# DenseNet

**Category:** CNN · **Code:**
[`timm/models/densenet.py`](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/densenet.py) ·
**Architectures:** 6 · **Pretrained weights:** 6 (6 with ImageNet-1k results)

Densely connected convolutional networks, where each layer in a dense block takes the concatenated outputs of all
preceding layers of the block. Adapted from torchvision, `densenetblur` uses blur pool anti-aliased downsampling.

## Papers

* [Densely Connected Convolutional Networks](https://arxiv.org/abs/1608.06993)
* [Making Convolutional Networks Shift-Invariant Again](https://arxiv.org/abs/1904.11486)

## 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('densenet161.tv_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)
```

## Model API support

| Feature | `DenseNet` |
|---|---|
| [`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 (`DenseNet` with `densenet121` 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 |
|---|---:|---:|---:|---:|
| `densenet121` | 7.98 | 2.87 | 224 | 2 |
| `densenet161` | 28.68 | 7.79 | 224 | 1 |
| `densenet169` | 14.15 | 3.40 | 224 | 1 |
| `densenet201` | 20.01 | 4.34 | 224 | 1 |
| `densenet264d` | 72.74 | 13.57 | 224 | 0 |
| `densenetblur121d` | 8.00 | 3.11 | 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.

Weights are grouped by source: the paper or origin given in the pretrained config, else a Hub organization other than
`timm`, else the first part of the pretrained tag (`architecture.tag`). Sources with a single weight are listed under
other.

### `tv` tags

| Weight | Input size | Top-1 (%) | Eval size | Classes | License | Hub |
|---|---:|---:|---:|---:|---|---|
| `densenet121.tv_in1k` | 224 | 74.750 | 224 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenet121.tv_in1k) |
| `densenet161.tv_in1k` | 224 | 77.384 | 224 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenet161.tv_in1k) |
| `densenet169.tv_in1k` | 224 | 75.922 | 224 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenet169.tv_in1k) |
| `densenet201.tv_in1k` | 224 | 77.284 | 224 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenet201.tv_in1k) |

### `ra` tags

| Weight | Input size | Top-1 (%) | Eval size | Classes | License | Hub |
|---|---:|---:|---:|---:|---|---|
| `densenet121.ra_in1k` | 224 (test 288) | 76.486 | 288 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenet121.ra_in1k) |
| `densenetblur121d.ra_in1k` | 224 (test 288) | 77.320 | 288 | 1000 | apache-2.0 | [timm](https://huggingface.co/timm/densenetblur121d.ra_in1k) |