# VGG

**Category:** CNN · **Code:**
[`timm/models/vgg.py`](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/vgg.py) ·
**Architectures:** 8 · **Pretrained weights:** 8 (8 with ImageNet-1k results)

VGG networks of stacked 3x3 convolutions, adapted from torchvision with `timm` functionality added.

## Papers

* [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://arxiv.org/abs/1409.1556)

## Usage

See the [Quickstart](../quickstart) for loading pretrained weights, image preprocessing and fine-tuning, and [Feature
Extraction](../feature_extraction) for using the models as backbones.

```py
>>> import timm
>>> model = timm.create_model('vgg11.tv_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)
```

## Model API support

| Feature | `VGG` |
|---|---|
| [`features_only=True` feature maps](../model_api#intermediate-features) | ✓ (strides 1, 2, 4, 8, 16, 32) |
| [`forward_intermediates()`](../model_api#intermediate-features) | ✗ |
| [`prune_intermediate_layers()`](../model_api#intermediate-features) | ✗ |
| [Any input size, no changes needed](../model_api#input-size) | ✓ |
| [`set_input_size()`](../model_api#input-size) | – |
| [`dynamic_img_size=True`](../model_api#input-size) | – |
| [`set_grad_checkpointing()`](../model_api#training-hooks) | ✗ |
| [Layer-wise LR decay (`group_matcher()`)](../model_api#training-hooks) | ✓ |
| [`reset_classifier()`](../model_api#classifier) | ✓ |

✓ supported, ✓ (square) works for square sizes only, ✗ not supported, – not needed (the model already accepts any input
size). Checked by running each feature on a randomly initialized model (`VGG` with `vgg11` at 224), other architectures
of the same class are expected to behave the same. See the [Model API](../model_api) page for details.

## Architectures

| Architecture | Params (M) | GMACs | Input size | Pretrained weights |
|---|---:|---:|---:|---:|
| `vgg11` | 132.86 | 7.61 | 224 | 1 |
| `vgg11_bn` | 132.87 | 7.62 | 224 | 1 |
| `vgg13` | 133.05 | 11.31 | 224 | 1 |
| `vgg13_bn` | 133.05 | 11.33 | 224 | 1 |
| `vgg16` | 138.36 | 15.47 | 224 | 1 |
| `vgg16_bn` | 138.37 | 15.50 | 224 | 1 |
| `vgg19` | 143.67 | 19.63 | 224 | 1 |
| `vgg19_bn` | 143.68 | 19.66 | 224 | 1 |

GMACs and parameter counts are for the default input size and classifier of each architecture, from the [benchmark
results](https://github.com/huggingface/pytorch-image-models/tree/main/results).

## Pretrained weights

Top-1 is ImageNet-1k validation accuracy from
[`results-imagenet.csv`](https://github.com/huggingface/pytorch-image-models/blob/main/results/results-imagenet.csv) at
the listed eval size, - if the weight has no ImageNet-1k result (e.g. pretraining or image-text weights). Classes is the
size of the pretrained classifier, none for weights without one.

| Weight | Input size | Top-1 (%) | Eval size | Classes | License | Hub |
|---|---:|---:|---:|---:|---|---|
| `vgg11.tv_in1k` | 224 | 69.050 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg11.tv_in1k) |
| `vgg11_bn.tv_in1k` | 224 | 70.374 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg11_bn.tv_in1k) |
| `vgg13.tv_in1k` | 224 | 69.950 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg13.tv_in1k) |
| `vgg13_bn.tv_in1k` | 224 | 71.560 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg13_bn.tv_in1k) |
| `vgg16.tv_in1k` | 224 | 71.594 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg16.tv_in1k) |
| `vgg16_bn.tv_in1k` | 224 | 73.354 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg16_bn.tv_in1k) |
| `vgg19.tv_in1k` | 224 | 72.400 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg19.tv_in1k) |
| `vgg19_bn.tv_in1k` | 224 | 74.230 | 224 | 1000 | bsd-3-clause | [timm](https://huggingface.co/timm/vgg19_bn.tv_in1k) |