure

• data/ contains a simple dataset for the logistic regresion example (which is out of scope of this demo).
• job_monitoring/ contains the main source code. This is where you can find the main main.py file to run the app.
  • job_monitoring/algo contains the code to be execute in Taipy pipelines: the file ml.py contains a simple logistic regression problem, and debug.py a simple code to make a long running task.
  • job_monitoring/pages contains the code for the pages of the app. The main page is monitoring.py. This code has been in such a way that you can just copy/paste it into your own app to get the job monitoring table.
• app.config.toml contains the configuration for this Taipy app. It basically defines the data nodes, tasks and pipelines configuration.
• CONTRIBUTING.md instructions to contribute to this demo
• INSTALLATION.md instructions to install this demo
• LICENSE the Apache 2.0 license.
• Pipfile: File used by the Pipenv virtual environment to manage project dependencies.
• README.md this file

Installation

Want to install Demo Job Monitoring? Check out our INSTALLATION.md file.

Contributing

Want to help build demo ? Check out our CONTRIBUTING.md file.

Code of conduct

Want to be part of the demo community? Check out our CODE_OF_CONDUCT.md file.