# SENet (legacy)

**Category:** CNN · **Code:**
[`timm/models/senet.py`](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/senet.py) ·
**Architectures:** 9 · **Pretrained weights:** 9 (9 with ImageNet-1k results)

SENet, SE-ResNet and SE-ResNeXt from Cadene's pretrained models (`legacy_*`). The module is deprecated in favour of the
SE variants of the `ResNet` class.

## Papers

* [Squeeze-and-Excitation Networks](https://arxiv.org/abs/1709.01507)

## 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('legacy_seresnext50_32x4d.in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)
```

## Model API support

| Feature | `SENet` |
|---|---|
| [`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 (`SENet` with `legacy_seresnet18` 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 |
|---|---:|---:|---:|---:|
| `legacy_senet154` | 115.09 | 20.77 | 224 | 1 |
| `legacy_seresnet18` | 11.78 | 1.82 | 224 | 1 |
| `legacy_seresnet34` | 21.96 | 3.67 | 224 | 1 |
| `legacy_seresnet50` | 28.09 | 3.88 | 224 | 1 |
| `legacy_seresnet101` | 49.33 | 7.61 | 224 | 1 |
| `legacy_seresnet152` | 66.82 | 11.33 | 224 | 1 |
| `legacy_seresnext26_32x4d` | 16.79 | 2.49 | 224 | 1 |
| `legacy_seresnext50_32x4d` | 27.56 | 4.26 | 224 | 1 |
| `legacy_seresnext101_32x4d` | 48.96 | 8.02 | 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 |
|---|---:|---:|---:|---:|---|---|
| `legacy_senet154.in1k` | 224 | 81.320 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/legacy_senet154-e9eb9fe6.pth) |
| `legacy_seresnet18.in1k` | 224 | 71.742 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet18-4bb0ce65.pth) |
| `legacy_seresnet34.in1k` | 224 | 74.788 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet34-a4004e63.pth) |
| `legacy_seresnet50.in1k` | 224 | 77.650 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-cadene/se_resnet50-ce0d4300.pth) |
| `legacy_seresnet101.in1k` | 224 | 78.370 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-cadene/se_resnet101-7e38fcc6.pth) |
| `legacy_seresnet152.in1k` | 224 | 78.662 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-cadene/se_resnet152-d17c99b7.pth) |
| `legacy_seresnext26_32x4d.in1k` | 224 | 77.100 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26_32x4d-65ebdb501.pth) |
| `legacy_seresnext50_32x4d.in1k` | 224 | 79.080 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/legacy_se_resnext50_32x4d-f3651bad.pth) |
| `legacy_seresnext101_32x4d.in1k` | 224 | 80.274 | 224 | 1000 | apache-2.0 | [original](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/legacy_se_resnext101_32x4d-37725eac.pth) |