[Docs] add readthedocs algorithm pages and fix some displaying error (#599)
* update algo readme * add script to generate algo pages in model zoo * fix lint * update requirements * fix lint * update title * fix readthedocs docstring displaying error * update gitignore * fix displaying * fix lintpull/616/head
Yixiao Fang
GitHub
parent
73cd764b5f
commit
b74c3bb25c
|
|
@ -64,8 +64,14 @@ instance/
|
|||
# Scrapy stuff:
|
||||
.scrapy
|
||||
|
||||
# Sphinx documentation
|
||||
docs/_build/
|
||||
# Sphinx documentation (readthedocs)
|
||||
src/
|
||||
docs/en/_build/
|
||||
docs/en/model_zoo_statistics.md
|
||||
docs/en/papers/
|
||||
docs/zh_cn/_build/
|
||||
docs/zh_cn/model_zoo_statistics.md
|
||||
docs/zh_cn/papers/
|
||||
|
||||
# PyBuilder
|
||||
target/
|
||||
|
|
@ -125,9 +131,3 @@ INFO
|
|||
|
||||
# Pytorch
|
||||
*.pth
|
||||
|
||||
|
||||
# readthedocs
|
||||
docs/zh_cn/_build
|
||||
src/
|
||||
docs/en/_build
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Self-supervised learning (SSL) is rapidly closing the gap with supervised method
|
|||
<img src="https://user-images.githubusercontent.com/36138628/163914714-082de804-0b5f-4024-94f9-880e6ef334fa.png" width="800" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -24,21 +22,52 @@ The classification benchmarks includes 1 downstream task datasets, **ImageNet**.
|
|||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-90e.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-90e.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_8xb32-steplr-100e_in1k.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [barlowtwins_resnet50_8xb256-coslr-300e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k.py) | 15.51 | 33.98 | 45.96 | 61.90 | 71.01 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [barlowtwins_resnet50_8xb256-coslr-300e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/arlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k.py) | 15.51 | 33.98 | 45.96 | 61.90 | 71.01 | 71.66 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>BarlowTwins</td>
|
||||
<td>ResNet50</td>
|
||||
<td>300</td>
|
||||
<td>2048</td>
|
||||
<td>71.8</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k/barlowtwins_resnet50_8xb256-coslr-300e_in1k_20220825-57307488.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k/barlowtwins_resnet50_8xb256-coslr-300e_in1k_20220726_033718.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-coslr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k/resnet50_linear-8xb32-coslr-100e_in1k/resnet50_linear-8xb32-coslr-100e_in1k_20220825-52fde35f.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k/resnet50_linear-8xb32-coslr-100e_in1k/resnet50_linear-8xb32-coslr-100e_in1k_20220730_093018.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [barlowtwins_resnet50_8xb256-coslr-300e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/arlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k.py) | 63.6 | 63.8 | 62.7 | 61.9 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [barlowtwins_resnet50_8xb256-coslr-300e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/barlowtwins/barlowtwins_resnet50_8xb256-coslr-300e_in1k.py) | 63.6 | 63.8 | 62.7 | 61.9 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149720208-5ffbee78-1437-44c7-9ddb-b8caab60d2c3.png" width="800" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,38 +26,69 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | feature5 | 86.31 | 45.37 | 56.83 | 68.47 | 74.12 | 78.30 | 81.53 | 83.56 | 84.73 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | feature5 | 86.31 | 45.37 | 56.83 | 68.47 | 74.12 | 78.30 | 81.53 | 83.56 | 84.73 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb512-coslr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 15.16 | 35.26 | 47.77 | 63.10 | 71.21 |
|
||||
| [resnet50_16xb256-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_16xb256-coslr-200e_in1k.py) | 15.41 | 35.15 | 47.77 | 62.59 | 71.85 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 15.16 | 35.26 | 47.77 | 63.10 | 71.21 | 71.72 |
|
||||
| [resnet50_16xb256-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_16xb256-coslr-200e_in1k.py) | 15.41 | 35.15 | 47.77 | 62.59 | 71.85 | 71.88 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>BYOL</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>4096</td>
|
||||
<td>71.8</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/byol/byol_resnet50_16xb256-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/byol/byol_resnet50_16xb256-coslr-200e_in1k/byol_resnet50_16xb256-coslr-200e_in1k_20220825-de817331.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/byol/byol_resnet50_16xb256-coslr-200e_in1k/byol_resnet50_16xb256-coslr-200e_in1k_20220721_150515.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/byol/byol_resnet50_16xb256-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-7596c6f5.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/byol/byol_resnet50_16xb256-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220724_130251.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 21.25 | 36.55 | 43.66 | 50.74 | 53.82 |
|
||||
| [resnet50_8xb32-accum16-coslr-300e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-300e_in1k.py) | 21.18 | 36.68 | 43.42 | 51.04 | 54.06 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 21.25 | 36.55 | 43.66 | 50.74 | 53.82 |
|
||||
| [resnet50_8xb32-accum16-coslr-300e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-300e_in1k.py) | 21.18 | 36.68 | 43.42 | 51.04 | 54.06 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 63.9 | 64.2 | 62.9 | 61.9 |
|
||||
| [resnet50_8xb32-accum16-coslr-300e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-300e_in1k.py) | 66.1 | 66.3 | 65.2 | 64.4 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 63.9 | 64.2 | 62.9 | 61.9 |
|
||||
| [resnet50_8xb32-accum16-coslr-300e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-300e_in1k.py) | 66.1 | 66.3 | 65.2 | 64.4 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -67,19 +96,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 80.35 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 80.35 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 40.9 | 61.0 | 44.6 | 36.8 | 58.1 | 39.5 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 40.9 | 61.0 | 44.6 | 36.8 | 58.1 | 39.5 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -87,11 +116,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 67.16 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-accum16-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/byol/byol_resnet50_8xb32-accum16-coslr-200e_in1k.py) | 67.16 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Clustering is a class of unsupervised learning methods that has been extensively
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149720586-5bfd213e-0638-47fc-b48a-a16689190e17.png" width="700" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,27 +26,27 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | feature5 | 74.26 | 29.37 | 37.99 | 45.85 | 55.57 | 62.48 | 66.15 | 70.00 | 71.37 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | feature5 | 74.26 | 29.37 | 37.99 | 45.85 | 55.57 | 62.48 | 66.15 | 70.00 | 71.37 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | 12.78 | 30.81 | 43.88 | 57.71 | 51.68 | 46.92 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | 12.78 | 30.81 | 43.88 | 57.71 | 51.68 | 46.92 |
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | 18.80 | 33.93 | 41.44 | 47.22 | 42.61 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [sobel_resnet50_8xb64-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/deepcluster/deepcluster-sobel_resnet50_8xb64-steplr-200e_in1k.py) | 18.80 | 33.93 | 41.44 | 47.22 | 42.61 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,15 +12,13 @@ To date, most existing self-supervised learning methods are designed and optimiz
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149721111-bab03a6d-a30d-418e-b338-43c3689cfc65.png" width="900" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
### Classification
|
||||
|
||||
The classification benchmarks includes 4 downstream task datasets, **VOC**, **ImageNet**, **iNaturalist2018** and **Places205**. If not specified, the results are Top-1 (%).
|
||||
The classification benchmarks includes 4 downstream task datasets, **VOC**, **ImageNet**, **iNaturalist2018** and **Places205**. If not specified, the results are Top-1 (%).
