import torch
import torch.nn as nn

class LoRALinearLayer(nn.Module):
    def __init__(self, in_features, out_features, rank=4, network_alpha=None, device=None, dtype=None):
        super().__init__()

        self.down = nn.Linear(in_features, rank, bias=False, device=device, dtype=dtype)
        self.up = nn.Linear(rank, out_features, bias=False, device=device, dtype=dtype)
        self.network_alpha = network_alpha
        self.rank = rank

        nn.init.normal_(self.down.weight, std=1 / rank)
        nn.init.zeros_(self.up.weight)

    def forward(self, hidden_states):
        orig_dtype = hidden_states.dtype
        dtype = self.down.weight.dtype

        down_hidden_states = self.down(hidden_states.to(dtype))
        up_hidden_states = self.up(down_hidden_states)

        if self.network_alpha is not None:
            up_hidden_states *= self.network_alpha / self.rank

        return up_hidden_states.to(orig_dtype)