/optical-levitation-raytracing-experiments

Primary LanguageJupyter Notebook

optical-levitation-raytracing-experiments

Created during an Undergraduate Research Experience Programme placement at Imperial College London 2020 by Jakub Dranczewski.

To contact me, try (in no particular order)

Setup

A virtual environment is recommended, but not strictly required. The non-standard packages used are listed in requirements.txt. Quite a few of these probably come with a standard Anaconda installation, but not necessarily all of them.

To install all the requirements automatically, just run:

pip install -r requirements.txt

Now, this command will slightly vary depending on your particular Python setup. This margin is to small to delve into the magical land of Python environments, but as a general rule, if you're in a virtual environment pip should be fine. If not, and you have more than one Python, you may need to use pip3. Or, if that doesn't work, maybe pythom -m pip will do the trick?

Structure of this repository

simulate.py

The most important file for most use cases. It allows for running configuration files and simulating the behaviour of a given target.

The behaviour of this script largely depends on the chosen configuration file, which is better described in configs\config_example.yaml. In general it will run a simulation and the display / save the results in the outputs directory.

The script runs the config.yaml file in its directory by default.

To run arbitrary files, you need to pass them as command line arguments. For example:

> python simulate.py configs/config_example.yaml

The output directory includes a README.md file explaining the output file format.

output-inspector.ipynb

A Jupyter Notebook that can be used to analyse output files produced by simulate.py. It includes a few convenient utilities, like orientation visualisation, graphing the ray tracer's output at any point in time, and quiver plotting.

It can also be used as a guide to how the data can be analysed.

ptracer.py

The ray tracer library. This can be used on its own to investigate particular situations, which is demonstrated in ptracer-examples.ipynb and, in more detail, in various-experiments\ptracer-experiments.ipynb.

jit_methods

This file includes some computationally heavy or often-called functions, which are then compiled by numba.

notes.pdf

A set of handwritten notes containing most of the derivations for maths used in the code.

Directories

  • configs contains example configuration files and is a good place to put your config files as well.
  • forces contains python files describing the various forces that can be used in simulate.py.
  • objs contains some .obj files, which describe 3D meshes. Note that some of them do not contain information about vertex normal and cannot be used in the SmoothMeshTO class.
  • output is where simulate.py puts output directories.
  • various-experiments contains code and Jupyter Notebooks used and created during the project to test various idea and implementations.

**Some of these directories contain README.md files with further information.

Some references