|
||||
|
||||
#### VOC SVM / Low-shot SVM
|
||||
|
||||
|
|
@ -28,35 +26,66 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ---- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 82.5 | 42.68 | 50.64 | 61.74 | 68.17 | 72.99 | 76.07 | 79.19 | 80.55 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ---- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 82.5 | 42.68 | 50.64 | 61.74 | 68.17 | 72.99 | 76.07 | 79.19 | 80.55 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 15.86 | 35.47 | 49.46 | 64.06 | 62.95 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 15.86 | 35.47 | 49.46 | 64.06 | 62.95 | 63.34 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>DenseCL</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>63.5</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/densecl/densecl_resnet50_8xb32-coslr-200e_in1k/densecl_resnet50_8xb32-coslr-200e_in1k_20220825-3078723b.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/densecl/densecl_resnet50_8xb32-coslr-200e_in1k/densecl_resnet50_8xb32-coslr-200e_in1k_20220727_221415.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/densecl/densecl_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220825-f0f0a579.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/densecl/densecl_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220730_091650.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 21.32 | 36.20 | 43.97 | 51.04 | 50.45 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 21.32 | 36.20 | 43.97 | 51.04 | 50.45 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 48.2 | 48.5 | 46.8 | 45.6 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 48.2 | 48.5 | 46.8 | 45.6 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -64,19 +93,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 82.14 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 82.14 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | | | | | | |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | | | | | | |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -84,11 +113,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 69.47 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/densecl/densecl_resnet50_8xb32-coslr-200e_in1k.py) | 69.47 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -13,3 +13,12 @@ Collections:
|
|||
URL: https://arxiv.org/abs/1911.05722
|
||||
Title: "Momentum Contrast for Unsupervised Visual Representation Learning"
|
||||
README: configs/selfsup/mocov1/README.md
|
||||
|
||||
Models:
|
||||
- Name: moco_resnet50_8xb32-coslr-200e_in1k
|
||||
In Collection: MoCo
|
||||
Metadata:
|
||||
Epochs: 200
|
||||
Batch Size: 256
|
||||
Results: null
|
||||
Config: configs/selfsup/mocov1/mocov1_resnet50_8xb32-coslr-200e_in1k.py
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Contrastive unsupervised learning has recently shown encouraging progress, e.g.,
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149720067-b65e5736-d425-45b3-93ed-6f2427fc6217.png" width="500" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,35 +26,66 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 84.04 | 43.14 | 53.29 | 65.34 | 71.03 | 75.42 | 78.48 | 80.88 | 82.23 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 84.04 | 43.14 | 53.29 | 65.34 | 71.03 | 75.42 | 78.48 | 80.88 | 82.23 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 15.96 | 34.22 | 45.78 | 61.11 | 66.24 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 15.96 | 34.22 | 45.78 | 61.11 | 66.24 | 67.58 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>MoCo v2</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>67.5</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k/mocov2_resnet50_8xb32-coslr-200e_in1k_20220825-b6d23c86.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k/mocov2_resnet50_8xb32-coslr-200e_in1k_20220721_215805.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220825-994c4128.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220724_172046.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 20.92 | 35.72 | 42.62 | 49.79 | 52.25 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 20.92 | 35.72 | 42.62 | 49.79 | 52.25 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 55.6 | 55.7 | 53.7 | 52.5 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 55.6 | 55.7 | 53.7 | 52.5 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -64,19 +93,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 81.06 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 81.06 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 40.2 | 59.7 | 44.2 | 36.1 | 56.7 | 38.8 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 40.2 | 59.7 | 44.2 | 36.1 | 56.7 | 38.8 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -84,11 +113,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 67.55 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/mocov2/mocov2_resnet50_8xb32-coslr-200e_in1k.py) | 67.55 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -16,7 +16,7 @@ Collections:
|
|||
|
||||
Models:
|
||||
- Name: mocov2_resnet50_8xb32-coslr-200e_in1k
|
||||
In Collection: MoCo
|
||||
In Collection: MoCov2
|
||||
Metadata:
|
||||
Epochs: 200
|
||||
Batch Size: 256
|
||||
|
|
|
|||
|
|
@ -18,8 +18,6 @@ Our method is also remarkable for consistently improving test performance with m
|
|||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
### Classification
|
||||
|
|
@ -32,35 +30,66 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | feature5 | 76.75 | 26.96 | 35.37 | 44.48 | 53.89 | 60.39 | 66.41 | 71.48 | 73.39 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | feature5 | 76.75 | 26.96 | 35.37 | 44.48 | 53.89 | 60.39 | 66.41 | 71.48 | 73.39 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 14.68 | 31.98 | 42.85 | 56.95 | 58.41 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 14.68 | 31.98 | 42.85 | 56.95 | 58.41 | 57.97 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>NPID</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>58.3</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/npid/npid_resnet50_8xb32-steplr-200e_in1k/npid_resnet50_8xb32-steplr-200e_in1k_20220825-a67c5440.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/npid/npid_resnet50_8xb32-steplr-200e_in1k/npid_resnet50_8xb32-steplr-200e_in1k_20220725_161221.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/npid/npid_resnet50_8xb32-steplr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220825-661b736e.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/npid/npid_resnet50_8xb32-steplr-200e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220728_150535.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 19.98 | 34.86 | 41.59 | 48.43 | 48.71 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 19.98 | 34.86 | 41.59 | 48.43 | 48.71 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 42.9 | 44.0 | 43.2 | 42.2 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 42.9 | 44.0 | 43.2 | 42.2 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -68,19 +97,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 79.52 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 79.52 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 38.5 | 57.7 | 42.0 | 34.6 | 54.8 | 37.1 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 38.5 | 57.7 | 42.0 | 34.6 | 54.8 | 37.1 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -88,11 +117,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 65.45 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-steplr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/npid/npid_resnet50_8xb32-steplr-200e_in1k.py) | 65.45 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Joint clustering and feature learning methods have shown remarkable performance
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149722645-8da8e5b2-8846-4554-aa3e-727d286b85cd.png" width="700" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,35 +26,35 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | feature5 | 78.42 | 32.42 | 40.27 | 49.95 | 59.96 | 65.71 | 69.99 | 73.64 | 75.13 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | feature5 | 78.42 | 32.42 | 40.27 | 49.95 | 59.96 | 65.71 | 69.99 | 73.64 | 75.13 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 14.76 | 31.82 | 42.44 | 55.76 | 57.70 | 53.42 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 14.76 | 31.82 | 42.44 | 55.76 | 57.70 | 53.42 |
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 19.28 | 34.09 | 40.90 | 47.04 | 48.35 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 19.28 | 34.09 | 40.90 | 47.04 | 48.35 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 38.5 | 39.1 | 37.8 | 36.9 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-440e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/odc/odc_resnet50_8xb64-steplr-440e_in1k.py) | 38.5 | 39.1 | 37.8 | 36.9 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ This work explores the use of spatial context as a source of free and plentiful
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149723222-76bc89e8-98bf-4ed7-b179-dfe5bc6336ba.png" width="400" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,35 +26,66 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | feature4 | 65.52 | 20.36 | 23.12 | 30.66 | 37.02 | 42.55 | 50.00 | 55.58 | 59.28 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | feature4 | 65.52 | 20.36 | 23.12 | 30.66 | 37.02 | 42.55 | 50.00 | 55.58 | 59.28 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 15.11 | 30.47 | 42.83 | 51.20 | 40.96 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 15.11 | 30.47 | 42.83 | 51.20 | 40.96 | 39.65 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>Relative-Loc</td>
|
||||
<td>ResNet50</td>
|
||||
<td>70</td>
|
||||
<td>512</td>
|
||||
