提交 a7a45dd1 编写于 作者: littletomatodonkey's avatar littletomatodonkey

update update_history

上级 cc1e805e
...@@ -7,7 +7,8 @@ ...@@ -7,7 +7,8 @@
飞桨图像分类套件PaddleClas是飞桨为工业界和学术界所准备的一个图像分类任务的工具集,助力使用者训练出更好的视觉模型和应用落地。 飞桨图像分类套件PaddleClas是飞桨为工业界和学术界所准备的一个图像分类任务的工具集,助力使用者训练出更好的视觉模型和应用落地。
**近期更新** **近期更新**
- 2020.07.14 添加Res2Net200_vd_26w_4s_ssld模型,在ImageNet上Top-1 Acc可达85.13%;添加Fix_ResNet50_vd_ssld_v2模型,,在ImageNet上Top-1 Acc可达84.0%。 - 2020.09.07 添加HRNet_W18_C_ssld模型,在ImageNet上Top-1 Acc可达0.81162;添加MobileNetV3_small_x0_35_ssld模型,在ImageNet上Top-1 Acc可达0.5555。
- 2020.07.14 添加Res2Net200_vd_26w_4s_ssld模型,在ImageNet上Top-1 Acc可达85.13%;添加Fix_ResNet50_vd_ssld_v2模型,在ImageNet上Top-1 Acc可达84.0%。
- 2020.06.17 添加英文文档。 - 2020.06.17 添加英文文档。
- 2020.06.12 添加对windows和CPU环境的训练与评估支持。 - 2020.06.12 添加对windows和CPU环境的训练与评估支持。
- 2020.05.17 添加混合精度训练,基于ResNet50模型,精度几乎无损的情况下,训练时间可以减少约40%。 - 2020.05.17 添加混合精度训练,基于ResNet50模型,精度几乎无损的情况下,训练时间可以减少约40%。
...@@ -122,30 +123,30 @@ SEResNeXt与Res2Net系列模型的精度、速度指标如下表所示,更多 ...@@ -122,30 +123,30 @@ SEResNeXt与Res2Net系列模型的精度、速度指标如下表所示,更多
| 模型 | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | Flops(G) | Params(M) | 下载地址 | | 模型 | Top-1 Acc | Top-5 Acc | time(ms)<br>bs=1 | time(ms)<br>bs=4 | Flops(G) | Params(M) | 下载地址 |
|---------------------------|-----------|-----------|-----------------------|----------------------|----------|-----------|----------------------------------------------------------------------------------------------------| |---------------------------|-----------|-----------|-----------------------|----------------------|----------|-----------|----------------------------------------------------------------------------------------------------|
| Res2Net50_26w_4s | 0.7933 | 0.9457 | 4.47188 | 9.65722 | 8.52 | 25.7 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_26w_4s_pretrained.tar) | | Res2Net50_<br>26w_4s | 0.7933 | 0.9457 | 4.47188 | 9.65722 | 8.52 | 25.7 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_26w_4s_pretrained.tar) |
| Res2Net50_vd_26w_4s | 0.7975 | 0.9491 | 4.52712 | 9.93247 | 8.37 | 25.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar) | | Res2Net50_vd_<br>26w_4s | 0.7975 | 0.9491 | 4.52712 | 9.93247 | 8.37 | 25.06 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar) |
| Res2Net50_14w_8s | 0.7946 | 0.9470 | 5.4026 | 10.60273 | 9.01 | 25.72 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_14w_8s_pretrained.tar) | | Res2Net50_<br>14w_8s | 0.7946 | 0.9470 | 5.4026 | 10.60273 | 9.01 | 25.72 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_14w_8s_pretrained.tar) |
| Res2Net101_vd_26w_4s | 0.8064 | 0.9522 | 8.08729 | 17.31208 | 16.67 | 45.22 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net101_vd_26w_4s_pretrained.tar) | | Res2Net101_vd_<br>26w_4s | 0.8064 | 0.9522 | 8.08729 | 17.31208 | 16.67 | 45.22 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net101_vd_26w_4s_pretrained.tar) |
| Res2Net200_vd_26w_4s | 0.8121 | 0.9571 | 14.67806 | 32.35032 | 31.49 | 76.21 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net200_vd_26w_4s_pretrained.tar) | | Res2Net200_vd_<br>26w_4s | 0.8121 | 0.9571 | 14.67806 | 32.35032 | 31.49 | 76.21 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net200_vd_26w_4s_pretrained.tar) |
| Res2Net200_vd_26w_4s_ssld | 0.8513 | 0.9742 | 14.67806 | 32.35032 | 31.49 | 76.21 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net200_vd_26w_4s_ssld_pretrained.tar) | | Res2Net200_vd_<br>26w_4s_ssld | 0.8513 | 0.9742 | 14.67806 | 32.35032 | 31.49 | 76.21 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net200_vd_26w_4s_ssld_pretrained.tar) |
