Update Demo Preview video
Browse files
README.md
CHANGED
|
@@ -43,6 +43,8 @@ The model outputs two vectors:
|
|
| 43 |
1. **Policy**: A probability distribution over `NUM_POSSIBLE_MOVES=4672` representing the probability of making each move, obtained using `softmax` activation.
|
| 44 |
2. **Value**: A single scalar value indicating win/loss probability from current player’s perspective, ranging from -1 (loss) to 1 (win), obtained using `tanh` activation.
|
| 45 |
|
|
|
|
|
|
|
| 46 |
### Model Architecture
|
| 47 |
|
| 48 |
The neural network architecture consists of:
|
|
|
|
| 43 |
1. **Policy**: A probability distribution over `NUM_POSSIBLE_MOVES=4672` representing the probability of making each move, obtained using `softmax` activation.
|
| 44 |
2. **Value**: A single scalar value indicating win/loss probability from current player’s perspective, ranging from -1 (loss) to 1 (win), obtained using `tanh` activation.
|
| 45 |
|
| 46 |
+
[](https://huggingface.co/nirajandhakal/StockZero/blob/main/v2-gameplay-svg-high-quality.mp4)
|
| 47 |
+
|
| 48 |
### Model Architecture
|
| 49 |
|
| 50 |
The neural network architecture consists of:
|