add lora finetune code

haoneng.lhn

2023-07-20 36c43d4c9f3ae98f026889b2f5f9726826a208d8

add lora finetune code

 egs_modelscope/asr/paraformer/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/finetune.py

@@ -19,7 +19,8 @@
        work_dir=params.output_dir,
        batch_bins=params.batch_bins,
        max_epoch=params.max_epoch,
        lr=params.lr)
        lr=params.lr,
        mate_params=params.param_dict)
    trainer = build_trainer(Trainers.speech_asr_trainer, default_args=kwargs)
    trainer.train()

@@ -30,7 +31,18 @@
    params.data_path = "./example_data/"            # 数据路径
    params.dataset_type = "small"                   # 小数据量设置small，若数据量大于1000小时，请使用large
    params.batch_bins = 2000                       # batch size，如果dataset_type="small"，batch_bins单位为fbank特征帧数，如果dataset_type="large"，batch_bins单位为毫秒，
    params.max_epoch = 50                           # 最大训练轮数
    params.max_epoch = 5                           # 最大训练轮数
    params.lr = 0.00005                             # 设置学习率
    init_param = []
    freeze_param = []
    ignore_init_mismatch = True
    use_lora = False
    params.param_dict = {"init_param":init_param, "freeze_param": freeze_param, "ignore_init_mismatch": ignore_init_mismatch}
    if use_lora:
        enable_lora = True
        lora_bias = "all"
        lora_params = {"lora_list":['q','v'], "lora_rank":8, "lora_alpha":16, "lora_dropout":0.1}
        lora_config = {"enable_lora": enable_lora, "lora_bias": lora_bias, "lora_params": lora_params}
        params.param_dict.update(lora_config)
    
    modelscope_finetune(params)

 funasr/bin/build_trainer.py

@@ -92,6 +92,14 @@
    for key, value in finetune_configs.items():
        if hasattr(args, key):
            setattr(args, key, value)
    if mate_params is not None:
        for key, value in mate_params.items():
            if hasattr(args, key):
                setattr(args, key, value)
    if mate_params is not None and "lora_params" in mate_params:
        lora_params = mate_params['lora_params']
        configs['encoder_conf'].update(lora_params) 
        configs['decoder_conf'].update(lora_params) 

    # prepare data
    args.dataset_type = dataset_type
@@ -106,6 +114,9 @@
    else:
        raise ValueError(f"Not supported dataset_type={args.dataset_type}")
    args.init_param = [init_param]
    if mate_params is not None and "init_param" in mate_params:
        if len(mate_params["init_param"]) != 0:
            args.init_param = mate_params["init_param"]
    args.cmvn_file = cmvn_file
    if os.path.exists(seg_dict_file):
        args.seg_dict_file = seg_dict_file
@@ -144,10 +155,6 @@
        args.patience = None
    args.local_rank = local_rank
    args.distributed = distributed
    if mate_params is not None:
        for key, value in mate_params.items():
            if hasattr(args, key):
                setattr(args, key, value)
    ASRTask.finetune_args = args

    return ASRTask

 funasr/modules/attention.py

@@ -338,7 +338,10 @@
            else:
                self.linear_out = nn.Linear(n_feat, n_feat)
            lora_qkv_list = ["q" in lora_list, "k" in lora_list, "v" in lora_list]
            self.linear_q_k_v = lora.MergedLinear(in_feat, n_feat * 3, r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout, enable_lora=lora_qkv_list)
            if lora_qkv_list == [False, False, False]:
                self.linear_q_k_v = nn.Linear(in_feat, n_feat * 3)
            else:
                self.linear_q_k_v = lora.MergedLinear(in_feat, n_feat * 3, r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout, enable_lora=lora_qkv_list)
        else:
            self.linear_out = nn.Linear(n_feat, n_feat)
            self.linear_q_k_v = nn.Linear(in_feat, n_feat * 3)
@@ -562,11 +565,18 @@
        if lora_list is not None:
            if "q" in lora_list:
                self.linear_q = lora.Linear(n_feat, n_feat, r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout)
            else:
                self.linear_q = nn.Linear(n_feat, n_feat)
            lora_kv_list = ["k" in lora_list, "v" in lora_list]
            self.linear_k_v = lora.MergedLinear(n_feat if encoder_output_size is None else encoder_output_size, n_feat * 2, 
                                  r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout, enable_lora=lora_kv_list)
            if lora_kv_list == [False, False]:
                self.linear_k_v = nn.Linear(n_feat if encoder_output_size is None else encoder_output_size, n_feat*2)
            else:
                self.linear_k_v = lora.MergedLinear(n_feat if encoder_output_size is None else encoder_output_size, n_feat * 2, 
                                      r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout, enable_lora=lora_kv_list)
            if "o" in lora_list:
                self.linear_out = lora.Linear(n_feat, n_feat, r=lora_rank, lora_alpha=lora_alpha, lora_dropout=lora_dropout)
            else:
                self.linear_out = nn.Linear(n_feat, n_feat)
        else:
            self.linear_q = nn.Linear(n_feat, n_feat)
            self.linear_k_v = nn.Linear(n_feat if encoder_output_size is None else encoder_output_size, n_feat*2)