| ResNeXt50_32x4d | 0.7775 | 0.9382 | 7.56327 | 10.6134 | 8.02 | 23.64 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_32x4d_pretrained.tar) | | ResNeXt50_<br>32x4d | 0.7775 | 0.9382 | 7.56327 | 10.6134 | 8.02 | 23.64 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_32x4d_pretrained.tar) |
| ResNeXt50_vd_32x4d | 0.7956 | 0.9462 | 7.62044 | 11.03385 | 8.5 | 23.66 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_vd_32x4d_pretrained.tar) | | ResNeXt50_vd_<br>32x4d | 0.7956 | 0.9462 | 7.62044 | 11.03385 | 8.5 | 23.66 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_vd_32x4d_pretrained.tar) |
| ResNeXt50_64x4d | 0.7843 | 0.9413 | 13.80962 | 18.4712 | 15.06 | 42.36 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_64x4d_pretrained.tar) | | ResNeXt50_<br>64x4d | 0.7843 | 0.9413 | 13.80962 | 18.4712 | 15.06 | 42.36 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_64x4d_pretrained.tar) |
| ResNeXt50_vd_64x4d | 0.8012 | 0.9486 | 13.94449 | 18.88759 | 15.54 | 42.38 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_vd_64x4d_pretrained.tar) | | ResNeXt50_vd_<br>64x4d | 0.8012 | 0.9486 | 13.94449 | 18.88759 | 15.54 | 42.38 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt50_vd_64x4d_pretrained.tar) |
| ResNeXt101_32x4d | 0.7865 | 0.9419 | 16.21503 | 19.96568 | 15.01 | 41.54 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x4d_pretrained.tar) | | ResNeXt101_<br>32x4d | 0.7865 | 0.9419 | 16.21503 | 19.96568 | 15.01 | 41.54 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_32x4d_pretrained.tar) |
| ResNeXt101_vd_32x4d | 0.8033 | 0.9512 | 16.28103 | 20.25611 | 15.49 | 41.56 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_32x4d_pretrained.tar) | | ResNeXt101_vd_<br>32x4d | 0.8033 | 0.9512 | 16.28103 | 20.25611 | 15.49 | 41.56 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_32x4d_pretrained.tar) |
| ResNeXt101_64x4d | 0.7835 | 0.9452 | 30.4788 | 36.29801 | 29.05 | 78.12 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_64x4d_pretrained.tar) | | ResNeXt101_<br>64x4d | 0.7835 | 0.9452 | 30.4788 | 36.29801 | 29.05 | 78.12 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_64x4d_pretrained.tar) |
| ResNeXt101_vd_64x4d | 0.8078 | 0.9520 | 30.40456 | 36.77324 | 29.53 | 78.14 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_64x4d_pretrained.tar) | | ResNeXt101_vd_<br>64x4d | 0.8078 | 0.9520 | 30.40456 | 36.77324 | 29.53 | 78.14 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_64x4d_pretrained.tar) |
| ResNeXt152_32x4d | 0.7898 | 0.9433 | 24.86299 | 29.36764 | 22.01 | 56.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_32x4d_pretrained.tar) | | ResNeXt152_<br>32x4d | 0.7898 | 0.9433 | 24.86299 | 29.36764 | 22.01 | 56.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_32x4d_pretrained.tar) |
| ResNeXt152_vd_32x4d | 0.8072 | 0.9520 | 25.03258 | 30.08987 | 22.49 | 56.3 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_32x4d_pretrained.tar) | | ResNeXt152_vd_<br>32x4d | 0.8072 | 0.9520 | 25.03258 | 30.08987 | 22.49 | 56.3 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_32x4d_pretrained.tar) |
| ResNeXt152_64x4d | 0.7951 | 0.9471 | 46.7564 | 56.34108 | 43.03 | 107.57 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_64x4d_pretrained.tar) | | ResNeXt152_<br>64x4d | 0.7951 | 0.9471 | 46.7564 | 56.34108 | 43.03 | 107.57 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_64x4d_pretrained.tar) |
