Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/mikolasan/ai_sandbox
Artificial Intelligence sandbox
https://github.com/mikolasan/ai_sandbox
artificial-intelligence artificial-neural-networks reinforcement-learning
Last synced: about 2 months ago
JSON representation
Artificial Intelligence sandbox
- Host: GitHub
- URL: https://github.com/mikolasan/ai_sandbox
- Owner: mikolasan
- License: mit
- Created: 2022-10-28T03:28:52.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2024-10-30T05:01:11.000Z (2 months ago)
- Last Synced: 2024-10-30T08:29:59.634Z (2 months ago)
- Topics: artificial-intelligence, artificial-neural-networks, reinforcement-learning
- Language: Jupyter Notebook
- Homepage:
- Size: 4.17 MB
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Artificial neural network sandbox
TODO: add description here. PRs are welcome ;)
## Install & Run
Pages from the `gui` folder are served by **vite**. Install modules
```
yarn install
```
Run in develpment mode
```
yarn dev
```
And open in the browser http://localhost:5173/
Other available pages:
- http://localhost:5173/emotions.html - dynamic model of emotions (only one bar)
- http://localhost:5173/index.html - artificial network visualization (WIP)
- http://localhost:5173/inverse-dynamics.html - 3D environment with a rolling wheel (solving with Euler and Newton equations)
- http://localhost:5173/qu-robot.html - visualization for **Qu Robot** project
- http://localhost:5173/robot-go.html - robot on a grid following the aim by listening ANN from the backend
### Backend for robot-go
Install dependencies
```
pip3 install --user -r requirements.txt
```
Run
```
python run-backend.py
```
## Links and notes
### PIXI js
- Guide https://pixijs.io/guides/basics/scene-graph.html
- API https://pixijs.download/release/docs/PIXI.Text.html
### A-star
- https://theory.stanford.edu/~amitp/GameProgramming/ImplementationNotes.html
- https://will.thimbleby.net/a-shortest-path-in-javascript/
### Polar plot
- https://matplotlib.org/stable/gallery/pie_and_polar_charts/polar_demo.html
- point colored by another vector https://stackoverflow.com/questions/14885895/color-by-column-values-in-matplotlib
### Pytorch
- mse loss https://pytorch.org/docs/stable/generated/torch.nn.MSELoss.html?highlight=mseloss
### Non-binary classifaction
- good article https://towardsdatascience.com/applied-deep-learning-part-1-artificial-neural-networks-d7834f67a4f6#106c
- notebook https://github.com/ardendertat/Applied-Deep-Learning-with-Keras/blob/master/notebooks/Part%201%20-%20Artificial%20Neural%20Networks.ipynb
n classes > 2 --> categorization / one-hot encoding --> softmax activation function