https://github.com/atharvaaalok/geosimilarity
Differentiable curve and surface similarity measures.
https://github.com/atharvaaalok/geosimilarity
autograd curve-fitting curves differentiable-optimization geometric-deep-learning pytorch shape-matching similarity-measures surface-fitting surfaces
Last synced: 8 months ago
JSON representation
Differentiable curve and surface similarity measures.
- Host: GitHub
- URL: https://github.com/atharvaaalok/geosimilarity
- Owner: atharvaaalok
- Created: 2025-02-07T04:24:34.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-07T05:05:08.000Z (8 months ago)
- Last Synced: 2025-02-07T06:19:17.727Z (8 months ago)
- Topics: autograd, curve-fitting, curves, differentiable-optimization, geometric-deep-learning, pytorch, shape-matching, similarity-measures, surface-fitting, surfaces
- Homepage:
- Size: 0 Bytes
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
GeoSimilarity is a differentiable curve and surface similarity loss function library.
It is built on top of [PyTorch](https://github.com/pytorch/pytorch) and is designed with
modularity and extensibility in mind.
> [!NOTE]
> We are in the very early stages of development and are inviting domain experts for collaboration.
> The usage examples below are stand-ins for future api.
For a flavor of the documentation please refer:
[Documentation Webpage](https://atharvaaalok.github.io/geosimilarity/)
## Basic Usage
Given a candidate curve $Xc$ and a target curve $Xt$ we can measure the similarity using a given
measure as follows:
```python
import torch
import geosimilarity as gs
Xc = torch.randn(10, 2, requires_grad = True)
Xt = torch.randn(10, 2)
# Define a loss object and compute the similarity between curves
loss_fn = gs.MSELoss()
loss = loss_fn(Xc, Xt)
# To modify Xc to fit the target use autograd capacity for gradient stepping
loss.backward()
```
## Inviting Collaborators
We are in the very early stages of development are are inviting collaborators for:
- **Code Optimization:** Benchmark your results and show a clear improvement.
- **Novel Loss Functions:** Developing new curve/surface/point cloud similarity measures.
- **Testing:** Incorporating testing codes.
- **Improving Documentation:** Improving doc-string clarity and including doc tests. Also for
improving the documentation website and adding explanations of similarity measures.
We'll use [Github issues](https://github.com/atharvaaalok/geosimilarity/issues) for tracking pull
requests and bugs.
## License
Distributed under the [MIT License](License).
## Project Plan:
- [x] Create first loss function. Mean Squared Error (MSE).
- [x] Generate proper documentation.
- [x] Proper docstrings. Follow Google python coding style guide.
- [x] Take inspiration from pytorch.
- [x] Use math equations.
- [x] Use type annotations.
- [x] Add .gitignore file.
- [x] Add testing code.
- [x] Use linear splines. Create spline generator function.
- [x] Create automated training function.
- [x] Create plot function. Parameterized and Target shape comparison.
- [x] Generate a bunch of target shapes. Use SVGs.
- [ ] Create documentation first cut using MyST markdown.
- [x] Create motivation/introduction.
- [ ] Include function docstrings in the documentation.
- [x] Launch web page for documentation using github pages.
- [x] Add license.