File size: 3,401 Bytes
829aac4 fcff0b3 829aac4 e3de7bf 829aac4 e3de7bf 829aac4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 |
---
datasets:
- imagenet-1k
library_name: timm
tags:
- image-classification
- timm
- rdnet
license: bsd-3-clause
---
# Model card for rdnet_base.nv_in1k
A RDNet image classification model. Trained on ImageNet-1k, original torchvision weights.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Imagenet-1k validation top-1 accuracy: 84.4%
- Params (M): 87
- GMACs: 15.4
- Image size: 224 x 224
- **Papers:**
- DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs: https://arxiv.org/abs/2403.19588
- **Dataset:** ImageNet-1k
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
import torch
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('rdnet_base.nv_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)
```
### Feature Map Extraction
```python
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(
'rdnet_base.nv_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, 408, 56, 56])
# torch.Size([1, 584, 28, 28])
# torch.Size([1, 1000, 14, 14])
# torch.Size([1, 1760, 7, 7])
print(o.shape)
```
### Image Embeddings
```python
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(
'rdnet_base.nv_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, 1760, 7, 7) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
### Citation
```
@misc{kim2024densenets,
title={DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs},
author={Donghyun Kim and Byeongho Heo and Dongyoon Han},
year={2024},
eprint={2403.19588},
archivePrefix={arXiv},
}
``` |