seresnext101d_32x8d.ah_in1k的模型卡片

一种带有挤压激活通道注意力的SE-ResNeXt-D图像分类模型。

该模型特点如下：

• ReLU激活函数
• 3层3x3卷积的干预层，带有池化
• 2x2平均池化 + 1x1卷积的快捷下采样
• 分组的3x3瓶颈卷积
• 挤压激活通道注意力

在timm中使用下面描述的配方模板对ImageNet-1k进行训练。

配方细节：

• 基于 ResNet Strikes Back A1 配方
• LAMB优化器
• 没有使用CutMix。比论文A1配方中的更强的dropout、随机深度和RandAugment
• 余弦学习率调度与预热

模型细节

• 模型类型：图像分类/特征主干
• 模型统计信息：
  • 参数数量(M)：93.6
  • GMACs：16.7
  • 激活数量(M)：32.0
  • 图像大小：训练=224 x 224，测试=288 x 288
• 论文：
  • ResNet反击：timm中的改进训练过程： https://arxiv.org/abs/2110.00476
  • 深度神经网络的聚合残差变换： https://arxiv.org/abs/1611.05431
  • 图像识别的深度残差学习： https://arxiv.org/abs/1512.03385
  • 挤压激励网络： https://arxiv.org/abs/1709.01507
  • 使用卷积神经网络的图像分类的各种技巧： https://arxiv.org/abs/1812.01187
• 原始链接： https://github.com/huggingface/pytorch-image-models

模型用法

图像分类

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model('seresnext101d_32x8d.ah_in1k', pretrained=True)
model = model.eval()

# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0))  # unsqueeze single image into batch of 1

top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)

特征图提取

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model(
    'seresnext101d_32x8d.ah_in1k',
    pretrained=True,
    features_only=True,
)
model = model.eval()

# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0))  # unsqueeze single image into batch of 1

for o in output:
    # print shape of each feature map in output
    # e.g.:
    #  torch.Size([1, 64, 112, 112])
    #  torch.Size([1, 256, 56, 56])
    #  torch.Size([1, 512, 28, 28])
    #  torch.Size([1, 1024, 14, 14])
    #  torch.Size([1, 2048, 7, 7])

    print(o.shape)

图像嵌入

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model(
    'seresnext101d_32x8d.ah_in1k',
    pretrained=True,
    num_classes=0,  # remove classifier nn.Linear
)
model = model.eval()

# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0))  # output is (batch_size, num_features) shaped tensor

# or equivalently (without needing to set num_classes=0)

output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 2048, 7, 7) shaped tensor

output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor

模型比较

在timm model results 中，可以探索此模型的数据集和运行时指标。

引用

@inproceedings{wightman2021resnet,
  title={ResNet strikes back: An improved training procedure in timm},
  author={Wightman, Ross and Touvron, Hugo and Jegou, Herve},
  booktitle={NeurIPS 2021 Workshop on ImageNet: Past, Present, and Future}
}

@misc{rw2019timm,
  author = {Ross Wightman},
  title = {PyTorch Image Models},
  year = {2019},
  publisher = {GitHub},
  journal = {GitHub repository},
  doi = {10.5281/zenodo.4414861},
  howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}

@article{Xie2016,
  title={Aggregated Residual Transformations for Deep Neural Networks},
  author={Saining Xie and Ross Girshick and Piotr Dollár and Zhuowen Tu and Kaiming He},
  journal={arXiv preprint arXiv:1611.05431},
  year={2016}
}

@article{He2015,
  author = {Kaiming He and Xiangyu Zhang and Shaoqing Ren and Jian Sun},
  title = {Deep Residual Learning for Image Recognition},
  journal = {arXiv preprint arXiv:1512.03385},
  year = {2015}
}

@inproceedings{hu2018senet,
  title={Squeeze-and-Excitation Networks},
  author={Jie Hu and Li Shen and Gang Sun},
  journal={IEEE Conference on Computer Vision and Pattern Recognition},
  year={2018}
}

@article{He2018BagOT,
  title={Bag of Tricks for Image Classification with Convolutional Neural Networks},
  author={Tong He and Zhi Zhang and Hang Zhang and Zhongyue Zhang and Junyuan Xie and Mu Li},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2018},
  pages={558-567}
}