mvitv2_large_cls.fb_inw21k 模型卡片

一种 MViT-v2（多尺度 ViT）图像分类模型。在 ImageNet-22k（Winter21 变种）上进行了预训练，并由论文作者在 ImageNet-1k 上进行了微调。该模型的分类器布局未公开，并且与预期的词典排序的 synset 顺序不匹配。

模型细节

• 模型类型：图像分类/特征骨干
• 模型统计数据：
  • 参数数目（M）：234.6
  • GMACs：42.2
  • 激活数目（M）：111.7
  • 图像尺寸：224 x 224
• 论文：
  • MViTv2：改进的用于分类和检测的多尺度视觉转换器： https://arxiv.org/abs/2112.01526
• 数据集：ImageNet-1k
• 预训练数据集：ImageNet-22k
• 原始论文： https://github.com/facebookresearch/mvit

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('mvitv2_large_cls.fb_inw21k', 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(
    'mvitv2_large_cls.fb_inw21k',
    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, 50, 1152) shaped tensor

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