GSoC 2021: Polaris Web Reports

A tool to generate customizable and rich analysis Reports and make interactive graphs


  1. Polaris Behave The first goal was to make a module which uses a pre-built autoencoder model fromBETSI libraryand detect anomalies in satellite telemetry data and output the results, models, history and normalizers in nicely formatted form.
  2. Polaris Report The second goal was to use the results of Polaris behave and show them in the interactive formats in a web browser so that users can read them more easily.



  1. I learnt the importance of unit and integration tests. I made test cases for several classes and modules in Polaris.
  2. I tried out a lot of charting and graph libraries to select the perfect library that was able to handle big data, aesthetically beautiful and future proof.
  3. I learnt a whole lot about machine learning and auto-encoders. I read a bunch of research papers about anomaly detection, feature importance detection and other stuff.
  4. I got to know about how important the description of the merge request is when you are working with the collaboration of others. I started to write clearer descriptions, workflow of the merge requests, proof of validation and results in the description for better understanding.

Future Work

  • Make Supports for Multi-Satellite by URL separation.
  • Improve look and feel and responsiveness of the reports.
  • Integrate Polaris dependency graph in it to unify the visualization process.
  • Use Graph Comparison to get information of telemetry responsible for an anomaly, percentage of responsibility of telemetry, order of importance of telemetry in an anomaly and other micro details.
  • Add Support for both normalized graph and un-normalized graph.

More than Just an Internship



Web Developer

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store