<td>40.4</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k/relative-loc_resnet50_8xb64-steplr-70e_in1k_20220825-daae1b41.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k/relative-loc_resnet50_8xb64-steplr-70e_in1k_20220802_223045.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220825-c2a0b188.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220804_194226.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 20.69 | 34.72 | 43.01 | 45.97 | 41.96 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 20.69 | 34.72 | 43.01 | 45.97 | 41.96 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 14.5 | 15.0 | 15.0 | 14.2 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 14.5 | 15.0 | 15.0 | 14.2 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -64,19 +93,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 79.70 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 79.70 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 37.5 | 56.2 | 41.3 | 33.7 | 53.3 | 36.1 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 37.5 | 56.2 | 41.3 | 33.7 | 53.3 | 36.1 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -84,11 +113,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 63.49 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb64-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/relative_loc/relative-loc_resnet50_8xb64-steplr-70e_in1k.py) | 63.49 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Over the last years, deep convolutional neural networks (ConvNets) have transfor
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149723477-8f63e237-362e-4962-b405-9bab0f579808.png" width="700" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,35 +26,66 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | feature4 | 67.70 | 20.60 | 24.35 | 31.41 | 39.17 | 46.56 | 53.37 | 59.14 | 62.42 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | feature4 | 67.70 | 20.60 | 24.35 | 31.41 | 39.17 | 46.56 | 53.37 | 59.14 | 62.42 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-steplr-100e_in1k.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 12.15 | 31.99 | 44.57 | 54.20 | 45.94 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 12.15 | 31.99 | 44.57 | 54.20 | 45.94 | 48.12 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>Rotation-Pred</td>
|
||||
<td>ResNet50</td>
|
||||
<td>70</td>
|
||||
<td>128</td>
|
||||
<td>47.0</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k/rotation-pred_resnet50_8xb16-steplr-70e_in1k_20220825-a8bf5f69.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k/rotation-pred_resnet50_8xb16-steplr-70e_in1k_20220805_113136.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-steplr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220825-7c6edcb3.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k/resnet50_linear-8xb32-steplr-100e_in1k/resnet50_linear-8xb32-steplr-100e_in1k_20220808_143921.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 18.94 | 34.72 | 44.53 | 46.30 | 44.12 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 18.94 | 34.72 | 44.53 | 46.30 | 44.12 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 11.0 | 11.9 | 12.6 | 12.4 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 11.0 | 11.9 | 12.6 | 12.4 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -64,19 +93,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 79.67 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 79.67 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 37.9 | 56.5 | 41.5 | 34.2 | 53.9 | 36.7 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 37.9 | 56.5 | 41.5 | 34.2 | 53.9 | 36.7 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -84,11 +113,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 64.31 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb16-steplr-70e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/rotation_pred/rotation-pred_resnet50_8xb16-steplr-70e_in1k.py) | 64.31 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -14,8 +14,6 @@ This paper presents SimCLR: a simple framework for contrastive learning of visua
|
|||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
### Classification
|
||||
|
|
@ -28,36 +26,87 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ---- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 79.98 | 35.02 | 42.79 | 54.87 | 61.91 | 67.38 | 71.88 | 75.56 | 77.4 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ---- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 79.98 | 35.02 | 42.79 | 54.87 | 61.91 | 67.38 | 71.88 | 75.56 | 77.4 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb512-coslr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 16.29 | 31.11 | 39.99 | 55.06 | 62.91 |
|
||||
| [resnet50_16xb256-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_16xb256-coslr-200e_in1k.py) | 15.44 | 31.47 | 41.83 | 59.44 | 66.41 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ---------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 16.29 | 31.11 | 39.99 | 55.06 | 62.91 | 62.56 |
|
||||
| [resnet50_16xb256-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_16xb256-coslr-200e_in1k.py) | 15.44 | 31.47 | 41.83 | 59.44 | 66.41 | 66.66 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td rowspan="3">SimCLR</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>62.7</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_8xb32-coslr-200e_in1k/simclr_resnet50_8xb32-coslr-200e_in1k_20220825-15f807a4.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_8xb32-coslr-200e_in1k/simclr_resnet50_8xb32-coslr-200e_in1k_20220721_103223.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-9596a505.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220724_210050.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>4096</td>
|
||||
<td>66.9</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/simclr/simclr_resnet50_16xb256-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-200e_in1k/simclr_resnet50_16xb256-coslr-200e_in1k_20220825-4d9cce50.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-200e_in1k/simclr_resnet50_16xb256-coslr-200e_in1k_20220721_150508.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-f12c0457.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220724_172050.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>ResNet50</td>
|
||||
<td>800</td>
|
||||
<td>4096</td>
|
||||
<td>69.2</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/simclr/simclr_resnet50_16xb256-coslr-800e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-800e_in1k/simclr_resnet50_16xb256-coslr-800e_in1k_20220825-85fcc4de.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-800e_in1k/simclr_resnet50_16xb256-coslr-800e_in1k_20220725_112248.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-800e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-b80ae1e5.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simclr/simclr_resnet50_16xb256-coslr-800e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220730_165101.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 20.60 | 33.62 | 38.86 | 45.25 | 50.91 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 20.60 | 33.62 | 38.86 | 45.25 | 50.91 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 47.8 | 48.4 | 46.7 | 45.2 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 47.8 | 48.4 | 46.7 | 45.2 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -65,19 +114,19 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 79.38 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 79.38 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 38.7 | 58.1 | 42.4 | 34.9 | 55.3 | 37.5 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 38.7 | 58.1 | 42.4 | 34.9 | 55.3 | 37.5 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -85,11 +134,11 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 64.03 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| --------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simclr/simclr_resnet50_8xb32-coslr-200e_in1k.py) | 64.03 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Siamese networks have become a common structure in various recent models for uns
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149724180-bc7bac6a-fcb8-421e-b8f1-9550c624d154.png" width="500" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -28,39 +26,80 @@ The **Best Layer** indicates that the best results are obtained from which layer
|
|||
|
||||
Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
||||
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | feature5 | 84.64 | 39.65 | 49.86 | 62.48 | 69.50 | 74.48 | 78.31 | 81.06 | 82.56 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 85.20 | 39.85 | 50.44 | 63.73 | 70.93 | 75.74 | 79.42 | 82.02 | 83.44 |
|
||||
| Self-Supervised Config | Best Layer | SVM | k=1 | k=2 | k=4 | k=8 | k=16 | k=32 | k=64 | k=96 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | feature5 | 84.64 | 39.65 | 49.86 | 62.48 | 69.50 | 74.48 | 78.31 | 81.06 | 82.56 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | feature5 | 85.20 | 39.85 | 50.44 | 63.73 | 70.93 | 75.74 | 79.42 | 82.02 | 83.44 |
|
||||
|
||||
#### ImageNet Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb512-coslr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 16.27 | 33.77 | 45.80 | 60.83 | 68.21 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 15.57 | 37.21 | 47.28 | 62.21 | 69.85 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 16.27 | 33.77 | 45.80 | 60.83 | 68.21 | 68.28 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 15.57 | 37.21 | 47.28 | 62.21 | 69.85 | 69.84 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td rowspan="2">SimSiam</td>
|
||||
<td>ResNet50</td>
|
||||
<td>100</td>
|
||||
<td>256</td>
|
||||
<td>68.3</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k/simsiam_resnet50_8xb32-coslr-100e_in1k_20220825-d07cb2e6.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k/simsiam_resnet50_8xb32-coslr-100e_in1k_20220725_224724.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-f53ba400.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220804_175115.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>69.8</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k/simsiam_resnet50_8xb32-coslr-200e_in1k_20220825-efe91299.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k/simsiam_resnet50_8xb32-coslr-200e_in1k_20220726_033722.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb512-coslr-90e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220825-519b5135.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k/resnet50_linear-8xb512-coslr-90e_in1k/resnet50_linear-8xb512-coslr-90e_in1k_20220802_120717.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_8xb32-steplr-28e_places205.py](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/classification/places205/resnet50_mhead_8xb32-steplr-28e_places205.py) for details of config.
