cs3sedarknet_l.c2ns_in1k模型卡片

一个CS3-SE-DarkNet（跨阶段部分，包含3个卷积和挤压和激励信道注意力）图像分类模型。使用timm中描述的配方模板在ImageNet-1k上进行训练。

配方详情：

• 基于 ResNet Strikes Back 个不重复增强和更强的混合的C个配方
• 使用SGD（带Nesterov的）优化器和AGC（自适应梯度剪裁）
• 此ns变体的配方中没有使用随机深度
• 使用余弦LR调度和热身启动

模型详情

• 模型类型：图像分类/特征骨干
• 模型统计数据：
  • 参数（M）：21.9
  • GMACs：4.9
  • 激活函数（M）：8.6
  • 图像尺寸：训练 = 256 x 256，测试 = 288 x 288
• 论文：
  • CSPNet：一个可以增强CNN学习能力的新骨干网络： https://arxiv.org/abs/1911.11929
  • YOLOv3：一种增量改进： https://arxiv.org/abs/1804.02767
  • ResNet回归：timm中的改进训练流程： https://arxiv.org/abs/2110.00476
• 原始： https://github.com/huggingface/pytorch-image-models

模型用途

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('cs3sedarknet_l.c2ns_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(
    'cs3sedarknet_l.c2ns_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, 128, 128])
    #  torch.Size([1, 128, 64, 64])
    #  torch.Size([1, 256, 32, 32])
    #  torch.Size([1, 512, 16, 16])
    #  torch.Size([1, 1024, 8, 8])

    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(
    'cs3sedarknet_l.c2ns_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, 1024, 8, 8) shaped tensor

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

模型对比

在timm中探索数据集和运行时指标以了解该模型的详情 model results .

引用

@article{Wang2019CSPNetAN,
  title={CSPNet: A New Backbone that can Enhance Learning Capability of CNN},
  author={Chien-Yao Wang and Hong-Yuan Mark Liao and I-Hau Yeh and Yueh-Hua Wu and Ping-Yang Chen and Jun-Wei Hsieh},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  year={2019},
  pages={1571-1580}
}

@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{Redmon2018YOLOv3AI,
  title={YOLOv3: An Incremental Improvement},
  author={Joseph Redmon and Ali Farhadi},
  journal={ArXiv},
  year={2018},
  volume={abs/1804.02767}
}

@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}
}