Week 7-GSoC 2020

This blog describes the 7th week of the program and the 3rd week of Phase 2. Some of the key highlights on the discussions and the implementations during this week are:


  • Matrix Distributions
    • This issues contains the discussions for adding the matrix distributions which are not possible to implement along with the joint distributions.

Open PRs:

  • MatrixGamma Distribution
    • This is a prototype PR for showing the Matrix Distribution additions. However, this needs to be changed and will be implemented in a new file as suggested by Gagandeep Singh.
  • Add symbolic Moment and CentralMoment
    • This PR fixes one of the old issue for adding symbolic classes for Moment and CentralMoment. This PR also links moment with Moment.doit() and cmoment with CentralMoment.doit().

Merged PRs:

  • Compound Distributions
    • Adding support for Compound Distributions was one of the main aim of the project as well as of the Phase 2 as it was stalled from 2018. It now has its own Probability Space and is implemented in a new file which was previously mixed up with Joint Rvs.
  • Add doit in symbolic Probability
    • This PR adds doit method in symbolic Probability. This also links P or probability with Probability.doit(). Hence, this forms the uniformity with P linked to Probability.doit() and E linked to Expectation.doit().


Varanasi, Uttar Pradesh, India