utils.py

# --------------------------------------------------------
# Swin Transformer
# Copyright (c) 2021 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ze Liu
# --------------------------------------------------------

import os

import torch
import torch.distributed as dist

try:
    from torch._six import inf
except:
    from torch import inf

import torch.nn as nn

def load_checkpoint(config, model, optimizer, lr_scheduler, loss_scaler, logger):
    logger.info(f"==============> Resuming form {config.MODEL.RESUME}....................")
    if config.MODEL.RESUME.startswith('https'):
        checkpoint = torch.hub.load_state_dict_from_url(
            config.MODEL.RESUME, map_location='cpu', check_hash=True)
    else:
        checkpoint = torch.load(config.MODEL.RESUME, map_location='cpu')

    msg = model.load_state_dict(checkpoint['model'], strict=False)
    logger.info(msg)
    max_accuracy = 0.0
    if not config.EVAL_MODE and 'optimizer' in checkpoint and 'lr_scheduler' in checkpoint and 'epoch' in checkpoint:
        optimizer.load_state_dict(checkpoint['optimizer'])
        lr_scheduler.load_state_dict(checkpoint['lr_scheduler'])
        config.defrost()
        config.TRAIN.START_EPOCH = checkpoint['epoch'] + 1
        config.freeze()
        if 'scaler' in checkpoint:
            loss_scaler.load_state_dict(checkpoint['scaler'])
        logger.info(f"=> loaded successfully '{config.MODEL.RESUME}' (epoch {checkpoint['epoch']})")
        if 'max_accuracy' in checkpoint:
            max_accuracy = checkpoint['max_accuracy']

    del checkpoint
    torch.cuda.empty_cache()
    return max_accuracy


def load_pretrained(config, model, logger):
    logger.info(f"==============> Loading weight {config.MODEL.PRETRAINED} for fine-tuning......")
    checkpoint = torch.load(config.MODEL.PRETRAINED, map_location='cpu')
    state_dict = checkpoint['model']

    # delete relative_position_index since we always re-init it
    relative_position_index_keys = [k for k in state_dict.keys() if "relative_position_index" in k]
    for k in relative_position_index_keys:
        del state_dict[k]

    # delete relative_coords_table since we always re-init it
    relative_position_index_keys = [k for k in state_dict.keys() if "relative_coords_table" in k]
    for k in relative_position_index_keys:
        del state_dict[k]

    # delete attn_mask since we always re-init it
    attn_mask_keys = [k for k in state_dict.keys() if "attn_mask" in k]
    for k in attn_mask_keys:
        del state_dict[k]

    # bicubic interpolate relative_position_bias_table if not match
    relative_position_bias_table_keys = [k for k in state_dict.keys() if "relative_position_bias_table" in k]
    for k in relative_position_bias_table_keys:
        relative_position_bias_table_pretrained = state_dict[k]
        relative_position_bias_table_current = model.state_dict()[k]
        L1, nH1 = relative_position_bias_table_pretrained.size()
        L2, nH2 = relative_position_bias_table_current.size()
        if nH1 != nH2:
            logger.warning(f"Error in loading {k}, passing......")
        else:
            if L1 != L2:
                # bicubic interpolate relative_position_bias_table if not match
                S1 = int(L1 ** 0.5)
                S2 = int(L2 ** 0.5)
                relative_position_bias_table_pretrained_resized = torch.nn.functional.interpolate(
                    relative_position_bias_table_pretrained.permute(1, 0).view(1, nH1, S1, S1), size=(S2, S2),
                    mode='bicubic')
                state_dict[k] = relative_position_bias_table_pretrained_resized.view(nH2, L2).permute(1, 0)

    # bicubic interpolate absolute_pos_embed if not match
    absolute_pos_embed_keys = [k for k in state_dict.keys() if "absolute_pos_embed" in k]
    for k in absolute_pos_embed_keys:
        # dpe
        absolute_pos_embed_pretrained = state_dict[k]
        absolute_pos_embed_current = model.state_dict()[k]
        _, L1, C1 = absolute_pos_embed_pretrained.size()
        _, L2, C2 = absolute_pos_embed_current.size()
        if C1 != C1:
            logger.warning(f"Error in loading {k}, passing......")
        else:
            if L1 != L2:
                S1 = int(L1 ** 0.5)
                S2 = int(L2 ** 0.5)
                absolute_pos_embed_pretrained = absolute_pos_embed_pretrained.reshape(-1, S1, S1, C1)
                absolute_pos_embed_pretrained = absolute_pos_embed_pretrained.permute(0, 3, 1, 2)
                absolute_pos_embed_pretrained_resized = torch.nn.functional.interpolate(
                    absolute_pos_embed_pretrained, size=(S2, S2), mode='bicubic')
                absolute_pos_embed_pretrained_resized = absolute_pos_embed_pretrained_resized.permute(0, 2, 3, 1)
                absolute_pos_embed_pretrained_resized = absolute_pos_embed_pretrained_resized.flatten(1, 2)
                state_dict[k] = absolute_pos_embed_pretrained_resized

    if config.MODEL.DEEP_SUPERVISION:
        for head_layer in model.head_layers_g:
            torch.nn.init.constant_(head_layer.bias, 0.)
            torch.nn.init.constant_(head_layer.weight, 0.)
        for head_layer in model.head_layers_l:
            torch.nn.init.constant_(head_layer.bias, 0.)
            torch.nn.init.constant_(head_layer.weight, 0.)

    torch.nn.init.constant_(model.head.bias, 0.)
    torch.nn.init.constant_(model.head.weight, 0.)
    del state_dict['head.weight']
    del state_dict['head.bias']
    logger.warning(f"Error in loading classifier head, re-init classifier head to 0")

    if config.MODEL.FINE_TUNE:
        msg = model.load_state_dict(state_dict, strict=False)
    else:
        # 初始化空的权重字典
        new_state_dict = {}

