https://github.com/mohamedelashri/python_bench_ising
Ising model simulation code as a python benchmark test
https://github.com/mohamedelashri/python_bench_ising
Last synced: about 2 months ago
JSON representation
Ising model simulation code as a python benchmark test
- Host: GitHub
- URL: https://github.com/mohamedelashri/python_bench_ising
- Owner: MohamedElashri
- License: mit
- Created: 2022-10-27T06:29:58.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-07-23T01:57:25.000Z (about 2 years ago)
- Last Synced: 2025-03-03T08:26:44.361Z (7 months ago)
- Language: Python
- Homepage: https://melashri.net/python_bench_ising/
- Size: 968 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Ising Model Python Benchmark
Ising model simulation as python Benchmarking test
## About
This repository contains code to perform simulations of the 2-d Ising model, a mathematical model in statistical mechanics. The model consists of discrete variables that represent magnetic dipole moments of atomic spins that can be in one of two states (+1 or -1). The simulations aim to study phase transitions, specifically the phenomenon of spontaneous magnetization. As new python releases are released I will test them vs the previous releases.
## packages
Previously I used `numpy` but this wasn't ideal situation because I introduced a dependency. So I decided to use the standard library `random` and `math` to implement the simulation.
The old `ising.py` file containts this code. But the standatd implementation is in `ising_model.py` file.
## Results
### Manual executions (Apple Silicon Macs - Standard)
The benchmarks are executed on Macbook pro Apple Silicon M1 version. The python version are installed using `homebrew`
The arguments of the script are L(length of the lattice), n(number of Monte Carlo cycles).
Also there is a `run_bench.sh` script that runs the benchmark for all the python versions (3.8-3.11) for different `n` and `L` values.
The results are stored in `results` folder. You can reproduce the results by running the `run_bench.sh` script.
These are the plots that are generated from the results (using `plot_results.py` script). I use the `png` as Github doesn't support embedded `pdf` files in the `README.md`.
### Manual executions (Apple Silicon Macs - Numpy)
The benchmarks are also executed on the same Macbook pro Apple Silicon M1 version. The python version are installed using `homebrew`
The arguments of the script are L(length of the lattice), n(number of Monte Carlo cycles).
#### python 3.11
```bash
time python3.11 ising.py 10 10000
Time it took in seconds is = 109.2954261302948
python3.11 ising.py 10 10000 110.82s user 0.79s system 100% cpu 1:51.38 total
```
#### python 3.10
```bash
time python3.10 ising.py 10 10000
Time it took in seconds is = 123.83638191223145
python3.10 ising.py 10 10000 124.80s user 0.69s system 101% cpu 2:04.06 total
```
#### python 3.9
```bash
time python3.9 ising.py 10 10000
Time it took in seconds is = 123.56476402282715
python3.9 ising.py 10 10000 123.94s user 1.08s system 101% cpu 2:03.67 total
```
#### python 3.8
```bash
time python3.8 ising.py 10 10000
Time it took in seconds is = 135.1741383075714
python3.8 ising.py 10 10000 137.05s user 1.00s system 100% cpu 2:17.24 total
```
#### Other versions
Python started supporting apple silicon starting from `python3.8` and I try to make things as consistent as possible. Not to mention that `numpy` optimization for different version might be unaccounted difference.
### Continuous integration
This repository is using github actions to run the simulation on `ubuntu` instance so that this is not limited to have `3.8` as a minimum version but I keep limited test versions before that to avoid having a long CI run. The results are published on a webpage [here](https://melashri.net/python_bench_ising/).
## Contributions
I welcome contributions to improve the code and the efficiency of the simulation. If you would like to contribute, feel free to fork this repository, make your changes, and submit a pull request. Please make sure that your code is clean, well-commented, and does not break the existing functionalities or introduce dependencies.