You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
pytorch-image-models/docs/models/swsl-resnext.md

278 lines
8.7 KiB

# SWSL ResNeXt
A **ResNeXt** repeats a [building block](https://paperswithcode.com/method/resnext-block) that aggregates a set of transformations with the same topology. Compared to a [ResNet](https://paperswithcode.com/method/resnet), it exposes a new dimension, *cardinality* (the size of the set of transformations) $C$, as an essential factor in addition to the dimensions of depth and width.
The models in this collection utilise semi-weakly supervised learning to improve the performance of the model. The approach brings important gains to standard architectures for image, video and fine-grained classification.
Please note the CC-BY-NC 4.0 license on theses weights, non-commercial use only.
## How do I use this model on an image?
To load a pretrained model:
```python
import timm
model = timm.create_model('swsl_resnext101_32x16d', pretrained=True)
model.eval()
```
To load and preprocess the image:
```python
import urllib
from PIL import Image
from timm.data import resolve_data_config
from timm.data.transforms_factory import create_transform
config = resolve_data_config({}, model=model)
transform = create_transform(**config)
url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
urllib.request.urlretrieve(url, filename)
img = Image.open(filename).convert('RGB')
tensor = transform(img).unsqueeze(0) # transform and add batch dimension
```
To get the model predictions:
```python
import torch
with torch.no_grad():
out = model(tensor)
probabilities = torch.nn.functional.softmax(out[0], dim=0)
print(probabilities.shape)
# prints: torch.Size([1000])
```
To get the top-5 predictions class names:
```python
# Get imagenet class mappings
url, filename = ("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt", "imagenet_classes.txt")
urllib.request.urlretrieve(url, filename)
with open("imagenet_classes.txt", "r") as f:
categories = [s.strip() for s in f.readlines()]
# Print top categories per image
top5_prob, top5_catid = torch.topk(probabilities, 5)
for i in range(top5_prob.size(0)):
print(categories[top5_catid[i]], top5_prob[i].item())
# prints class names and probabilities like:
# [('Samoyed', 0.6425196528434753), ('Pomeranian', 0.04062102362513542), ('keeshond', 0.03186424449086189), ('white wolf', 0.01739676296710968), ('Eskimo dog', 0.011717947199940681)]
```
Replace the model name with the variant you want to use, e.g. `swsl_resnext101_32x16d`. You can find the IDs in the model summaries at the top of this page.
To extract image features with this model, follow the [timm feature extraction examples](https://rwightman.github.io/pytorch-image-models/feature_extraction/), just change the name of the model you want to use.
## How do I finetune this model?
You can finetune any of the pre-trained models just by changing the classifier (the last layer).
```python
model = timm.create_model('swsl_resnext101_32x16d', pretrained=True).reset_classifier(NUM_FINETUNE_CLASSES)
```
To finetune on your own dataset, you have to write a training loop or adapt [timm's training
script](https://github.com/rwightman/pytorch-image-models/blob/master/train.py) to use your dataset.
## How do I train this model?
You can follow the [timm recipe scripts](https://rwightman.github.io/pytorch-image-models/scripts/) for training a new model afresh.
## Citation
```BibTeX
@article{DBLP:journals/corr/abs-1905-00546,
author = {I. Zeki Yalniz and
Herv{\'{e}} J{\'{e}}gou and
Kan Chen and
Manohar Paluri and
Dhruv Mahajan},
title = {Billion-scale semi-supervised learning for image classification},
journal = {CoRR},
volume = {abs/1905.00546},
year = {2019},
url = {http://arxiv.org/abs/1905.00546},
archivePrefix = {arXiv},
eprint = {1905.00546},
timestamp = {Mon, 28 Sep 2020 08:19:37 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-1905-00546.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
<!--
Type: model-index
Collections:
- Name: SWSL ResNext
Paper:
Title: Billion-scale semi-supervised learning for image classification
URL: https://paperswithcode.com/paper/billion-scale-semi-supervised-learning-for
Models:
- Name: swsl_resnext101_32x16d
In Collection: SWSL ResNext
Metadata:
FLOPs: 46623691776
Parameters: 194030000
File Size: 777518664
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- IG-1B-Targeted
- ImageNet
Training Resources: 64x GPUs
ID: swsl_resnext101_32x16d
LR: 0.0015
Epochs: 30
Layers: 101
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L1009
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x16-f3559a9c.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 83.34%
Top 5 Accuracy: 96.84%
- Name: swsl_resnext101_32x4d
In Collection: SWSL ResNext
Metadata:
FLOPs: 10298145792
Parameters: 44180000
File Size: 177341913
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- IG-1B-Targeted
- ImageNet
Training Resources: 64x GPUs
ID: swsl_resnext101_32x4d
LR: 0.0015
Epochs: 30
Layers: 101
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L987
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x4-3f87e46b.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 83.22%
Top 5 Accuracy: 96.77%
- Name: swsl_resnext101_32x8d
In Collection: SWSL ResNext
Metadata:
FLOPs: 21180417024
Parameters: 88790000
File Size: 356056638
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- IG-1B-Targeted
- ImageNet
Training Resources: 64x GPUs
ID: swsl_resnext101_32x8d
LR: 0.0015
Epochs: 30
Layers: 101
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L998
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x8-b4712904.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 84.27%
Top 5 Accuracy: 97.17%
- Name: swsl_resnext50_32x4d
In Collection: SWSL ResNext
Metadata:
FLOPs: 5472648192
Parameters: 25030000
File Size: 100428550
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- IG-1B-Targeted
- ImageNet
Training Resources: 64x GPUs
ID: swsl_resnext50_32x4d
LR: 0.0015
Epochs: 30
Layers: 50
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L976
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext50_32x4-72679e44.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 82.17%
Top 5 Accuracy: 96.23%
-->