Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/tpvasconcelos/mdsea
Pure Python Framework for Computational Fluid Dynamics
https://github.com/tpvasconcelos/mdsea
blender cfd fluid-dynamics fluid-simulation python python3
Last synced: about 15 hours ago
JSON representation
Pure Python Framework for Computational Fluid Dynamics
- Host: GitHub
- URL: https://github.com/tpvasconcelos/mdsea
- Owner: tpvasconcelos
- License: mit
- Created: 2018-09-28T12:30:17.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2024-08-29T10:40:24.000Z (2 months ago)
- Last Synced: 2024-08-29T11:58:26.510Z (2 months ago)
- Topics: blender, cfd, fluid-dynamics, fluid-simulation, python, python3
- Language: Python
- Homepage:
- Size: 126 MB
- Stars: 5
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# mdsea
mdsea is a stand-alone Python molecular dynamics library equipped with a
flexible simulation engine and multiple analysis tools, including
integrated beautiful visualization in 1, 2, and 3 dimensions.
## Getting started
### Prerequisites
mdsea requires **Python >= 3.6**. The
**minimal** requirements for this project will be automatically
installed with `setup.py`.
### Installing
To install mdsea you need to get the source and execute the `setup.py`
script
```sh
git clone https://github.com/TPVasconcelos/mdsea.git
cd mdsea
python setup.py install
```
### Extra requirements
In order to install the complete set of requirements and be able to
make the most use of mdsea, run
```sh
pip install -r requirements.txt
```
To render 3D animation with [Blender](https://www.blender.org), you will
need to [download](https://www.blender.org/download/) and install it on
your machine first. If you want mdsea to render mp4 animation files
for you, you will also need to
[install ffmpeg](https://www.ffmpeg.org/download.html).
## Running the tests
You can find the test scripts inside the `/tests` directory.
You should start by running
```sh
cd tests
python initsetup.py
```
which will run a default simulation and store in in
`./simfiles/_mdsea_testsimulation/*`. You can then run tests for the
multiple visualization tools (`/tests/vis_*.py`) - See bellow how to run
the Blender tests! - The `/tests/cleanup.py` script will remove the
`/simfiles/_mdsea_testsimulation` directory.
### Running Blender tests
Blender uses its own python interpreter, so you will need to install
mdsea there as well. Find out where the python binary is stored and
execute the following command. e.g. on macOS:
```sh
cd mdsea
/path/to/blender.app/Contents/Resources/2.XX/python/bin/python setup.py install --force
```
To run the tests, execute
```sh
cd tests
/path/to/blender.app/Contents/MacOS/blender --python vis_blender.py
```
I'd also suggest that you add
```
alias blender=/path/to/blender.app/Contents/MacOS/blender
```
to `.bash_profile` and `.profile`.
```sh
echo "alias blender=/path/to/blender.app/Contents/MacOS/blender" >> ~/.profile
echo "alias blender=/path/to/blender.app/Contents/MacOS/blender" >> ~/.bash_profile
```
In
[Blender’s documentation](https://docs.blender.org/manual/en/dev/render/workflows/command_line.html#platforms)
they only state the alias in `~/.profile` but this is only read by
_login_ shells. On the other hand, `~/.bash_profile` is read by
_interactive_ shells (like the one linked to Terminal).
## Example
```python
from mdsea.core import SysManager
from mdsea.helpers import setup_logging
from mdsea.potentials import Potential
from mdsea.simulator import ContinuousPotentialSolver
setup_logging(level="DEBUG")
# Instantiate system manager ---
sm = SysManager.new(
simid="_mdsea_docs_example",
ndim=2,
num_particles=6 ** 2,
steps=500,
vol_fraction=0.2,
radius_particle=0.5,
)
# Instantiate simulation ---
sim = ContinuousPotentialSolver(sm, pot=Potential.lennardjones(1, 1))
# Run the simulation ---
sim.run_simulation()
# Run 2D animation with matplotlib ---
from mdsea.vis.matplotlib import MLPAnimationScatter
anim = MLPAnimationScatter(sm, frame_step=6, colorspeed=True)
anim.anim()
```
The output should look something like this (you can ignore the warning):
```
[INFO:mdsea.core] New simulation: {'id': '_mdsea_docs_example', 'path': 'simfiles/_mdsea_docs_example'}
[WARNING:mdsea.core] A ValueError ('Not a location (invalid object ID)') was catched while trying to update a dataset. However, we noticed that the datafile in question wasn't open. We'll try to fix this and retry...
[INFO:mdsea.simulator] ProgresBar (Simulator): {'step': '28/500', 'percentage': 6, 'ETA': '1s'}
[INFO:mdsea.simulator] ProgresBar (Simulator): {'step': '73/500', 'percentage': 15, 'ETA': '1s'}
[INFO:mdsea.simulator] ProgresBar (Simulator): {'step': '128/500', 'percentage': 26, 'ETA': '1s'}
...
[INFO:mdsea.simulator] ProgresBar (Simulator): {'step': '428/500', 'percentage': 86, 'ETA': '0s'}
[INFO:mdsea.simulator] ProgresBar (Simulator): {'step': '473/500', 'percentage': 95, 'ETA': '0s'}
[INFO:mdsea.simulator] ProgresBar (Simulator): {'status': 'finished', 'lifetime': '1.1s'}
[INFO:mdsea.vis.mpl] Plotting anim_loop...
```
## Gallery
### 3D renderings
### GIFs
**1D Simulation:**
**2D Simulation:**
**3D Simulation with Blender's Cycles rendering engine:**
---
## Licence
[MIT License](./LICENSE)