crossvit_base_240.in1k模型卡片

一种CrossViT图像分类模型。由论文作者在ImageNet-1k上进行训练。

模型详细信息

• 模型类型：图像分类/特征骨干
• 模型统计数据：
  • 参数（百万）：105.0
  • GMACs：21.2
  • 激活数（百万）：36.3
  • 图像尺寸：240 x 240
• 论文：
  • CrossViT：用于图像分类的交叉注意力多尺度视觉变换器： https://arxiv.org/abs/2103.14899
• 数据集：ImageNet-1k
• 原始论文： https://github.com/IBM/CrossViT

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('crossvit_base_240.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)

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(
    'crossvit_base_240.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 (torch.Size([1, 401, 384]), torch.Size([1, 197, 768])) 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 中探索该模型的数据集和运行时指标。

引用

@inproceedings{
  chen2021crossvit,
  title={{CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification}},
  author={Chun-Fu (Richard) Chen and Quanfu Fan and Rameswar Panda},
  booktitle={International Conference on Computer Vision (ICCV)},
  year={2021}
}