Add README.md
Browse files
README.md
CHANGED
|
@@ -37,41 +37,12 @@ For more information please refer to the paper at: https://doi.org/10.13140/RG.2
|
|
| 37 |
|
| 38 |
| File | Purpose |
|
| 39 |
|----------------------------|--------|
|
| 40 |
-
| `chess_mcvs.py`
|
| 41 |
|
| 42 |
-
|
| 43 |
-
## Requirements
|
| 44 |
## Requirements
|
| 45 |
|
| 46 |
Install the minimal dependencies required to run `chess_mcvs.py` and the handler:
|
| 47 |
|
| 48 |
-
```bash
|
| 49 |
-
pip install chess torch numpy huggingface_hub
|
| 50 |
-
```
|
| 51 |
-
|
| 52 |
-
## Quick Start
|
| 53 |
-
|
| 54 |
-
Run the main program (self-play + training):
|
| 55 |
-
|
| 56 |
-
```bash
|
| 57 |
-
python chess_mcvs.py
|
| 58 |
-
```
|
| 59 |
-
|
| 60 |
-
Run a tournament between MCVS and UCT:
|
| 61 |
-
|
| 62 |
-
```bash
|
| 63 |
-
python mcvs_vs_uct.py
|
| 64 |
-
```
|
| 65 |
-
|
| 66 |
-
Handler usage (for inference / HF-style endpoints):
|
| 67 |
-
|
| 68 |
-
```python
|
| 69 |
-
from handler import EndpointHandler
|
| 70 |
-
h = EndpointHandler(path='.')
|
| 71 |
-
res = h({"inputs": {"fen": "<FEN>", "player": "w", "time_budget": 1.0}})
|
| 72 |
-
print(res)
|
| 73 |
-
```
|
| 74 |
-
|
| 75 |
## Notes
|
| 76 |
|
| 77 |
The repository contains the following important file:
|
|
|
|
| 37 |
|
| 38 |
| File | Purpose |
|
| 39 |
|----------------------------|--------|
|
| 40 |
+
| `chess_mcvs.py` | Main implementation: game logic, ABC model, Zone Database, MCVS, neural networks, incremental training |
|
| 41 |
|
|
|
|
|
|
|
| 42 |
## Requirements
|
| 43 |
|
| 44 |
Install the minimal dependencies required to run `chess_mcvs.py` and the handler:
|
| 45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
## Notes
|
| 47 |
|
| 48 |
The repository contains the following important file:
|