File size: 2,012 Bytes
7d48e7e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: mit
library_name: libreyolo
pipeline_tag: object-detection
datasets:
  - detection-datasets/coco
tags:
  - object-detection
  - centernet
  - libreyolo
---

# LibreCenterNetdla34

CenterNet DLA-34 with deformable aggregation COCO detector, repackaged for LibreYOLO.

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreCenterNetdla34.pt")
results = model.predict("image.jpg")
```

## Source

Derived from [xingyizhou/CenterNet](https://github.com/xingyizhou/CenterNet)
at commit
[`4c50fd3a46bdf63dbf2082c5cbb3458d39579e6c`](https://github.com/xingyizhou/CenterNet/commit/4c50fd3a46bdf63dbf2082c5cbb3458d39579e6c).
Copyright (c) 2019 Xingyi Zhou. The source implementation is MIT licensed.

Official checkpoint: [ctdet_coco_dla_2x.pth](https://drive.google.com/file/d/18Q3fzzAsha_3Qid6mn4jcIFPeOGUaj1d/view)  
Official SHA-256: `43bf4cc2efe00e02c1ae8484035b062a35543872d276c7dcfeb4db3e64203e4f`  
Published COCO test-dev AP without test-time augmentation: 37.4.

## Modifications

The data-parallel `module.` prefix was removed and LibreYOLO v1 checkpoint
metadata was added. Learned tensors are unchanged. The native graph strictly
loads the official state dict and its `hm`, `wh`, and `reg` outputs are
bit-exact against the pinned implementation. LibreYOLO replaces the legacy
DCNv2 extension with torchvision deformable convolution.

See `weights/convert_centernet_weights.py` and
`docs/provenance/centernet.md` in the
[LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo).

## Benchmarks

Independent accuracy and speed benchmarks:
[visionanalysis.org/model/centernet-dla34](https://www.visionanalysis.org/model/centernet-dla34)

## License

The checkpoint publisher did not attach a standalone per-object license file.
This mirror applies the releasing project's MIT license on an **implied**, not
publisher-confirmed, basis. COCO annotations are CC BY 4.0; source images
retain their individual Flickr terms. See [`LICENSE`](./LICENSE) and
[`NOTICE`](./NOTICE).