feat(actpred): 动作预测模块入库（PyTorch 训练/预测/ONNX 导出）

actpred/config.py

@@ -0,0 +1,110 @@
+import argparse
+from dataclasses import dataclass, asdict, field
+from typing import Optional
+
+
+@dataclass
+class DataConfig:
+    image_size: int = 64
+    action_dim: int = 8
+    history_len: int = 5
+    data_dir: str = "./data"
+    normalize_actions: bool = False
+    action_stats_path: Optional[str] = None
+
+
+@dataclass
+class ModelConfig:
+    cnn_backbone: str = "simple"
+    cnn_feature_dim: int = 128
+    action_hidden_dim: int = 64
+    fused_hidden_dim: int = 128
+    dropout: float = 0.1
+    history_encoder: str = "mlp"
+
+
+@dataclass
+class TrainConfig:
+    batch_size: int = 32
+    epochs: int = 100
+    lr: float = 1e-3
+    weight_decay: float = 1e-4
+    val_split: float = 0.15
+    checkpoint_dir: str = "./checkpoints"
+    patience: int = 15
+    num_workers: int = 2
+
+
+@dataclass
+class Config:
+    data: DataConfig = field(default_factory=DataConfig)
+    model: ModelConfig = field(default_factory=ModelConfig)
+    train: TrainConfig = field(default_factory=TrainConfig)
+
+    def to_dict(self):
+        return asdict(self)
+
+    @classmethod
+    def from_dict(cls, d):
+        return cls(
+            data=DataConfig(**d.get("data", {})),
+            model=ModelConfig(**d.get("model", {})),
+            train=TrainConfig(**d.get("train", {})),
+        )
+
+
+def build_argparser(description: str = "") -> argparse.ArgumentParser:
+    parser = argparse.ArgumentParser(description=description)
+    parser.add_argument("--image-size", type=int, default=None)
+    parser.add_argument("--action-dim", type=int, default=None)
+    parser.add_argument("--history-len", type=int, default=None)
+    parser.add_argument("--data-dir", type=str, default=None)
+    parser.add_argument("--normalize-actions", action="store_true", default=None)
+    parser.add_argument("--action-stats-path", type=str, default=None)
+    parser.add_argument("--backbone", type=str, default=None, choices=["simple", "mobilenet"])
+    parser.add_argument("--cnn-feature-dim", type=int, default=None)
+    parser.add_argument("--action-hidden-dim", type=int, default=None)
+    parser.add_argument("--fused-hidden-dim", type=int, default=None)
+    parser.add_argument("--dropout", type=float, default=None)
+    parser.add_argument("--history-encoder", type=str, default=None, choices=["mlp", "gru"])
+    parser.add_argument("--batch-size", type=int, default=None)
+    parser.add_argument("--epochs", type=int, default=None)
+    parser.add_argument("--lr", type=float, default=None)
+    parser.add_argument("--weight-decay", type=float, default=None)
+    parser.add_argument("--val-split", type=float, default=None)
+    parser.add_argument("--checkpoint-dir", type=str, default=None)
+    parser.add_argument("--patience", type=int, default=None)
+    parser.add_argument("--num-workers", type=int, default=None)
+    parser.add_argument("--save-config", type=str, default=None, help="保存配置到 JSON 文件")
+    return parser
+
+
+def apply_overrides(cfg: Config, args: argparse.Namespace) -> Config:
+    mapping = {
+        "image_size": ("data", "image_size"),
+        "action_dim": ("data", "action_dim"),
+        "history_len": ("data", "history_len"),
+        "data_dir": ("data", "data_dir"),
+        "normalize_actions": ("data", "normalize_actions"),
+        "action_stats_path": ("data", "action_stats_path"),
+        "backbone": ("model", "cnn_backbone"),
+        "cnn_feature_dim": ("model", "cnn_feature_dim"),
+        "action_hidden_dim": ("model", "action_hidden_dim"),
+        "fused_hidden_dim": ("model", "fused_hidden_dim"),
+        "dropout": ("model", "dropout"),
+        "history_encoder": ("model", "history_encoder"),
+        "batch_size": ("train", "batch_size"),
+        "epochs": ("train", "epochs"),
+        "lr": ("train", "lr"),
+        "weight_decay": ("train", "weight_decay"),
+        "val_split": ("train", "val_split"),
+        "checkpoint_dir": ("train", "checkpoint_dir"),
+        "patience": ("train", "patience"),
+        "num_workers": ("train", "num_workers"),
+    }
+    args_dict = vars(args)
+    for arg_name, (section, field) in mapping.items():
+        val = args_dict.get(arg_name)
+        if val is not None:
+            setattr(getattr(cfg, section), field, val)
+    return cfg