tresnet_v2_l.miil_in21k 模型卡片

一种 TResNet 图像分类模型。由论文作者在 ImageNet-21K-P（"大规模 ImageNet-21K 预训练"，ImageNet-22k 的 11k 子集）上进行训练。

此模型的权重已经重新映射和修改，以适应标准的 BatchNorm 而不是 InplaceABN。近期构建 inplace_abn 非常困难，并且使用 memory_format=channels_last、torch.compile() 等效果更慢。

模型详细信息

• 模型类型: 图像分类 / 特征提取
• 模型统计信息:
  • 参数（百万）: 67.1
  • GMACs: 8.9
  • 激活数（百万）: 16.4
  • 图像尺寸: 224 x 224
• 论文:
  • TResNet: 高性能 GPU 专用架构: https://arxiv.org/abs/2003.13630
  • 大规模 ImageNet-21K 预训练: https://arxiv.org/abs/2104.10972
• 预训练数据集: ImageNet-21K-P
• 原始版本:
  • https://github.com/Alibaba-MIIL/TResNet
  • https://github.com/Alibaba-MIIL/ImageNet21K

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('tresnet_v2_l.miil_in21k', 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(
    'tresnet_v2_l.miil_in21k',
    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, 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(
    'tresnet_v2_l.miil_in21k',
    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

引用

@misc{ridnik2020tresnet,
    title={TResNet: High Performance GPU-Dedicated Architecture},
    author={Tal Ridnik and Hussam Lawen and Asaf Noy and Itamar Friedman},
    year={2020},
    eprint={2003.13630},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

@misc{ridnik2021imagenet21k,
  title={ImageNet-21K Pretraining for the Masses}, 
  author={Tal Ridnik and Emanuel Ben-Baruch and Asaf Noy and Lihi Zelnik-Manor},
  year={2021},
  eprint={2104.10972},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}