"""Centralised configuration for ai_mouse.

All magic numbers and hyperparameters live here so they can be tuned
from one place, overridden per-instance, or serialised to JSON.
"""
from __future__ import annotations

from dataclasses import dataclass, field
from pathlib import Path


# ---------------------------------------------------------------------------
# Training configuration
# ---------------------------------------------------------------------------


@dataclass
class TrainConfig:
    """Hyperparameters for Flow Matching training."""

    epochs: int = 300
    batch_size: int = 64
    lr: float = 3e-4
    seq_len: int = 64
    # Transformer backbone
    d_model: int = 128
    nhead: int = 4
    num_layers: int = 4
    dim_feedforward: int = 256
    dropout: float = 0.1
    # Flow matching
    sigma_min: float = 1e-4
    # Data augmentation
    augment: bool = True
    # Duration distribution bins
    dist_bins: list[float] = field(
        default_factory=lambda: [0, 50, 100, 200, 400, 600, 800, 1200, float("inf")]
    )


# ---------------------------------------------------------------------------
# Generation configuration
# ---------------------------------------------------------------------------


@dataclass
class GenerateConfig:
    """Tuneable knobs for Flow Matching inference."""

    n_steps: int = 10           # Euler ODE steps
    seq_len: int = 64
    dt_clip_min_ms: float = 2.0
    dt_clip_max_ms: float = 150.0


# ---------------------------------------------------------------------------
# Scroll subsystem configuration
# ---------------------------------------------------------------------------


@dataclass
class ScrollModeConfig:
    """Parameters for a single scroll collection mode."""

    dist_min: int
    dist_max: int
    success_radius: int


SCROLL_MODES: dict[str, ScrollModeConfig] = {
    "target": ScrollModeConfig(dist_min=500, dist_max=3000, success_radius=80),
    "fast": ScrollModeConfig(dist_min=3000, dist_max=8000, success_radius=120),
    "precise": ScrollModeConfig(dist_min=200, dist_max=800, success_radius=40),
}


@dataclass
class ScrollTrainConfig:
    """Hyperparameters for ScrollCVAE training."""

    epochs: int = 100
    batch_size: int = 32
    lr: float = 5e-4
    seq_len: int = 32
    beta_max: float = 0.3
    beta_warmup_epochs: int = 15
    weight_delta: float = 1.0
    weight_log_dt: float = 1.5


# ---------------------------------------------------------------------------
# Server configuration
# ---------------------------------------------------------------------------


@dataclass
class ServerConfig:
    """Web server and data directory settings."""

    host: str = "127.0.0.1"
    port: int = 8765
    data_dir: Path = field(default_factory=lambda: Path("data"))
    canvas_width: int = 800
    canvas_height: int = 600
    open_browser: bool = True


# ---------------------------------------------------------------------------
# Pretraining (Balabit) configuration
# ---------------------------------------------------------------------------


@dataclass
class PretrainConfig:
    """Hyperparameters for Balabit pretraining stage."""

    epochs: int = 200
    batch_size: int = 128
    lr: float = 3e-4
    seq_len: int = 64


@dataclass
class FinetuneConfig:
    """Hyperparameters for fine-tuning on user-collected data."""

    epochs: int = 50
    batch_size: int = 64
    lr: float = 1e-5  # 比预训练小一个数量级，防止灾难性遗忘
    seq_len: int = 64


# ---------------------------------------------------------------------------
# Balabit adapter configuration
# ---------------------------------------------------------------------------


@dataclass
class BalabitAdapterConfig:
    """Settings for Balabit CSV → traces.jsonl conversion."""

    window_ms: int = 1200       # click 前回溯窗口
    min_dist: int = 50          # 最小起终点距离 (px)
    min_events: int = 5         # 最小 Move 事件数
    max_span_ms: int = 5000     # 最大段时间跨度 (ms)
    max_gap_ms: int = 200       # 段内相邻 Move 最大时间差


# ---------------------------------------------------------------------------
# Eval configuration
# ---------------------------------------------------------------------------


@dataclass
class EvalConfig:
    """Settings for evaluation report generation."""

    n_samples: int = 1000
    fft_freq_band: tuple[float, float] = (4.0, 12.0)  # 生理震颤频段 (Hz)
    kl_bins: int = 50