namndh / quantile_computation

Techcombank Interview Coding Test

Problem definition:

• In my opinion, the requirement of the test is to create an application with two APIs that:
  • Add or update a pool of values
  • Compute the quantile value of a pool with known p-th percentile

Installation:

• Requirement: docker, docker-compose
• To build a new container: docker-compose build
• To up the container: docker-compose up

Solution

• The algorithm I used in this project:
  • Let n be the length of the sorted_pool and 0 < p <= 100 be the desired percentile.
  • If n = 1 return the unique array element (regardless of the value of p); otherwise
  • Compute the estimated percentile position in the array pos = (p * (n + 1) / 100) and the difference, d between pos and floor(pos).
  • If pos < 1 return the smallest element in the array.
  • Else if pos >= n return the largest element in the array.
  • Else let lower = sorted_pool[floor(pos)-1] and let upper = sorted_pool[floor(pos)] (in Python, array start with 0). Return lower + d * (upper - lower)
• APIs:
  • http://127.0.0.1:5000/pool - it takes a json form that contains id of the pool and its value as input, it is used to update value of a pool or insert a new pool. A pool is defined as a dictionary with key is its id and value is list of its samples
  • http://127.0.0.1:5000/quantile - it takes a json form that contains id of the pool and the p-th percentile as input, it is used to compute p-th percentile value of a pool
• You can use: Postman, curl or the add-on Swagger in the url: http://127.0.0.1:5000 to test the API

Contributing

Pull request are welcome. Please open any issues to help me improve the project

Reference

• Apache Commons Math API docs
• Quantile: Definition and How to Find Them in Easy Steps