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 21.32 | 35.66 | 43.05 | 50.79 | 53.27 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 21.17 | 35.85 | 43.49 | 50.99 | 54.10 |
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 21.32 | 35.66 | 43.05 | 50.79 | 53.27 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 21.17 | 35.85 | 43.49 | 50.99 | 54.10 |
|
||||
|
||||
#### ImageNet Nearest-Neighbor Classification
|
||||
|
||||
The results are obtained from the features after GlobalAveragePooling. Here, k=10 to 200 indicates different number of nearest neighbors.
|
||||
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 57.4 | 57.6 | 55.8 | 54.2 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 60.2 | 60.4 | 58.8 | 57.4 |
|
||||
| Self-Supervised Config | k=10 | k=20 | k=100 | k=200 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ---- | ---- | ----- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 57.4 | 57.6 | 55.8 | 54.2 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 60.2 | 60.4 | 58.8 | 57.4 |
|
||||
|
||||
### Detection
|
||||
|
||||
|
|
@ -68,21 +107,21 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 79.80 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 79.85 |
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 79.80 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 79.85 |
|
||||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 38.6 | 57.6 | 42.3 | 34.6 | 54.8 | 36.9 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 38.8 | 58.0 | 42.3 | 34.9 | 55.3 | 37.6 |
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 38.6 | 57.6 | 42.3 | 34.6 | 54.8 | 36.9 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 38.8 | 58.0 | 42.3 | 34.9 | 55.3 | 37.6 |
|
||||
|
||||
### Segmentation
|
||||
|
||||
|
|
@ -90,12 +129,12 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| -------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 48.35 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 46.27 |
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
| [resnet50_8xb32-coslr-100e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-100e_in1k.py) | 48.35 |
|
||||
| [resnet50_8xb32-coslr-200e](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/selfsup/simsiam/simsiam_resnet50_8xb32-coslr-200e_in1k.py) | 46.27 |
|
||||
|
||||
## Citation
|
||||
|
||||
|
|
|
|||
|
|
@ -12,9 +12,7 @@ Unsupervised image representations have significantly reduced the gap with super
|
|||
<img src="https://user-images.githubusercontent.com/36138628/149724517-9f1e7bdf-04c7-43e3-92f4-2b8fc1399123.png" width="500" />
|
||||
</div>
|
||||
|
||||
## Results and Models
|
||||
|
||||
**Back to [model_zoo.md](https://github.com/open-mmlab/mmselfsup/blob/master/docs/en/model_zoo.md) to download models.**
|
||||
## Models and Benchmarks
|
||||
|
||||
In this page, we provide benchmarks as much as possible to evaluate our pre-trained models. If not mentioned, all models are pre-trained on ImageNet-1k dataset.
|
||||
|
||||
|
|
@ -36,11 +34,42 @@ Besides, k=1 to 96 indicates the hyper-parameter of Low-shot SVM.
|
|||
|
||||
The **Feature1 - Feature5** don't have the GlobalAveragePooling, the feature map is pooled to the specific dimensions and then follows a Linear layer to do the classification. Please refer to [resnet50_mhead_linear-8xb32-steplr-90e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_mhead_linear-8xb32-steplr-90e_in1k.py) for details of config.
|
||||
|
||||
The **AvgPool** result is obtained from Linear Evaluation with GlobalAveragePooling. Please refer to [resnet50_linear-8xb32-coslr-100e_in1k](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-coslr-100e_in1k.py) for details of config.
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- |
|
||||
| [resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96.py) | 16.98 | 34.96 | 49.26 | 65.98 | 70.74 |
|
||||
|
||||
| Self-Supervised Config | Feature1 | Feature2 | Feature3 | Feature4 | Feature5 | AvgPool |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | -------- | -------- | -------- | -------- | ------- |
|
||||
| [resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96](https://github.com/open-mmlab/mmselfsup/blob/master/configs/selfsup/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96.py) | 16.98 | 34.96 | 49.26 | 65.98 | 70.74 | 70.47 |
|
||||
<table class="docutils">
|
||||
<thead>
|
||||
<tr>
|
||||
<th rowspan="2">Algorithm</th>
|
||||
<th rowspan="2">Backbone</th>
|
||||
<th rowspan="2">Epoch</th>
|
||||
<th rowspan="2">Batch Size</th>
|
||||
<th colspan="2" align="center">Results (Top-1 %)</th>
|
||||
<th colspan="3" align="center">Links</th>
|
||||
</tr>
|
||||
<tr>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
<th>Pretrain</th>
|
||||
<th>Linear Eval</th>
|
||||
<th>Fine-tuning</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr>
|
||||
<td>SwAV</td>
|
||||
<td>ResNet50</td>
|
||||
<td>200</td>
|
||||
<td>256</td>
|
||||
<td>70.5</td>
|
||||
<td>/</td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/selfsup/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96_20220825-5b3fc7fc.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96_20220728_141003.json'>log</a></td>
|
||||
<td><a href='https://github.com/open-mmlab/mmselfsup/blob/dev-1.x/configs/benchmarks/classification/imagenet/resnet50_linear-8xb32-coslr-100e_in1k.py'>config</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96/resnet50_linear-8xb32-coslr-100e_in1k/resnet50_linear-8xb32-coslr-100e_in1k_20220825-80341e08.pth'>model</a> | <a href='https://download.openmmlab.com/mmselfsup/1.x/swav/swav_resnet50_8xb32-mcrop-2-6-coslr-200e_in1k-224-96/resnet50_linear-8xb32-coslr-100e_in1k/resnet50_linear-8xb32-coslr-100e_in1k_20220802_145230.json'>log</a></td>
|
||||
<td>/</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
||||
#### Places205 Linear Evaluation
|
||||
|
||||
|
|
@ -64,7 +93,7 @@ The detection benchmarks includes 2 downstream task datasets, **Pascal VOC 2007
|
|||
|
||||
#### Pascal VOC 2007 + 2012
|
||||
|
||||
Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/voc0712/faster_rcnn_r50_c4_mstrain_24k_voc0712.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/voc0712/faster-rcnn_r50-c4_ms-24k_voc0712.py) for details.
|
||||
|
||||
| Self-Supervised Config | AP50 |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
|
|
@ -72,7 +101,7 @@ Please refer to [faster_rcnn_r50_c4_mstrain_24k_voc0712.py](https://github.com/o
|
|||
|
||||
#### COCO2017
|
||||
|
||||
Please refer to [mask_rcnn_r50_fpn_mstrain_1x_coco.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmdetection/coco/mask_rcnn_r50_fpn_mstrain_1x_coco.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmdetection/coco/mask-rcnn_r50_fpn_ms-1x_coco.py) for details.
|
||||
|
||||
| Self-Supervised Config | mAP(Box) | AP50(Box) | AP75(Box) | mAP(Mask) | AP50(Mask) | AP75(Mask) |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | --------- | --------- | --------- | ---------- | ---------- |
|
||||
|
|
@ -84,7 +113,7 @@ The segmentation benchmarks includes 2 downstream task datasets, **Cityscapes**
|
|||
|
||||
#### Pascal VOC 2012 + Aug
|
||||
|
||||
Please refer to [fcn_r50-d8_512x512_20k_voc12aug.py](https://github.com/open-mmlab/mmselfsup/blob/master/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_512x512_20k_voc12aug.py) for details of config.
|
||||
Please refer to [config](https://github.com/open-mmlab/mmselfsup/blob/1.x/configs/benchmarks/mmsegmentation/voc12aug/fcn_r50-d8_4xb4-20k_voc12aug-512x512.py) for details.
|
||||
|
||||
| Self-Supervised Config | mIOU |
|
||||
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- |
|
||||
|
|
|
|||
|
|
@ -12,6 +12,7 @@
|
|||
# documentation root, use os.path.abspath to make it absolute, like shown here.
|
||||
#
|
||||
import os
|
||||
import subprocess
|
||||
import sys
|
||||
|
||||
import pytorch_sphinx_theme
|
||||
|
|
@ -21,7 +22,7 @@ sys.path.insert(0, os.path.abspath('..'))
