from __future__ import annotations

import torch
import torch.nn as nn

from .grad_reverse import GradientReversalLayer


class SensorDiscriminator(nn.Module):
    """Adversarial sensor classifier attached to shared features."""

    def __init__(self, in_dim: int, num_sensors: int, lambda_grl: float = 0.3):
        super().__init__()
        self.grl = GradientReversalLayer(lambda_=lambda_grl)
        self.mlp = nn.Sequential(
            nn.Linear(in_dim, 64),
            nn.ReLU(inplace=True),
            nn.Linear(64, num_sensors),
        )

    def set_grl_lambda(self, value: float) -> None:
        """Update GRL lambda dynamically (DANN warm-up)."""
        self.grl.set_lambda(value)

    def forward(self, features: torch.Tensor) -> torch.Tensor:
        return self.mlp(self.grl(features))