        # 处理预训练模型的权重
        for key, value in state_dict.items():
            if key.startswith('patch_embed'):
                # 将权重分配给两个分支
                new_state_dict[key.replace('patch_embed', 'patch_embed_g')] = value
                new_state_dict[key.replace('patch_embed', 'patch_embed_l')] = value.clone()
            elif key.startswith('absolute_pos_embed'):
                # 同样的处理
                new_state_dict[key.replace('absolute_pos_embed', 'absolute_pos_embed_g')] = value
                new_state_dict[key.replace('absolute_pos_embed', 'absolute_pos_embed_l')] = value.clone()
            elif key.startswith('layers'):
                # 处理每一层的权重
                if 'layer' in key:
                    # 分支权重
                    new_state_dict[key.replace('layers.', 'layers_g.')] = value
                    new_state_dict[key.replace('layers.', 'layers_l.')] = value.clone()
            elif key.startswith('norm'):
                # 处理norm层的权重
                new_state_dict[key.replace('norm', 'norm_g')] = value
                new_state_dict[key.replace('norm', 'norm_l')] = value.clone()
            else:
                # 其他权重
                new_state_dict[key] = value

        # 加载新的权重到模型中
        msg = model.load_state_dict(new_state_dict, strict=False)

    logger.warning(msg)
    logger.info(f"=> loaded successfully '{config.MODEL.PRETRAINED}'")

    del checkpoint
    torch.cuda.empty_cache()


def save_checkpoint(config, epoch, model, max_accuracy, optimizer, lr_scheduler, loss_scaler, logger, tag=None):
    save_state = {'model': model.state_dict(),
                  'optimizer': optimizer.state_dict(),
                  'lr_scheduler': lr_scheduler.state_dict(),
                  'max_accuracy': max_accuracy,
                  'scaler': loss_scaler.state_dict(),
                  'epoch': epoch,
                  'config': config}
    if tag is not None:
        save_path = os.path.join(config.OUTPUT, f'{tag}.pth')
        logger.info(f"{save_path} saving......")
        torch.save(save_state, save_path)
        logger.info(f"{save_path} saved !!!")
    else:
        save_path = os.path.join(config.OUTPUT, f'ckpt_epoch_{epoch}.pth')
        logger.info(f"{save_path} saving......")
        torch.save(save_state, save_path)
        logger.info(f"{save_path} saved !!!")


def get_grad_norm(parameters, norm_type=2):
    if isinstance(parameters, torch.Tensor):
        parameters = [parameters]
    parameters = list(filter(lambda p: p.grad is not None, parameters))
    norm_type = float(norm_type)
    total_norm = 0
    for p in parameters:
        param_norm = p.grad.data.norm(norm_type)
        total_norm += param_norm.item() ** norm_type
    total_norm = total_norm ** (1. / norm_type)
    return total_norm


def auto_resume_helper(output_dir):
    checkpoints = os.listdir(output_dir)
    checkpoints = [ckpt for ckpt in checkpoints if ckpt.endswith('pth')]
    print(f"All checkpoints founded in {output_dir}: {checkpoints}")
    if len(checkpoints) > 0:
        latest_checkpoint = max([os.path.join(output_dir, d) for d in checkpoints], key=os.path.getmtime)
        print(f"The latest checkpoint founded: {latest_checkpoint}")
        resume_file = latest_checkpoint
    else:
        resume_file = None
    return resume_file


def reduce_tensor(tensor):
    rt = tensor.clone()
    dist.all_reduce(rt, op=dist.ReduceOp.SUM)
    rt /= dist.get_world_size()
    return rt


def ampscaler_get_grad_norm(parameters, norm_type: float = 2.0) -> torch.Tensor:
    if isinstance(parameters, torch.Tensor):
        parameters = [parameters]
    parameters = [p for p in parameters if p.grad is not None]
    norm_type = float(norm_type)
    if len(parameters) == 0:
        return torch.tensor(0.)
    device = parameters[0].grad.device
    if norm_type == inf:
        total_norm = max(p.grad.detach().abs().max().to(device) for p in parameters)
    else:
        total_norm = torch.norm(torch.stack([torch.norm(p.grad.detach(),
                                                        norm_type).to(device) for p in parameters]), norm_type)
    return total_norm


class NativeScalerWithGradNormCount:
    state_dict_key = "amp_scaler"

    def __init__(self):
        self._scaler = torch.cuda.amp.GradScaler()

    def __call__(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True):
        self._scaler.scale(loss).backward(create_graph=create_graph)
        if update_grad:
            if clip_grad is not None:
                assert parameters is not None
                self._scaler.unscale_(optimizer)  # unscale the gradients of optimizer's assigned params in-place
                norm = torch.nn.utils.clip_grad_norm_(parameters, clip_grad)
            else:
                self._scaler.unscale_(optimizer)
                norm = ampscaler_get_grad_norm(parameters)
            self._scaler.step(optimizer)
            self._scaler.update()
        else:
            norm = None
        return norm

    def state_dict(self):
        return self._scaler.state_dict()

    def load_state_dict(self, state_dict):
        self._scaler.load_state_dict(state_dict)


class LabelSmoothingBCEWithLogitsLoss(nn.Module):
    def __init__(self, smoothing=0.1):
        super(LabelSmoothingBCEWithLogitsLoss, self).__init__()
        self.smoothing = smoothing
        self.bce_loss = nn.BCEWithLogitsLoss()

    def forward(self, logits, targets):
        smoothed_targets = targets * (1 - self.smoothing) + (self.smoothing / 2.0)

        # 计算损失
        return self.bce_loss(logits, smoothed_targets)