# VoVNet

**Category:** CNN · **Code:**
[`timm/models/vovnet.py`](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/vovnet.py) ·
**Architectures:** 10 · **Pretrained weights:** 3 (3 with ImageNet-1k results)

VoVNet uses One-Shot Aggregation (OSA) modules, which concatenate the outputs of all layers of a block once at its end.
VoVNetV2 (from CenterMask) adds residual connections and effective SE (eSE) attention.

## Papers

* [An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object
  Detection](https://arxiv.org/abs/1904.09730)
* [CenterMask : Real-Time Anchor-Free Instance Segmentation](https://arxiv.org/abs/1911.06667)

## 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('ese_vovnet39b.ra_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)
```

## Model API support

| Feature | `VovNet` |
|---|---|
| [`features_only=True` feature maps](../model_api#intermediate-features) | ✓ (strides 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 (`VovNet` with `ese_vovnet19b_slim_dw` 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 |
|---|---:|---:|---:|---:|
| `eca_vovnet39b` | 22.60 | 7.09 | 224 | 0 |
| `ese_vovnet19b_dw` | 6.54 | 1.34 | 224 | 1 |
| `ese_vovnet19b_slim` | 3.17 | 1.69 | 224 | 0 |
| `ese_vovnet19b_slim_dw` | 1.90 | 0.40 | 224 | 0 |
| `ese_vovnet39b` | 24.57 | 7.09 | 224 | 1 |
| `ese_vovnet39b_evos` | 24.58 | 7.07 | 224 | 0 |
| `ese_vovnet57b` | 38.61 | 11.69 | 256 | 1 |
| `ese_vovnet99b` | 63.20 | 16.51 | 224 | 0 |
| `vovnet39a` | 22.60 | 7.09 | 224 | 0 |
| `vovnet57a` | 36.64 | 8.95 | 224 | 0 |

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 |
|---|---:|---:|---:|---:|---|---|
| `ese_vovnet19b_dw.ra_in1k` | 224 (test 288) | 77.788 | 288 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/ese_vovnet19b_dw.ra_in1k) |
| `ese_vovnet39b.ra_in1k` | 224 (test 288) | 80.378 | 288 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/ese_vovnet39b.ra_in1k) |
| `ese_vovnet57b.ra4_e3600_r256_in1k` | 256 (test 320) | 83.298 | 320 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/ese_vovnet57b.ra4_e3600_r256_in1k) |