 funasr/modules/lora/layers.py

@@ -11,9 +11,9 @@

class LoRALayer():
    def __init__(
        self, 
        r: int, 
        lora_alpha: int, 
        self,
        r: int,
        lora_alpha: int,
        lora_dropout: float,
        merge_weights: bool,
    ):
@@ -61,40 +61,42 @@

    def train(self, mode: bool = True):
        nn.Embedding.train(self, mode)
        if mode:
            if self.merge_weights and self.merged:
                # Make sure that the weights are not merged
                if self.r > 0:
                    self.weight.data -= (self.lora_B @ self.lora_A).transpose(0, 1) * self.scaling
                self.merged = False
        else:
            if self.merge_weights and not self.merged:
                # Merge the weights and mark it
                if self.r > 0:
                    self.weight.data += (self.lora_B @ self.lora_A).transpose(0, 1) * self.scaling
                self.merged = True
        
        if self.merge_weights and self.merged:
            # Make sure that the weights are not merged
            if self.r > 0:
                self.weight.data -= (self.lora_B @ self.lora_A).T * self.scaling
            self.merged = False

    def eval(self):
        nn.Linear.eval(self)
        if self.merge_weights and not self.merged:
            # Merge the weights and mark it
            if self.r > 0:
                self.weight.data += (self.lora_B @ self.lora_A) * self.scaling
            self.merged = True

    def forward(self, x: torch.Tensor):
        if self.r > 0 and not self.merged:
            result = nn.Embedding.forward(self, x)
            after_A = F.embedding(
                x, self.lora_A.transpose(0, 1), self.padding_idx, self.max_norm,
                self.norm_type, self.scale_grad_by_freq, self.sparse
            )
            result += (after_A @ self.lora_B.transpose(0, 1)) * self.scaling
            if self.r > 0:
                after_A = F.embedding(
                    x, self.lora_A.T, self.padding_idx, self.max_norm,
                    self.norm_type, self.scale_grad_by_freq, self.sparse
                )
                result += (after_A @ self.lora_B.T) * self.scaling
            return result
        else:
            return nn.Embedding.forward(self, x)
            


class Linear(nn.Linear, LoRALayer):
    # LoRA implemented in a dense layer
    def __init__(
        self, 
        in_features: int, 
        out_features: int, 
        r: int = 0, 
        lora_alpha: int = 1, 
        self,
        in_features: int,
        out_features: int,
        r: int = 0,
        lora_alpha: int = 1,
        lora_dropout: float = 0.,
        fan_in_fan_out: bool = False, # Set this to True if the layer to replace stores weight like (fan_in, fan_out)
        merge_weights: bool = True,
@@ -114,7 +116,7 @@
            self.weight.requires_grad = False
        self.reset_parameters()
        if fan_in_fan_out:
            self.weight.data = self.weight.data.transpose(0, 1)
            self.weight.data = self.weight.data.T

    def reset_parameters(self):
        nn.Linear.reset_parameters(self)
@@ -125,27 +127,31 @@

    def train(self, mode: bool = True):
        def T(w):
            return w.transpose(0, 1) if self.fan_in_fan_out else w
            return w.T if self.fan_in_fan_out else w
        nn.Linear.train(self, mode)
        if mode:
            if self.merge_weights and self.merged:
                # Make sure that the weights are not merged
                if self.r > 0:
                    self.weight.data -= T(self.lora_B @ self.lora_A) * self.scaling
                self.merged = False
        else:
            if self.merge_weights and not self.merged:
                # Merge the weights and mark it
                if self.r > 0:
                    self.weight.data += T(self.lora_B @ self.lora_A) * self.scaling
                self.merged = True       
        if self.merge_weights and self.merged:
            # Make sure that the weights are not merged
            if self.r > 0:
                self.weight.data -= T(self.lora_B @ self.lora_A) * self.scaling
            self.merged = False

    def eval(self):
        def T(w):
            return w.T if self.fan_in_fan_out else w
        nn.Linear.eval(self)
        if self.merge_weights and not self.merged:
            # Merge the weights and mark it
            if self.r > 0:
                self.weight.data += T(self.lora_B @ self.lora_A) * self.scaling
            self.merged = True

