docs: add NOTICE attribution per Apache-2.0 s4(d) (code AND weights)
Browse files
NOTICE
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
RTMDet-tiny — acaua mirror (pure-PyTorch port)
|
| 2 |
+
==============================================
|
| 3 |
+
|
| 4 |
+
This product includes:
|
| 5 |
+
|
| 6 |
+
1. PORTED SOURCE CODE: a pure-PyTorch port of the RTMDet architecture
|
| 7 |
+
(located in the acaua repository at src/acaua/adapters/rtmdet/) is a
|
| 8 |
+
derivative work of:
|
| 9 |
+
|
| 10 |
+
- open-mmlab/mmdetection
|
| 11 |
+
https://github.com/open-mmlab/mmdetection @ commit cfd5d3a985b0249de009b67d04f37263e11cdf3d
|
| 12 |
+
Apache-2.0
|
| 13 |
+
Files derived from:
|
| 14 |
+
mmdet/models/backbones/cspnext.py
|
| 15 |
+
mmdet/models/necks/cspnext_pafpn.py
|
| 16 |
+
mmdet/models/dense_heads/rtmdet_head.py
|
| 17 |
+
mmdet/models/layers/csp_layer.py
|
| 18 |
+
mmdet/models/layers/se_layer.py (ChannelAttention)
|
| 19 |
+
mmdet/models/backbones/csp_darknet.py (SPPBottleneck)
|
| 20 |
+
mmdet/models/task_modules/prior_generators/point_generator.py
|
| 21 |
+
mmdet/structures/bbox/transforms.py (distance2bbox)
|
| 22 |
+
Plus support utilities ported from open-mmlab/mmcv:
|
| 23 |
+
mmcv/cnn/bricks/conv_module.py (ConvModule)
|
| 24 |
+
mmcv/cnn/bricks/depthwise_separable_conv_module.py (DepthwiseSeparableConvModule)
|
| 25 |
+
|
| 26 |
+
2. CONVERTED WEIGHTS: the model.safetensors file in this mirror is a
|
| 27 |
+
key-remapped conversion of the upstream pretrained checkpoint:
|
| 28 |
+
|
| 29 |
+
- upstream URL: https://download.openmmlab.com/mmdetection/v3.0/rtmdet/rtmdet_tiny_8xb32-300e_coco/rtmdet_tiny_8xb32-300e_coco_20220902_112414-78e30dcc.pth
|
| 30 |
+
- upstream SHA256: 78e30dcce0c6f594eaff0d6977b84b4103688b4aff0ad1aa16008a8cc854a7fb
|
| 31 |
+
- upstream paper: Lyu et al., "RTMDet: An Empirical Study of Designing
|
| 32 |
+
Real-Time Object Detectors", arXiv:2212.07784, 2022.
|
| 33 |
+
|
| 34 |
+
Conversion was performed by scripts/convert_rtmdet.py in the acaua
|
| 35 |
+
repository. The conversion is lossless (no quantization or pruning)
|
| 36 |
+
and every rename is eager — load_state_dict(strict=True) is the
|
| 37 |
+
correctness gate.
|
| 38 |
+
|
| 39 |
+
Mirrored on 2026-04-20 by CondadosAI.
|
| 40 |
+
|
| 41 |
+
License
|
| 42 |
+
-------
|
| 43 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 44 |
+
you may not use this file except in compliance with the License.
|
| 45 |
+
You may obtain a copy of the License at
|
| 46 |
+
|
| 47 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 48 |
+
|
| 49 |
+
Unless required by applicable law or agreed to in writing, software
|
| 50 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 51 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 52 |
+
See the License for the specific language governing permissions and
|
| 53 |
+
limitations under the License.
|