许可协议: mit
库名称: timm
标签:

• 图像特征提取
• timm
• transformers
  数据集:
• imagenet-22k

模型卡片 focalnet_huge_fl4.ms_in22k

一个FocalNet图像分类模型。由论文作者在ImageNet-22k上预训练完成。

模型详情

• 模型类型: 图像分类/特征主干网络
• 模型统计:
  • 参数量(M): 686.5
  • GMACs运算量: 118.9
  • 激活值(M): 113.3
  • 图像尺寸: 224 x 224
• 相关论文:
  • 焦点调制网络: https://arxiv.org/abs/2203.11926
• 原始实现: https://github.com/microsoft/FocalNet
• 数据集: ImageNet-22k

模型使用

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('focalnet_huge_fl4.ms_in22k', pretrained=True)  
model = model.eval()  

# 获取模型特定的变换（归一化、调整尺寸）  
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))  # 将单张图像扩展为批次大小为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(  
    'focalnet_huge_fl4.ms_in22k',  
    pretrained=True,  
    features_only=True,  
)  
model = model.eval()  

# 获取模型特定的变换（归一化、调整尺寸）  
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))  # 将单张图像扩展为批次大小为1  

for o in output:  
    # 打印输出中每个特征图的形状  
    # 例如对于focalnet_base_srf:  
    #  torch.Size([1, 128, 56, 56])  
    #  torch.Size([1, 256, 28, 28])  
    #  torch.Size([1, 512, 14, 14])  
    #  torch.Size([1, 1024, 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(  
    'focalnet_huge_fl4.ms_in22k',  
    pretrained=True,  
    num_classes=0,  # 移除分类器nn.Linear  
)  
model = model.eval()  

# 获取模型特定的变换（归一化、调整尺寸）  
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))  # 输出为(batch_size, num_features)形状的张量  

# 或等效方式（无需设置num_classes=0）  

output = model.forward_features(transforms(img).unsqueeze(0))  
# 输出为非池化形式（即(batch_size, num_features, H, W)张量）  

output = model.forward_head(output, pre_logits=True)  
# 输出为(batch_size, num_features)张量  

模型比较

在timm的模型结果中探索该模型的数据集和运行时指标。

引用

@misc{yang2022focal,  
  title={Focal Modulation Networks},   
  author={Jianwei Yang and Chunyuan Li and Xiyang Dai and Jianfeng Gao},  
  journal={Advances in Neural Information Processing Systems (NeurIPS)},  
  year={2022}  
}  

@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}}  
}