RTFormer 预训练

[mandylyin]

问题确认 Search before asking

• 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer.

请提出你的问题 Please ask your question

你好，请问怎么复现论文里的ImageNet预训练结果？我把RTFormer拷贝到paddleclas里，添加了分类头，修改了forward函数，其它没修改。仅使用RandCropImage和RandFlipImage数据增强top1大概66%，添加了其它数据增强后top1就只有不到20%，请问是什么原因导致的？我用4张卡训练。

分类头代码如下：

self.pool2d_avg = AdaptiveAvgPool2D(1)
self.fc = nn.Linear(
           base_chs * 4,
           class_num,
           weight_attr=ParamAttr(name="fc_weights"),
           bias_attr=ParamAttr(name="fc_offset"))

forward代码：

def forward(self, x):
      x1 = self.layer1(self.conv1(x))  # c, 1/4
      x2 = self.layer2(self.relu(x1))  # 2c, 1/8
      x3 = self.layer3(self.relu(x2))  # 4c, 1/16
      x3_ = x2 + F.interpolate(
           self.compression3(x3), size=paddle.shape(x2)[2:], mode='bilinear')
      x3_ = self.layer3_(self.relu(x3_))  # 2c, 1/8

      x4_, x4 = self.layer4(
           [self.relu(x3_), self.relu(x3)])  # 2c, 1/8; 8c, 1/16
      x5_, x5 = self.layer5(
           [self.relu(x4_), self.relu(x4)])  # 2c, 1/8; 8c, 1/32

      x6 = self.spp(x5)
      x6 = F.interpolate(
           x6, size=paddle.shape(x5_)[2:], mode='bilinear')  # 2c, 1/8
      x7 = paddle.concat([x5_, x6], axis=1) # 4c, 1/8
      x8 = self.pool2d_avg(x7)
      x9 = paddle.flatten(x8, start_axis=1, stop_axis=-1)
      out = self.fc(x9)

      return out

# global configs
Global:
  checkpoints: null
  pretrained_model: null
  output_dir: ./output/
  device: gpu
  save_interval: 1
  eval_during_train: True
  eval_interval: 1
  epochs: 300
  print_batch_step: 10
  use_visualdl: True
  # used for static mode and model export
  image_shape: [3, 224, 224]
  save_inference_dir: ./inference
  # use_dali: True

# model architecture
Arch:
  name: RTFormer_slim
  class_num: 1000
 
# loss function config for traing/eval process
Loss:
  Train:
    - CELoss:
        weight: 1.0
        epsilon: 0.1
  Eval:
    - CELoss:
        weight: 1.0

Optimizer:
  name: AdamW
  beta1: 0.9
  beta2: 0.999
  epsilon: 1e-8
  weight_decay: 0.04
  no_weight_decay_name: .bias norm
  one_dim_param_no_weight_decay: True
  lr:
    name: Cosine
    learning_rate: 0.0005
    eta_min: 0.0002
    warmup_epoch: 5
    warmup_start_lr: 5e-7

# data loader for train and eval
DataLoader:
  Train:
    dataset:
      name: ImageNetDataset
      image_root: ../../datasets/ILSVRC2012/
      cls_label_path: ../../datasets/ILSVRC2012/train.txt
      relabel: True
      transform_ops:
        - DecodeImage:
            to_rgb: True
            channel_first: False
        - RandCropImage:
            size: 224
            interpolation: bicubic
            backend: pil
        - RandFlipImage:
            flip_code: 1
        # - TimmAutoAugment:
        #     config_str: rand-m9-mstd0.5-inc1
        #     interpolation: bicubic
        #     img_size: 224
        - NormalizeImage:
            scale: 1.0/255.0
            mean: [0.485, 0.456, 0.406]
            std: [0.229, 0.224, 0.225]
            order: ''
        - RandomErasing:
            EPSILON: 0.25
            sl: 0.02
            sh: 1.0/3.0
            r1: 0.3
            attempt: 10
            use_log_aspect: True
            mode: pixel
      # batch_transform_ops:
      #   - OpSampler:
      #       MixupOperator:
      #         alpha: 0.2
      #         prob: 0.5
      #       CutmixOperator:
      #         alpha: 1.0
      #         prob: 0.5

    sampler:
      name: DistributedBatchSampler
      batch_size: 256
      drop_last: False
      shuffle: True
    loader:
      num_workers: 8
      use_shared_memory: True

  Eval:
    dataset: 
      name: ImageNetDataset
      image_root: ../../datasets/ILSVRC2012/
      cls_label_path: ../../datasets/ILSVRC2012/val.txt
      relabel: True
      transform_ops:
        - DecodeImage:
            to_rgb: True
            channel_first: False
        - ResizeImage:
            resize_short: 256
        - CropImage:
            size: 224
        - NormalizeImage:
            scale: 1.0/255.0
            mean: [0.485, 0.456, 0.406]
            std: [0.229, 0.224, 0.225]
            order: ''
    sampler:
      name: DistributedBatchSampler
      batch_size: 256
      drop_last: False
      shuffle: False
    loader:
      num_workers: 4
      use_shared_memory: True

Infer:
  infer_imgs: docs/images/inference_deployment/whl_demo.jpg
  batch_size: 10
  transforms:
    - DecodeImage:
        to_rgb: True
        channel_first: False
    - ResizeImage:
        resize_short: 256
    - CropImage:
        size: 224
    - NormalizeImage:
        scale: 1.0/255.0
        mean: [0.485, 0.456, 0.406]
        std: [0.229, 0.224, 0.225]
        order: ''
    - ToCHWImage:
  PostProcess:
    name: Topk
    topk: 5
    class_id_map_file: ppcls/utils/imagenet1k_label_list.txt

Metric:
  Train:
    - TopkAcc:
        topk: [1, 5]
  Eval:
    - TopkAcc:
        topk: [1, 5]

[zhang-prog]

论文中的 minimum lr: 5e-6，你这里配置的 eta_min: 0.0002，差的有点多，可以改掉这里再试试。

[mandylyin]

最一开始用的是5e-6，加了其它数据增强后top1就总上不去，被我一通改，也没跑出个所以然来。我再改回去试试。

[zhang-prog]

好的，可以参考论文里面的参数再试试看