    def forward(self, x: torch.Tensor):
        def T(w):
            return w.transpose(0, 1) if self.fan_in_fan_out else w
            return w.T if self.fan_in_fan_out else w
        if self.r > 0 and not self.merged:
            result = F.linear(x, T(self.weight), bias=self.bias)            
            result += (self.lora_dropout(x) @ self.lora_A.transpose(0, 1) @ self.lora_B.transpose(0, 1)) * self.scaling
            result = F.linear(x, T(self.weight), bias=self.bias)
            if self.r > 0:
                result += (self.lora_dropout(x) @ self.lora_A.T @ self.lora_B.T) * self.scaling
            return result
        else:
            return F.linear(x, T(self.weight), bias=self.bias)
@@ -154,11 +160,11 @@
class MergedLinear(nn.Linear, LoRALayer):
    # LoRA implemented in a dense layer
    def __init__(
        self, 
        in_features: int, 
        out_features: int, 
        r: int = 0, 
        lora_alpha: int = 1, 
        self,
        in_features: int,
        out_features: int,
        r: int = 0,
        lora_alpha: int = 1,
        lora_dropout: float = 0.,
        enable_lora: List[bool] = [False],
        fan_in_fan_out: bool = False,
@@ -190,7 +196,7 @@
            self.lora_ind = self.lora_ind.view(-1)
        self.reset_parameters()
        if fan_in_fan_out:
            self.weight.data = self.weight.data.transpose(0, 1)
            self.weight.data = self.weight.data.T

    def reset_parameters(self):
        nn.Linear.reset_parameters(self)
@@ -209,34 +215,37 @@

    def train(self, mode: bool = True):
        def T(w):
            return w.transpose(0, 1) if self.fan_in_fan_out else w
            return w.T if self.fan_in_fan_out else w
        nn.Linear.train(self, mode)
        if mode:
            if self.merge_weights and self.merged:
                # Make sure that the weights are not merged
                if self.r > 0 and any(self.enable_lora):
                    delta_w = F.conv1d(
                        self.lora_A.data.unsqueeze(0), 
                        self.lora_B.data.unsqueeze(-1), 
                        groups=sum(self.enable_lora)
                    ).squeeze(0)
                    self.weight.data -= self.zero_pad(T(delta_w * self.scaling))
                self.merged = False
        else:
            if self.merge_weights and not self.merged:
                # Merge the weights and mark it
                if self.r > 0 and any(self.enable_lora):
                    delta_w = F.conv1d(
                        self.lora_A.data.unsqueeze(0), 
                        self.lora_B.data.unsqueeze(-1), 
                        groups=sum(self.enable_lora)
                    ).squeeze(0)
                    self.weight.data += self.zero_pad(T(delta_w * self.scaling))
                self.merged = True        
        if self.merge_weights and self.merged:
            # Make sure that the weights are not merged
            if self.r > 0 and any(self.enable_lora):
                delta_w = F.conv1d(
                    self.lora_A.data.unsqueeze(0),
                    self.lora_B.data.unsqueeze(-1),
                    groups=sum(self.enable_lora)
                ).squeeze(0)
                self.weight.data -= self.zero_pad(T(delta_w * self.scaling))
            self.merged = False

    def eval(self):
        def T(w):
            return w.T if self.fan_in_fan_out else w
        nn.Linear.eval(self)
        if self.merge_weights and not self.merged:
            # Merge the weights and mark it
            if self.r > 0 and any(self.enable_lora):
                delta_w = F.conv1d(
                    self.lora_A.data.unsqueeze(0),
                    self.lora_B.data.unsqueeze(-1),
                    groups=sum(self.enable_lora)
                ).squeeze(0)
                self.weight.data += self.zero_pad(T(delta_w * self.scaling))
            self.merged = True

    def forward(self, x: torch.Tensor):
        def T(w):
            return w.transpose(0, 1) if self.fan_in_fan_out else w
            return w.T if self.fan_in_fan_out else w
        if self.merged:
            return F.linear(x, T(self.weight), bias=self.bias)
        else:
@@ -244,76 +253,71 @@
            if self.r > 0:
                after_A = F.linear(self.lora_dropout(x), self.lora_A)
                after_B = F.conv1d(
                    after_A.transpose(-2, -1), 
                    self.lora_B.unsqueeze(-1), 
                    after_A.transpose(-2, -1),
                    self.lora_B.unsqueeze(-1),
                    groups=sum(self.enable_lora)
                ).transpose(-2, -1)
                result += self.zero_pad(after_B) * self.scaling
            return result
        