|
|||
# -- Project information -----------------------------------------------------
|
||||
|
||||
project = 'MMSelfSup'
|
||||
copyright = '2020-2021, OpenMMLab'
|
||||
copyright = '2021, OpenMMLab'
|
||||
author = 'MMSelfSup Authors'
|
||||
|
||||
# The full version, including alpha/beta/rc tags
|
||||
|
|
@ -103,12 +104,12 @@ html_theme_options = {
|
|||
'children': [
|
||||
{
|
||||
'name': 'MMSelfSup 0.x',
|
||||
'url': 'https://mmselfsup.readthedocs.io/zh_CN/latest/',
|
||||
'url': 'https://mmselfsup.readthedocs.io/en/latest/',
|
||||
'description': 'Main branch docs'
|
||||
},
|
||||
{
|
||||
'name': 'MMSelfSup 1.x',
|
||||
'url': 'https://mmselfsup.readthedocs.io/zh_CN/dev-1.x/',
|
||||
'url': 'https://mmselfsup.readthedocs.io/en/dev-1.x/',
|
||||
'description': '1.x branch docs'
|
||||
},
|
||||
],
|
||||
|
|
@ -134,3 +135,11 @@ myst_enable_extensions = ['colon_fence']
|
|||
myst_heading_anchors = 4
|
||||
|
||||
master_doc = 'index'
|
||||
|
||||
|
||||
def builder_inited_handler(app):
|
||||
subprocess.run(['./stat.py'])
|
||||
|
||||
|
||||
def setup(app):
|
||||
app.connect('builder-inited', builder_inited_handler)
|
||||
|
|
|
|||
|
|
@ -28,8 +28,11 @@ Welcome to MMSelfSup's documentation!
|
|||
.. toctree::
|
||||
:maxdepth: 1
|
||||
:caption: Model Zoo
|
||||
:glob:
|
||||
|
||||
model_zoo_statistics.md
|
||||
model_zoo.md
|
||||
papers/*
|
||||
|
||||
.. toctree::
|
||||
:maxdepth: 1
|
||||
|
|
|
|||
|
|
@ -3,16 +3,9 @@
|
|||
All models and part of benchmark results are recorded below.
|
||||
|
||||
- [Model Zoo](#model-zoo)
|
||||
- [Statistics](#statistics)
|
||||
- [Benchmarks](#benchmarks)
|
||||
- [ImageNet](#imagenet)
|
||||
|
||||
## Statistics
|
||||
|
||||
- Number of papers: 18
|
||||
|
||||
- Number of checkpoints: 62
|
||||
|
||||
## Benchmarks
|
||||
|
||||
### ImageNet
|
||||
|
|
|
|||
|
|
@ -0,0 +1,115 @@
|
|||
#!/usr/bin/env python
|
||||
import re
|
||||
from collections import defaultdict
|
||||
from pathlib import Path
|
||||
|
||||
from modelindex.load_model_index import load
|
||||
|
||||
MMSELFSUP_ROOT = Path(__file__).absolute().parents[2]
|
||||
PAPERS_ROOT = Path('papers') # Path to save generated paper pages.
|
||||
GITHUB_PREFIX = 'https://github.com/open-mmlab/mmselfsup/blob/1.x/'
|
||||
MODELZOO_TEMPLATE = """
|
||||
# Model Zoo Statistics
|
||||
|
||||
* Number of papers: {num_papers}
|
||||
{type_msg}
|
||||
|
||||
* Number of checkpoints: {num_ckpts}
|
||||
{paper_msg}
|
||||
"""
|
||||
|
||||
model_index = load(str(MMSELFSUP_ROOT / 'model-index.yml'))
|
||||
|
||||
|
||||
def build_collections(model_index):
|
||||
col_by_name = {}
|
||||
for col in model_index.collections:
|
||||
setattr(col, 'models', [])
|
||||
col_by_name[col.name] = col
|
||||
|
||||
for model in model_index.models:
|
||||
col = col_by_name[model.in_collection]
|
||||
col.models.append(model)
|
||||
setattr(model, 'collection', col)
|
||||
|
||||
|
||||
build_collections(model_index)
|
||||
|
||||
|
||||
def count_papers(model_index):
|
||||
ckpt_dict = dict()
|
||||
type_count = defaultdict(int)
|
||||
paper_msgs = []
|
||||
|
||||
for model in model_index.models:
|
||||
if model.collection.name in ckpt_dict.keys():
|
||||
if model.weights:
|
||||
ckpt_dict[model.collection.name] += 1
|
||||
else:
|
||||
ckpt_dict[model.collection.name] = 1
|
||||
|
||||
downstream_info = model.data.get('Downstream', [])
|
||||
for downstream_task in downstream_info:
|
||||
if downstream_task.get('Weights', None):
|
||||
ckpt_dict[model.collection.name] += 1
|
||||
|
||||
for collection in model_index.collections:
|
||||
name = collection.name
|
||||
title = collection.paper['Title']
|
||||
papertype = collection.data.get('type', 'Algorithm')
|
||||
type_count[papertype] += 1
|
||||
|
||||
with open(MMSELFSUP_ROOT / collection.readme) as f:
|
||||
readme = f.read()
|
||||
readme = PAPERS_ROOT / Path(
|
||||
collection.filepath).parent.with_suffix('.md').name
|
||||
paper_msgs.append(
|
||||
f'\t- [{papertype}] [{title}]({readme}) ({ckpt_dict[name]} '
|
||||
f'ckpts)')
|
||||
|
||||
type_msg = '\n'.join(
|
||||
[f'\t- {type_}: {count}' for type_, count in type_count.items()])
|
||||
paper_msg = '\n'.join(paper_msgs)
|
||||
|
||||
modelzoo = MODELZOO_TEMPLATE.format(
|
||||
num_papers=sum(type_count.values()),
|
||||
num_ckpts=sum(ckpt_dict.values()),
|
||||
type_msg=type_msg,
|
||||
paper_msg=paper_msg,
|
||||
)
|
||||
|
||||
with open('model_zoo_statistics.md', 'w') as f:
|
||||
f.write(modelzoo)
|
||||
|
||||
|
||||
count_papers(model_index)
|
||||
|
||||
|
||||
def generate_paper_page(collection):
|
||||
PAPERS_ROOT.mkdir(exist_ok=True)
|
||||
|
||||
# Write a copy of README
|
||||
with open(MMSELFSUP_ROOT / collection.readme) as f:
|
||||
readme = f.read()
|
||||
folder = Path(collection.filepath).parent
|
||||
copy = PAPERS_ROOT / folder.with_suffix('.md').name
|
||||
|
||||
def replace_link(matchobj):
|
||||
# Replace relative link to GitHub link.
|
||||
name = matchobj.group(1)
|
||||
link = matchobj.group(2)
|
||||
if not link.startswith('http'):
|
||||
assert (folder / link).exists(), \
|
||||
f'Link not found:\n{collection.readme}: {link}'
|
||||
rel_link = (folder / link).absolute().relative_to(MMSELFSUP_ROOT)
|
||||
link = GITHUB_PREFIX + str(rel_link)
|
||||
return f'[{name}]({link})'
|
||||
|
||||
content = re.sub(r'\[([^\]]+)\]\(([^)]+)\)', replace_link, readme)
|
||||
|
||||
with open(copy, 'w') as copy_file:
|
||||
copy_file.write(content)
|
||||
|
||||
|
||||
for collection in model_index.collections:
|
||||
generate_paper_page(collection)
|
||||
|
|
@ -12,6 +12,7 @@
|
|||
# documentation root, use os.path.abspath to make it absolute, like shown here.
|
||||
#
|
||||
import os
|
||||
import subprocess
|
||||
import sys
|
||||
|
||||
import pytorch_sphinx_theme
|
||||
|
|
@ -21,7 +22,7 @@ sys.path.insert(0, os.path.abspath('..'))
