//! PPO Separate Learning Rates Demo
//!
//! Demonstrates how to use separate actor/critic learning rates in PPO trainer.
//!
//! # Usage
//! ```bash
//! cargo run -p ml --example ppo_separate_lr_demo --release
//! ```

use ml::trainers::ppo::PpoHyperparameters;

fn main() {
    println!("PPO Separate Learning Rates Demo\n");

    // Example 1: Using default separate learning rates
    println!("Example 1: Default Configuration");
    let default_params = PpoHyperparameters::conservative();
    println!("  Actor LR:  {:?}", default_params.actor_learning_rate);
    println!("  Critic LR: {:?}", default_params.critic_learning_rate);
    println!();

    // Example 2: Custom separate learning rates
    println!("Example 2: Custom Separate Learning Rates");
    let mut custom_params = PpoHyperparameters::conservative();
    custom_params.actor_learning_rate = Some(1e-6); // Conservative for policy stability
    custom_params.critic_learning_rate = Some(0.001); // Aggressive for faster value convergence
    println!("  Actor LR:  {:?}", custom_params.actor_learning_rate);
    println!("  Critic LR: {:?}", custom_params.critic_learning_rate);
    println!();

    // Example 3: For train_ppo_parquet integration
    println!("Example 3: Recommended Settings for Parquet Training");
    let mut parquet_params = PpoHyperparameters::conservative();
    parquet_params.actor_learning_rate = Some(1e-6); // Actor: 1e-6
    parquet_params.critic_learning_rate = Some(0.001); // Critic: 0.001 (1000x faster)
    parquet_params.epochs = 100;
    parquet_params.batch_size = 64;
    println!("  Actor LR:  {:?}", parquet_params.actor_learning_rate);
    println!("  Critic LR: {:?}", parquet_params.critic_learning_rate);
    println!("  Epochs:    {}", parquet_params.epochs);
    println!("  Batch Size: {}", parquet_params.batch_size);
    println!();

    println!("✅ Separate learning rates are now supported!");
    println!("   - Actor (policy) learns slowly for stability");
    println!("   - Critic (value) learns faster for better returns estimation");
}