utils/lora_utils.py

import yaml
import torch
import torch.nn as nn
import torch.nn.functional as F
import re

class LoRAConfig:
    def __init__(self, config_file):
        # Load the YAML configuration file
        with open(config_file, 'r') as file:
            config = yaml.safe_load(file)
        # self.config = config

        # Set class attributes based on the loaded YAML config
        for key, value in config.items():
            setattr(self, key, value)

class LoRALinear(nn.Module):
    def __init__(self, linear_layer, rank, scaling_rank, init_scale):
        super().__init__()
        self.in_features = linear_layer.in_features
        self.out_features = linear_layer.out_features
        self.rank = rank
        self.scaling_rank = scaling_rank
        self.weight = linear_layer.weight
        self.bias = linear_layer.bias
        if self.rank > 0:
            self.lora_a = nn.Parameter(torch.randn(rank, linear_layer.in_features) * init_scale)
            if init_scale < 0:
                self.lora_b = nn.Parameter(torch.randn(linear_layer.out_features, rank) * init_scale)
            else:
                self.lora_b = nn.Parameter(torch.zeros(linear_layer.out_features, rank))
        if self.scaling_rank:
            self.multi_lora_a = nn.Parameter(
                torch.ones(self.scaling_rank, linear_layer.in_features)
                + torch.randn(self.scaling_rank, linear_layer.in_features) * init_scale
            )
            if init_scale < 0:
                self.multi_lora_b = nn.Parameter(
                    torch.ones(linear_layer.out_features, self.scaling_rank)
                    + torch.randn(linear_layer.out_features, self.scaling_rank) * init_scale
                )
            else:
                self.multi_lora_b = nn.Parameter(torch.ones(linear_layer.out_features, self.scaling_rank))

    def forward(self, input):
        if self.scaling_rank == 1 and self.rank == 0:
            # parsimonious implementation for ia3 and lora scaling
            if self.multi_lora_a.requires_grad:
                hidden = F.linear((input * self.multi_lora_a.flatten()), self.weight, self.bias)
            else:
                hidden = F.linear(input, self.weight, self.bias)
            if self.multi_lora_b.requires_grad:
                hidden = hidden * self.multi_lora_b.flatten()
            return hidden
        else:
            # general implementation for lora (adding and scaling)
            weight = self.weight
            if self.scaling_rank:
                weight = weight * torch.matmul(self.multi_lora_b, self.multi_lora_a) / self.scaling_rank
            if self.rank:
                weight = weight + torch.matmul(self.lora_b, self.lora_a) / self.rank
            return F.linear(input, weight, self.bias)

    def extra_repr(self):
        return "in_features={}, out_features={}, bias={}, rank={}, scaling_rank={}".format(
            self.in_features, self.out_features, self.bias is not None, self.rank, self.scaling_rank
        )


def modify_with_lora(transformer, config):
    for m_name, module in dict(transformer.named_modules()).items():
        if re.fullmatch(config.lora_modules, m_name):
            for c_name, layer in dict(module.named_children()).items():
                if re.fullmatch(config.lora_layers, c_name):
                    assert isinstance(
                        layer, nn.Linear
                    ), f"LoRA can only be applied to torch.nn.Linear, but {layer} is {type(layer)}."
                    setattr(
                        module,
                        c_name,
                        LoRALinear(layer, config.lora_rank, config.lora_scaling_rank, config.lora_init_scale),
                    )
    return transformer