class ConvLoRA(nn.Module, LoRALayer):
    def __init__(self, conv_module, in_channels, out_channels, kernel_size, r=0, lora_alpha=1, lora_dropout=0., merge_weights=True, **kwargs):
        super(ConvLoRA, self).__init__()
        self.conv = conv_module(in_channels, out_channels, kernel_size, **kwargs)
        LoRALayer.__init__(self, r=r, lora_alpha=lora_alpha, lora_dropout=lora_dropout, merge_weights=merge_weights)
        assert isinstance(kernel_size, int)

class Conv2d(nn.Conv2d, LoRALayer):
    # LoRA implemented in a dense layer
    def __init__(
        self,
        in_channels: int,
        out_channels: int,
        kernel_size: int,
        r: int = 0,
        lora_alpha: int = 1,
        lora_dropout: float = 0.,
        merge_weights: bool = True,
        **kwargs
    ):
        nn.Conv2d.__init__(self, in_channels, out_channels, kernel_size, **kwargs)
        LoRALayer.__init__(self, r=r, lora_alpha=lora_alpha, lora_dropout=lora_dropout,
                           merge_weights=merge_weights)
        assert type(kernel_size) is int
        # Actual trainable parameters
        if r > 0:
            self.lora_A = nn.Parameter(
                self.conv.weight.new_zeros((r * kernel_size, in_channels * kernel_size))
                self.weight.new_zeros((r*kernel_size, in_channels*kernel_size))
            )
            self.lora_B = nn.Parameter(
              self.conv.weight.new_zeros((out_channels//self.conv.groups*kernel_size, r*kernel_size))
                self.weight.new_zeros((out_channels*kernel_size, r*kernel_size))
            )
            self.scaling = self.lora_alpha / self.r
            # Freezing the pre-trained weight matrix
            self.conv.weight.requires_grad = False
            self.weight.requires_grad = False
        self.reset_parameters()
        self.merged = False

    def reset_parameters(self):
        self.conv.reset_parameters()
        nn.Conv2d.reset_parameters(self)
        if hasattr(self, 'lora_A'):
            # initialize A the same way as the default for nn.Linear and B to zero
            nn.init.kaiming_uniform_(self.lora_A, a=math.sqrt(5))
            nn.init.zeros_(self.lora_B)

    def train(self, mode=True):
        super(ConvLoRA, self).train(mode)
        if mode:
            if self.merge_weights and self.merged:
                if self.r > 0:
                    # Make sure that the weights are not merged
                    self.conv.weight.data -= (self.lora_B @ self.lora_A).view(self.conv.weight.shape) * self.scaling
                self.merged = False
        else:
            if self.merge_weights and not self.merged:
                if self.r > 0:
                    # Merge the weights and mark it
                    self.conv.weight.data += (self.lora_B @ self.lora_A).view(self.conv.weight.shape) * self.scaling
                self.merged = True
    def train(self, mode: bool = True):
        nn.Conv2d.train(self, mode)
        if self.merge_weights and self.merged:
            # Make sure that the weights are not merged
            self.weight.data -= (self.lora_B @ self.lora_A).view(self.weight.shape) * self.scaling
            self.merged = False

    def forward(self, x):
    def eval(self):
        nn.Conv2d.eval(self)
        if self.merge_weights and not self.merged:
            # Merge the weights and mark it
            self.weight.data += (self.lora_B @ self.lora_A).view(self.weight.shape) * self.scaling
            self.merged = True

    def forward(self, x: torch.Tensor):
        if self.r > 0 and not self.merged:
            return self.conv._conv_forward(
                x, 
                self.conv.weight + (self.lora_B @ self.lora_A).view(self.conv.weight.shape) * self.scaling,
                self.conv.bias
            return F.conv2d(
                x,
                self.weight + (self.lora_B @ self.lora_A).view(self.weight.shape) * self.scaling,
                self.bias, self.stride, self.padding, self.dilation, self.groups
            )
        return self.conv(x)

class Conv2d(ConvLoRA):
    def __init__(self, *args, **kwargs):
        super(Conv2d, self).__init__(nn.Conv2d, *args, **kwargs)

class Conv1d(ConvLoRA):
    def __init__(self, *args, **kwargs):
        super(Conv1d, self).__init__(nn.Conv1d, *args, **kwargs)

# Can Extend to other ones like this

class Conv3d(ConvLoRA):
    def __init__(self, *args, **kwargs):
        super(Conv3d, self).__init__(nn.Conv3d, *args, **kwargs)
        return nn.Conv2d.forward(self, x)