|
|||
# -- Project information -----------------------------------------------------
|
||||
|
||||
project = 'MMSelfSup'
|
||||
copyright = '2020-2021, OpenMMLab'
|
||||
copyright = '2021, OpenMMLab'
|
||||
author = 'MMSelfSup Authors'
|
||||
|
||||
# The full version, including alpha/beta/rc tags
|
||||
|
|
@ -130,3 +131,11 @@ myst_enable_extensions = ['colon_fence']
|
|||
myst_heading_anchors = 4
|
||||
|
||||
master_doc = 'index'
|
||||
|
||||
|
||||
def builder_inited_handler(app):
|
||||
subprocess.run(['./stat.py'])
|
||||
|
||||
|
||||
def setup(app):
|
||||
app.connect('builder-inited', builder_inited_handler)
|
||||
|
|
|
|||
|
|
@ -30,8 +30,11 @@
|
|||
.. toctree::
|
||||
:maxdepth: 1
|
||||
:caption: 模型库
|
||||
:glob:
|
||||
|
||||
model_zoo_statistics.md
|
||||
model_zoo.md
|
||||
papers/*
|
||||
|
||||
.. toctree::
|
||||
:maxdepth: 1
|
||||
|
|
|
|||
|
|
@ -3,16 +3,9 @@
|
|||
本部分内容主要介绍 MMSelfSup 支持的模型和部分下游任务的评测结果。
|
||||
|
||||
- [模型库](#模型库)
|
||||
- [数据统计](#数据统计)
|
||||
- [下游任务评测](#下游任务评测)
|
||||
- [ImageNet](#imagenet)
|
||||
|
||||
## 数据统计
|
||||
|
||||
- 论文数量: 18
|
||||
|
||||
- 模型文件数量: 62
|
||||
|
||||
## 下游任务评测
|
||||
|
||||
### ImageNet
|
||||
|
|
|
|||
|
|
@ -0,0 +1,115 @@
|
|||
#!/usr/bin/env python
|
||||
import re
|
||||
from collections import defaultdict
|
||||
from pathlib import Path
|
||||
|
||||
from modelindex.load_model_index import load
|
||||
|
||||
MMSELFSUP_ROOT = Path(__file__).absolute().parents[2]
|
||||
PAPERS_ROOT = Path('papers') # Path to save generated paper pages.
|
||||
GITHUB_PREFIX = 'https://github.com/open-mmlab/mmselfsup/blob/1.x/'
|
||||
MODELZOO_TEMPLATE = """
|
||||
# 模型库数据汇总
|
||||
|
||||
* Number of papers: {num_papers}
|
||||
{type_msg}
|
||||
|
||||
* Number of checkpoints: {num_ckpts}
|
||||
{paper_msg}
|
||||
"""
|
||||
|
||||
model_index = load(str(MMSELFSUP_ROOT / 'model-index.yml'))
|
||||
|
||||
|
||||
def build_collections(model_index):
|
||||
col_by_name = {}
|
||||
for col in model_index.collections:
|
||||
setattr(col, 'models', [])
|
||||
col_by_name[col.name] = col
|
||||
|
||||
for model in model_index.models:
|
||||
col = col_by_name[model.in_collection]
|
||||
col.models.append(model)
|
||||
setattr(model, 'collection', col)
|
||||
|
||||
|
||||
build_collections(model_index)
|
||||
|
||||
|
||||
def count_papers(model_index):
|
||||
ckpt_dict = dict()
|
||||
type_count = defaultdict(int)
|
||||
paper_msgs = []
|
||||
|
||||
for model in model_index.models:
|
||||
if model.collection.name in ckpt_dict.keys():
|
||||
if model.weights:
|
||||
ckpt_dict[model.collection.name] += 1
|
||||
else:
|
||||
ckpt_dict[model.collection.name] = 1
|
||||
|
||||
downstream_info = model.data.get('Downstream', [])
|
||||
for downstream_task in downstream_info:
|
||||
if downstream_task.get('Weights', None):
|
||||
ckpt_dict[model.collection.name] += 1
|
||||
|
||||
for collection in model_index.collections:
|
||||
name = collection.name
|
||||
title = collection.paper['Title']
|
||||
papertype = collection.data.get('type', 'Algorithm')
|
||||
type_count[papertype] += 1
|
||||
|
||||
with open(MMSELFSUP_ROOT / collection.readme) as f:
|
||||
readme = f.read()
|
||||
readme = PAPERS_ROOT / Path(
|
||||
collection.filepath).parent.with_suffix('.md').name
|
||||
paper_msgs.append(
|
||||
f'\t- [{papertype}] [{title}]({readme}) ({ckpt_dict[name]} '
|
||||
f'ckpts)')
|
||||
|
||||
type_msg = '\n'.join(
|
||||
[f'\t- {type_}: {count}' for type_, count in type_count.items()])
|
||||
paper_msg = '\n'.join(paper_msgs)
|
||||
|
||||
modelzoo = MODELZOO_TEMPLATE.format(
|
||||
num_papers=sum(type_count.values()),
|
||||
num_ckpts=sum(ckpt_dict.values()),
|
||||
type_msg=type_msg,
|
||||
paper_msg=paper_msg,
|
||||
)
|
||||
|
||||
with open('model_zoo_statistics.md', 'w') as f:
|
||||
f.write(modelzoo)
|
||||
|
||||
|
||||
count_papers(model_index)
|
||||
|
||||
|
||||
def generate_paper_page(collection):
|
||||
PAPERS_ROOT.mkdir(exist_ok=True)
|
||||
|
||||
# Write a copy of README
|
||||
with open(MMSELFSUP_ROOT / collection.readme) as f:
|
||||
readme = f.read()
|
||||
folder = Path(collection.filepath).parent
|
||||
copy = PAPERS_ROOT / folder.with_suffix('.md').name
|
||||
|
||||
def replace_link(matchobj):
|
||||
# Replace relative link to GitHub link.
|
||||
name = matchobj.group(1)
|
||||
link = matchobj.group(2)
|
||||
if not link.startswith('http'):
|
||||
assert (folder / link).exists(), \
|
||||
f'Link not found:\n{collection.readme}: {link}'
|
||||
rel_link = (folder / link).absolute().relative_to(MMSELFSUP_ROOT)
|
||||
link = GITHUB_PREFIX + str(rel_link)
|
||||
return f'[{name}]({link})'
|
||||
|
||||
content = re.sub(r'\[([^\]]+)\]\(([^)]+)\)', replace_link, readme)
|
||||
|
||||
with open(copy, 'w') as copy_file:
|
||||
copy_file.write(content)
|
||||
|
||||
|
||||
for collection in model_index.collections:
|
||||
generate_paper_page(collection)
|
||||
|
|
@ -392,9 +392,9 @@ class RandomResizedCropAndInterpolationWithTwoPic(BaseTransform):
|
|||
|
||||
@TRANSFORMS.register_module()
|
||||
class RandomGaussianBlur(BaseTransform):
|
||||
"""GaussianBlur augmentation refers to `SimCLR.
|
||||
"""GaussianBlur augmentation refers to SimCLR.
|
||||
|
||||
<https://arxiv.org/abs/2002.05709>`_.
|
||||
`Paper link <https://arxiv.org/abs/2002.05709>`_.
|
||||
|
||||
Required Keys:
|
||||
|
||||
|
|
@ -448,9 +448,9 @@ class RandomGaussianBlur(BaseTransform):
|
|||
|
||||
@TRANSFORMS.register_module()
|
||||
class RandomSolarize(BaseTransform):
|
||||
"""Solarization augmentation refers to `BYOL.
|
||||
"""Solarization augmentation refers to BYOL.
|
||||
|
||||
<https://arxiv.org/abs/2006.07733>`_.
|
||||
`Paper link <https://arxiv.org/abs/2006.07733>`_.
|
||||
|
||||
Required Keys:
|
||||
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ class LARS(Optimizer):
|
|||
|
||||
Based on Algorithm 1 of the following paper by You, Gitman, and Ginsburg.
|
||||
`Large Batch Training of Convolutional Networks:
|
||||
<https://arxiv.org/abs/1708.03888>`_.
|
||||
<https://arxiv.org/abs/1708.03888>`_.
|
||||
|
||||
Args:
|
||||
params (Iterable): Iterable of parameters to optimize or dicts defining
|
||||
|
|
|
|||
|
|
@ -100,7 +100,8 @@ class BaseModel(_BaseModel):
|
|||
data_samples (List[BaseDataElement], optional):
|
||||
data samples collated by :attr:`data_preprocessor`.
|
||||
mode (str): mode should be one of ``loss``, ``predict`` and
|
||||
``tensor``
|
||||
``tensor``.
