File size: 1,892 Bytes
c22e381
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a7d577e
c22e381
 
 
 
 
a7d577e
c22e381
 
 
a7d577e
c22e381
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
---
library_name: lucid
license: bsd-3-clause
tags:
  - image-classification
  - convnext
  - lucid
datasets:
  - imagenet-1k
pipeline_tag: image-classification
model-index:
  - name: convnext-tiny
    results:
      - task: { type: image-classification }
        dataset: { name: ImageNet-1K, type: imagenet-1k }
        metrics:
          - { type: acc@1, value: 82.52 }
          - { type: acc@5, value: 96.146 }
---

# ConvNeXt-Tiny

> Liu et al., 2022 — *A ConvNet for the 2020s* (arXiv:2201.03545)

[Lucid](https://github.com/ChanLumerico/lucid) port of `torchvision/ConvNeXt_Tiny_Weights.IMAGENET1K_V1`,
converted to Lucid-native safetensors.

## Available weights

| Tag | acc@1 | acc@5 | Params | GFLOPs | Size | Source |
|---|---|---|---|---|---|---|
| `IMAGENET1K_V1` *(default)* | 82.52 | 96.146 | 28.6M | 4.456 | 109.07 MB | torchvision |

## Usage

```python
import lucid.models as models
from lucid.models.weights import ConvNeXtTinyWeights

# default tag
model = models.convnext_tiny_cls(pretrained=True)

# explicit tag (enum or string)
model = models.convnext_tiny_cls(weights=ConvNeXtTinyWeights.IMAGENET1K_V1)
model = models.convnext_tiny_cls(pretrained="IMAGENET1K_V1")

# preprocessing travels with the weights
weights = ConvNeXtTinyWeights.IMAGENET1K_V1
preprocess = weights.transforms()
logits = model(preprocess(image)[None]).logits
```

## Conversion

Converted from `torchvision/ConvNeXt_Tiny_Weights.IMAGENET1K_V1` via
`python -m tools.convert_weights convnext_tiny --tag IMAGENET1K_V1`.
Key mapping + numerical parity verified against the source.

## License

`bsd-3-clause` — inherited from the original weights.

## Citation

```
@inproceedings{liu2022convnet,
  title={A ConvNet for the 2020s},
  author={Liu, Zhuang and Mao, Hanzi and Wu, Chao-Yuan and Feichtenhofer, Christoph and Darrell, Trevor and Xie, Saining},
  booktitle={CVPR}, year={2022}
}
```