Upload README.md with huggingface_hub

README.md

@@ -3,39 +3,25 @@ license: mit
 tags:
 - chess
 - reinforcement-learning
-- game-ai
+- pytorch
 ---
 
-# ChessHacks Chess Bot Model
+# ChessHacks Model
 
-This model was trained to play chess using a neural network with policy and value heads.
+Neural network trained to play chess using PyTorch.
 
 ## Model Architecture
-
 - Convolutional neural network with residual blocks
-- Policy head: Predicts move probabilities
-- Value head: Evaluates position strength
+- Policy head: predicts move probabilities
+- Value head: evaluates board positions
 
 ## Usage
-
 ```python
-import torch
 from huggingface_hub import hf_hub_download
 
-# Download the model
-model_path = hf_hub_download(
+# Download weights
+weights_path = hf_hub_download(
     repo_id="HamzaAmmar/chesshacks-model",
     filename="best.pt"
 )
-
-# Load checkpoint
-checkpoint = torch.load(model_path, map_location="cpu")
 ```
-
-## Training
-
-Trained on Lichess games using:
-- 10 epochs
-- 10,000 steps per epoch
-- Batch size: 256
-- Lichess standard chess games dataset