ML Training Dashboard
Train, evaluate, and deploy the AI opponent policy network
↻
Refresh
📊
Training Data
--
decisions available
🗃
Dataset
--
not built
🧠
Policy Model
--
no checkpoint
⚡
Inference Server
--
not loaded
Training Pipeline
Epochs
Learning Rate
Batch Size
Early Stop Patience
Rebuild dataset from decision logs
▶
Start Training
↻
Reload Model
BUILDING
Initializing...
Training Data Files
Loading data status...
Evaluation Results
No evaluation results yet. Train a model first.
PPO Reinforcement Learning
Iterations
Episodes/Iter
Learning Rate
Opponent
Heuristic
Random
Self-Play
Initialize from supervised checkpoint (warm start)
▶
Start PPO Training
TRAINING
Initializing...
Tournament Evaluation
Games per Matchup
Playstyle
Midrange
Aggro
Control
Combo
🏆
Run Tournament
RUNNING
Running matches...
Model Checkpoints
No checkpoints found.