| ResNeXt152_vd_64x4d | 0.8108 | 0.9534 | 47.18638 | 57.16257 | 43.52 | 107.59 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_64x4d_pretrained.tar) | | ResNeXt152_vd_<br>64x4d | 0.8108 | 0.9534 | 47.18638 | 57.16257 | 43.52 | 107.59 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt152_vd_64x4d_pretrained.tar) |
| SE_ResNet18_vd | 0.7333 | 0.9138 | 1.7691 | 4.19877 | 4.14 | 11.8 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet18_vd_pretrained.tar) | | SE_ResNet18_vd | 0.7333 | 0.9138 | 1.7691 | 4.19877 | 4.14 | 11.8 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet18_vd_pretrained.tar) |
| SE_ResNet34_vd | 0.7651 | 0.9320 | 2.88559 | 7.03291 | 7.84 | 21.98 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet34_vd_pretrained.tar) | | SE_ResNet34_vd | 0.7651 | 0.9320 | 2.88559 | 7.03291 | 7.84 | 21.98 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet34_vd_pretrained.tar) |
| SE_ResNet50_vd | 0.7952 | 0.9475 | 4.28393 | 10.38846 | 8.67 | 28.09 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet50_vd_pretrained.tar) | | SE_ResNet50_vd | 0.7952 | 0.9475 | 4.28393 | 10.38846 | 8.67 | 28.09 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNet50_vd_pretrained.tar) |
| SE_ResNeXt50_32x4d | 0.7844 | 0.9396 | 8.74121 | 13.563 | 8.02 | 26.16 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_32x4d_pretrained.tar) | | SE_ResNeXt50_<br>32x4d | 0.7844 | 0.9396 | 8.74121 | 13.563 | 8.02 | 26.16 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_32x4d_pretrained.tar) |
| SE_ResNeXt50_vd_32x4d | 0.8024 | 0.9489 | 9.17134 | 14.76192 | 10.76 | 26.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_vd_32x4d_pretrained.tar) | | SE_ResNeXt50_vd_<br>32x4d | 0.8024 | 0.9489 | 9.17134 | 14.76192 | 10.76 | 26.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt50_vd_32x4d_pretrained.tar) |
| SE_ResNeXt101_32x4d | 0.7912 | 0.9420 | 18.82604 | 25.31814 | 15.02 | 46.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar) | | SE_ResNeXt101_<br>32x4d | 0.7912 | 0.9420 | 18.82604 | 25.31814 | 15.02 | 46.28 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SE_ResNeXt101_32x4d_pretrained.tar) |
| SENet154_vd | 0.8140 | 0.9548 | 53.79794 | 66.31684 | 45.83 | 114.29 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar) | | SENet154_vd | 0.8140 | 0.9548 | 53.79794 | 66.31684 | 45.83 | 114.29 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar) |
......
# Release Notes # Release Notes
* 2020.09.07
* Add `HRNet_W18_C_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 81.16%.
* Add `MobileNetV3_small_x0_35_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 55.55%.
* 2020.07.14 * 2020.07.14
* Add `Res2Net200_vd_26w_4s_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 85.13%. * Add `Res2Net200_vd_26w_4s_ssld` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 85.13%.
* Add `Fix_ResNet50_vd_ssld_v2` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 84.00%. * Add `Fix_ResNet50_vd_ssld_v2` pretrained model, whose Top-1 Acc on ImageNet1k dataset reaches 84.00%.
......
# 更新日志 # 更新日志
* 2020.09.07
* 添加HRNet_W18_C_ssld模型,在ImageNet上Top-1 Acc可达0.81162;添加MobileNetV3_small_x0_35_ssld模型,在ImageNet上Top-1 Acc可达0.5555。
* 2020.07.14 * 2020.07.14
* 添加Res2Net200_vd_26w_4s_ssld模型,在ImageNet上Top-1 Acc可达85.13%。 * 添加Res2Net200_vd_26w_4s_ssld模型,在ImageNet上Top-1 Acc可达85.13%。
* 添加Fix_ResNet50_vd_ssld_v2模型,,在ImageNet上Top-1 Acc可达84.0%。 * 添加Fix_ResNet50_vd_ssld_v2模型,,在ImageNet上Top-1 Acc可达84.0%。
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册