Metadata-Version: 2.4
Name: flexibot
Version: 0.1.3
Summary: A reinforcement learning library for robotics.
Author: Luke William Warren
Keywords: reinforcement-learning,robotics,ai,machine-learning
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=2.0
Dynamic: license-file

# Flexibot

This is an efficient Python library made for the intended use of reinforcement learning in robotics.

!!WARNING THIS LIBRARY IS NOT FOR THE USE OF LLMS or Agents (e.g. ChatGPT, Gemini)!!

## Installation

```bash
pip install flexibot
```
## Example of how Flexibot is used

```python
import flexibot as fl
import numpy as np
import time as t

isizess = [(1, 10), (1, 5), (1, 4), (1, 8)]
wsizess = [(10, 5), (5, 4), (4, 8), (8, 10)]  # Inner numbers match input columns!
bsizess = [(1, 5),  (1, 4), (1, 8), (1, 10)]



model = fl.network(isizes=isizess, wsizes=wsizess, bsizes=bsizess)

model.ReLU(activate=True)

while True:
    targets = np.random.randint(1, 10, size=(1, 10))
    g = model.MatrixMultiplication(2)

    l, err = model.MSE(targets)

    fg = model.Smoothen(1.2)
    t.sleep(2)
    print(fg)
    
    model.Optim(0.15)
```

## License

MIT License