|
||||
|
||||
- ``loss``: Called by ``train_step`` and return loss ``dict``
|
||||
used for logging
|
||||
- ``predict``: Called by ``val_step`` and ``test_step``
|
||||
|
|
@ -108,15 +109,16 @@ class BaseModel(_BaseModel):
|
|||
computing metric.
|
||||
- ``tensor``: Called by custom use to get ``Tensor`` type
|
||||
results.
|
||||
|
||||
Returns:
|
||||
ForwardResults:
|
||||
- If ``mode == loss``, return a ``dict`` of loss tensor used
|
||||
for backward and logging.
|
||||
- If ``mode == predict``, return a ``list`` of
|
||||
:obj:`BaseDataElement` for computing metric
|
||||
and getting inference result.
|
||||
- If ``mode == tensor``, return a tensor or ``tuple`` of tensor
|
||||
or ``dict of tensor for custom use.
|
||||
ForwardResults (dict or list):
|
||||
- If ``mode == loss``, return a ``dict`` of loss tensor used
|
||||
for backward and logging.
|
||||
- If ``mode == predict``, return a ``list`` of
|
||||
:obj:`BaseDataElement` for computing metric
|
||||
and getting inference result.
|
||||
- If ``mode == tensor``, return a tensor or ``tuple`` of tensor
|
||||
or ``dict of tensor for custom use.
|
||||
"""
|
||||
if mode == 'tensor':
|
||||
feats = self.extract_feat(inputs, data_samples=data_samples)
|
||||
|
|
|
|||
|
|
@ -13,8 +13,8 @@ class BEiT(BaseModel):
|
|||
"""BEiT v1/v2.
|
||||
|
||||
Implementation of `BEiT: BERT Pre-Training of Image Transformers
|
||||
<https://arxiv.org/abs/2106.08254>`_. Implementation of `BEiT v2: Masked
|
||||
Image Modeling with Vector-Quantized Visual Tokenizers
|
||||
<https://arxiv.org/abs/2106.08254>`_ and `BEiT v2: Masked Image Modeling
|
||||
with Vector-Quantized Visual Tokenizers
|
||||
<https://arxiv.org/abs/2208.06366>`_.
|
||||
"""
|
||||
|
||||
|
|
|
|||
|
|
@ -17,7 +17,7 @@ class ODC(BaseModel):
|
|||
Official implementation of `Online Deep Clustering for Unsupervised
|
||||
Representation Learning <https://arxiv.org/abs/2006.10645>`_.
|
||||
The operation w.r.t. memory bank and loss re-weighting is in
|
||||
`engine/hooks/odc_hook.py`.
|
||||
`engine/hooks/odc_hook.py`.
|
||||
|
||||
Args:
|
||||
backbone (dict): Config dict for module of backbone.
|
||||
|
|
|
|||
|
|
@ -19,8 +19,8 @@ class BEiTViT(BEiT):
|
|||
|
||||
Args:
|
||||
arch (str | dict): Vision Transformer architecture. If use string,
|
||||
choose from 'small', 'base', 'large', 'deit-tiny', 'deit-small'
|
||||
and 'deit-base'. If use dict, it should have below keys:
|
||||
choose from 'small', 'base' and 'large'. If use dict, it should
|
||||
have below keys:
|
||||
|
||||
- **embed_dims** (int): The dimensions of embedding.
|
||||
- **num_layers** (int): The number of transformer encoder layers.
|
||||
|
|
|
|||
|
|
@ -107,11 +107,11 @@ class MAEViT(VisionTransformer):
|
|||
Defaults to 0.75.
|
||||
|
||||
Returns:
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor]: masked image,
|
||||
mask and the ids to restore original image.
|
||||
- x_masked (torch.Tensor): masked image.
|
||||
- mask (torch.Tensor): mask used to mask image.
|
||||
- ids_restore (torch.Tensor): ids to restore original image.
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
|
||||
masked image, mask and the ids to restore original image.
|
||||
- x_masked (torch.Tensor): masked image.
|
||||
- mask (torch.Tensor): mask used to mask image.
|
||||
- ids_restore (torch.Tensor): ids to restore original image.
|
||||
"""
|
||||
N, L, D = x.shape # batch, length, dim
|
||||
len_keep = int(L * (1 - mask_ratio))
|
||||
|
|
@ -148,10 +148,12 @@ class MAEViT(VisionTransformer):
|
|||
x (torch.Tensor): Input images, which is of shape B x C x H x W.
|
||||
|
||||
Returns:
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor]: hidden features,
|
||||
mask and the ids to restore original image.
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
|
||||
|
||||
Hidden features, mask and the ids to restore original image.
|
||||
|
||||
- x (torch.Tensor): hidden features, which is of shape
|
||||
B x (L * mask_ratio) x C.
|
||||
B x (L * mask_ratio) x C.
|
||||
- mask (torch.Tensor): mask used to mask image.
|
||||
- ids_restore (torch.Tensor): ids to restore original image.
|
||||
"""
|
||||
|
|
|
|||
|
|
@ -35,10 +35,10 @@ class MILANViT(MAEViT):
|
|||
shape B x L.
|
||||
|
||||
Returns:
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor
|
||||
torch.Tensor, torch.Tensor]: masked image, mask, the ids to
|
||||
restore original image, ids of the shuffled patches, ids of
|
||||
the kept patches, ids of the removed patches.
|
||||
Tuple[torch.Tensor, ...]:
|
||||
masked image, mask, the ids to restore original image,
|
||||
ids of the shuffled patches, ids of the kept patches,
|
||||
ids of the removed patches.
|
||||
"""
|
||||
N, L, D = x.shape # batch, length, dim
|
||||
len_keep = int(L * (1 - mask_ratio))
|
||||
|
|
@ -82,15 +82,14 @@ class MILANViT(MAEViT):
|
|||
shape B x L.
|
||||
|
||||
Returns:
|
||||
Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor
|
||||
torch.Tensor, torch.Tensor]: masked image, the ids to
|
||||
restore original image, ids of the kept patches, ids of the
|
||||
removed patches.
|
||||
- x (torch.Tensor): hidden features, which is of shape
|
||||
Tuple[torch.Tensor, ...]:
|
||||
masked image, the ids to restore original image, ids of the
|
||||
kept patches, ids of the removed patches.
|
||||
- x (torch.Tensor): hidden features, which is of shape
|
||||
B x (L * mask_ratio) x C.
|
||||
- ids_restore (torch.Tensor): ids to restore original image.
|
||||
- ids_keep (torch.Tensor): ids of the kept patches.
|
||||
- ids_dump (torch.Tensor): ids of the removed patches.
|
||||
- ids_restore (torch.Tensor): ids to restore original image.
|
||||
- ids_keep (torch.Tensor): ids of the kept patches.
|
||||
- ids_dump (torch.Tensor): ids of the removed patches.
|
||||
"""
|
||||
B = x.shape[0]
|
||||
x = self.patch_embed(x)[0]
|
||||
|
|
|
|||
|
|
@ -48,9 +48,10 @@ class LatentPredictHead(BaseModule):
|
|||
|
||||
@MODELS.register_module()
|
||||
class LatentCrossCorrelationHead(BaseModule):
|
||||
"""Head for latent feature cross correlation. Part of the code is borrowed
|
||||
from:
|
||||
`https://github.com/facebookresearch/barlowtwins/blob/main/main.py>`_.
|
||||
"""Head for latent feature cross correlation.
|
||||
|
||||
Part of the code is borrowed from `script
|
||||
<https://github.com/facebookresearch/barlowtwins/blob/main/main.py>`_.
|
||||
|
||||
Args:
|
||||
in_channels (int): Number of input channels.
|
||||
|
|
|
|||
|
|
@ -17,9 +17,9 @@ class SwAVLoss(BaseModule):
|
|||
"""The Loss for SwAV.
|
||||
|
||||
This Loss contains clustering and sinkhorn algorithms to compute Q codes.
|
||||
Part of the code is borrowed from:
|
||||
`<https://github.com/facebookresearch/swav`_.
|
||||
The queue is built in `core/hooks/swav_hook.py`.
|
||||
Part of the code is borrowed from `script
|
||||
<https://github.com/facebookresearch/swav>`_.
|
||||
The queue is built in `engine/hooks/swav_hook.py`.
|
||||
|
||||
Args:
|
||||
feat_dim (int): feature dimension of the prototypes.
