subhash-pal / monte-carlo Goto Github PK

Monte-Carlo exercise
====================

The goal is to implement Monte-Carlo numerical solver module in Python.

Day 1.
- Explain the concept and provide examples.
- Learn about the random module.
- Prototype with toy examples (compute circle area).
- Think about design considerations and tests.

Day 2-3.
- Implement initial version for solving the 2D problem.
- Run test suite via py.test.
- Collect and maximize code coverage.
- Validate pep8/pylint.
- Use logging with various verbosity levels.
- Use tox for automation (tests, coverage, lint).
- Generalize to 3D case.
- Optimize memory usage -> O(1)

Day 4.
- Add command-line tool for invoking the solver on specific problem.
	* use logging for initialization and result reporting
	* use argparse for `iters` specification
- Use `matplotlib` to visualize resulting sample points.
- Use `cProfile` to optimize the solver's performance.

Day 5.
- Use `multiprocessing` to spread the work on CPU cores.
	* use argparse for pool size specification.
	* experiment and report optimal pool size.
- Use `numpy` for vectorization of loops.

Extra.
- Import function from configuration Python file.
	* use "path.to.module:func" notation.
- Specify domain from command-line arguments.
 	* use your own notation.