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ class DenseCLNeck(BaseModule):
|
|||
"""The non-linear neck of DenseCL.
|
||||
|
||||
Single and dense neck in parallel: fc-relu-fc, conv-relu-conv.
|
||||
Borrowed from the authors' code: `<https://github.com/WXinlong/DenseCL`_.
|
||||
Borrowed from the authors' `code <https://github.com/WXinlong/DenseCL>`_.
|
||||
|
||||
Args:
|
||||
in_channels (int): Number of input channels.
|
||||
|
|
|
|||
|
|
@ -32,9 +32,9 @@ class CLIPGenerator(BaseModule):
|
|||
x (torch.Tensor): The input image, which is of shape (N, 3, H, W).
|
||||
|
||||
Returns:
|
||||
Tuple[torch.Tensor, torch.Tensor]: The features and attention from
|
||||
the last layer of CLIP, which are of shape (N, L, C) and
|
||||
(N, L, L), respectively.
|
||||
Tuple[torch.Tensor, torch.Tensor]:
|
||||
The features and attention from the last layer of CLIP,
|
||||
which are of shape (N, L, C) and (N, L, L), respectively.
|
||||
"""
|
||||
# use the visual branch of CLIP to get the features
|
||||
clip_features = self.tokenizer.encode_image(x)
|
||||
|
|
|
|||
|
|
@ -8,86 +8,83 @@ class SelfSupDataSample(BaseDataElement):
|
|||
between different components.
|
||||
|
||||
Meta field:
|
||||
- ``img_shape`` (Tuple): The shape of the corresponding input image.
|
||||
Used for visualization.
|
||||
- ``ori_shape`` (Tuple): The original shape of the corresponding image.
|
||||
Used for visualization.
|
||||
- ``img_path`` (str): The path of original image.
|
||||
|
||||
- ``img_shape`` (Tuple): The shape of the corresponding input image.
|
||||
Used for visualization.
|
||||
|
||||
- ``ori_shape`` (Tuple): The original shape of the corresponding image.
|
||||
Used for visualization.
|
||||
|
||||
- ``img_path`` (str): The path of original image.
|
||||
|
||||
Data field:
|
||||
- ``gt_label``(LabelData): The ground truth label of an image.
|
||||
- ``sample_idx``(InstanceData): The idx of an image in the dataset.
|
||||
- ``mask``(BaseDataElement): Mask used in masks image modeling.
|
||||
- ``pred_label`` (LabelData): The predicted label.
|
||||
- ``pseudo_label``(InstanceData): Label used in pretext task,
|
||||
|
||||
- ``gt_label`` (LabelData): The ground truth label of an image.
|
||||
|
||||
- ``sample_idx`` (InstanceData): The idx of an image in the dataset.
|
||||
|
||||
- ``mask`` (BaseDataElement): Mask used in masks image modeling.
|
||||
|
||||
- ``pred_label`` (LabelData): The predicted label.
|
||||
|
||||
- ``pseudo_label`` (InstanceData): Label used in pretext task,
|
||||
e.g. Relative Location.
|
||||
|
||||
Examples:
|
||||
>>> import torch
|
||||
>>> import numpy as np
|
||||
>>> from mmengine.structure import InstanceData
|
||||
>>> from mmselfsup.structures import SelfSupDataSample
|
||||
>>> import torch
|
||||
>>> import numpy as np
|
||||
>>> from mmengine.structure import InstanceData
|
||||
>>> from mmselfsup.structures import SelfSupDataSample
|
||||
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> gt_label = LabelData()
|
||||
>>> gt_label.value = [1]
|
||||
>>> data_sample.gt_label = gt_label
|
||||
>>> len(data_sample.gt_label)
|
||||
1
|
||||
>>> print(data_sample)
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> gt_label = LabelData()
|
||||
>>> gt_label.value = [1]
|
||||
>>> data_sample.gt_label = gt_label
|
||||
>>> len(data_sample.gt_label)
|
||||
1
|
||||
>>> print(data_sample)
|
||||
<SelfSupDataSample(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
gt_label: <InstanceData(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
value: [1]
|
||||
) at 0x7f15c08f9d10>
|
||||
_gt_label: <InstanceData(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
value: [1]
|
||||
) at 0x7f15c08f9d10>
|
||||
) at 0x7f15c077ef10>
|
||||
>>> idx = InstanceData()
|
||||
>>> idx.value = [0]
|
||||
>>> data_sample = SelfSupDataSample(idx=idx)
|
||||
>>> assert 'idx' in data_sample
|
||||
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> mask = dict(value=np.random.rand(48, 48))
|
||||
>>> mask = PixelData(**mask)
|
||||
>>> data_sample.mask = mask
|
||||
>>> assert 'mask' in data_sample
|
||||
>>> assert 'value' in data_sample.mask
|
||||
>>> idx = InstanceData()
|
||||
>>> idx.value = [0]
|
||||
>>> data_sample = SelfSupDataSample(idx=idx)
|
||||
>>> assert 'idx' in data_sample
|
||||
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> pred_label = dict(pred_label=[3])
|
||||
>>> pred_label = LabelData(**pred_label)
|
||||
>>> data_sample.pred_label = pred_label
|
||||
>>> print(data_sample)
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> mask = dict(value=np.random.rand(48, 48))
|
||||
>>> mask = PixelData(**mask)
|
||||
>>> data_sample.mask = mask
|
||||
>>> assert 'mask' in data_sample
|
||||
>>> assert 'value' in data_sample.mask
|
||||
|
||||
>>> data_sample = SelfSupDataSample()
|
||||
>>> pred_label = dict(pred_label=[3])
|
||||
>>> pred_label = LabelData(**pred_label)
|
||||
>>> data_sample.pred_label = pred_label
|
||||
>>> print(data_sample)
|
||||
<SelfSupDataSample(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
_pred_label: <InstanceData(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
pred_label: [3]
|
||||
) at 0x7f15c06a3990>
|
||||
pred_label: <InstanceData(
|
||||
|
||||
META INFORMATION
|
||||
|
||||
DATA FIELDS
|
||||
pred_label: [3]
|
||||
) at 0x7f15c06a3990>
|
||||
|
|
|
|||
|
|
@ -161,14 +161,16 @@ class SelfSupVisualizer(Visualizer):
|
|||
step: int = 0) -> None:
|
||||
"""Draw datasample and save to all backends.
|
||||
|
||||
- If GT and prediction are plotted at the same time, they are
|
||||
displayed in a stitched image where the left image is the
|
||||
ground truth and the right image is the prediction.
|
||||
- If ``show`` is True, all storage backends are ignored, and
|
||||
the images will be displayed in a local window.
|
||||
- If ``out_file`` is specified, the drawn image will be
|
||||
saved to ``out_file``. t is usually used when the display
|
||||
is not available.
|
||||
- If GT and prediction are plotted at the same time, they are
|
||||
displayed in a stitched image where the left image is the
|
||||
ground truth and the right image is the prediction.
|
||||
|
||||
- If ``show`` is True, all storage backends are ignored, and
|
||||
the images will be displayed in a local window.
|
||||
|
||||
- If ``out_file`` is specified, the drawn image will be
|
||||
saved to ``out_file``. t is usually used when the display
|
||||
is not available.
|
||||
|
||||
Args:
|
||||
name (str): The image identifier.
|
||||
|
|
|
|||
|
|
@ -1,5 +1,6 @@
|
|||
docutils==0.16.0
|
||||
markdown>=3.4.0
|
||||
modelindex
|
||||
myst_parser
|
||||
-e git+https://github.com/open-mmlab/pytorch_sphinx_theme.git#egg=pytorch_sphinx_theme
|
||||
recommonmark
|
||||
|
|
|
|||
Loading…
Reference in New Issue