diff --git a/docs/zh_CN_tmp/.gitkeep b/docs/zh_CN_tmp/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000 diff --git a/docs/zh_CN_tmp/advanced_tutorials/.gitkeep b/docs/zh_CN_tmp/advanced_tutorials/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000 diff --git a/docs/zh_CN_tmp/advanced_tutorials/DataAugmentation.md b/docs/zh_CN_tmp/advanced_tutorials/DataAugmentation.md index 0b1e0459454b5987ae1a60182545cd2ca4c84a76..cad5d04e4ab54a2df0118dbaaed6c203d35adddf 100644 --- a/docs/zh_CN_tmp/advanced_tutorials/DataAugmentation.md +++ b/docs/zh_CN_tmp/advanced_tutorials/DataAugmentation.md @@ -1,6 +1,6 @@ # 一、数据增强分类实战 -本节将基于ImageNet-1K的数据集详细介绍数据增强实验，如果想快速体验此方法，可以参考[**30分钟玩转PaddleClas（进阶版）**](../../tutorials/quick_start_professional.md)中基于CIFAR100的数据增强实验。如果想了解相关算法的内容，请参考[数据增强算法介绍](../algorithm_introduction/DataAugmentation.md)。 +本节将基于ImageNet-1K的数据集详细介绍数据增强实验，如果想快速体验此方法，可以参考[**30分钟玩转PaddleClas（进阶版）**](../quick_start/quick_start_classification.md)中基于CIFAR100的数据增强实验。如果想了解相关算法的内容，请参考[数据增强算法介绍](../algorithm_introduction/DataAugmentation.md)。 ## 1.1 参数配置 diff --git a/docs/zh_CN_tmp/algorithm_introduction/.gitkeep b/docs/zh_CN_tmp/algorithm_introduction/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000 diff --git a/docs/zh_CN_tmp/algorithm_introduction/ImageNet_models.md b/docs/zh_CN_tmp/algorithm_introduction/ImageNet_models.md index ec1a1775c1fa62c72140233a37a8b798db5a61ac..35cc9429fc54f45fac285eac3e9ff3f07d130c1f 100644 --- a/docs/zh_CN_tmp/algorithm_introduction/ImageNet_models.md +++ b/docs/zh_CN_tmp/algorithm_introduction/ImageNet_models.md @@ -25,7 +25,7 @@ ### SSLD知识蒸馏预训练模型 -基于SSLD知识蒸馏的预训练模型列表如下所示，更多关于SSLD知识蒸馏方案的介绍可以参考：[SSLD知识蒸馏文档](./advanced_tutorials/distillation/distillation.md)。 +基于SSLD知识蒸馏的预训练模型列表如下所示，更多关于SSLD知识蒸馏方案的介绍可以参考：[SSLD知识蒸馏文档](./knowledge_distillation.md)。 * 服务器端知识蒸馏模型 @@ -298,7 +298,7 @@ ViT（Vision Transformer）与DeiT（Data-efficient Image Transformers）系列 | ViT_large_
patch16_224 | 0.8323 | 0.9650 | - | - | | 307 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ViT_large_patch16_224_pretrained.pdparams) | | ViT_large_
patch16_384 | 0.8513 | 0.9736 | - | - | | | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ViT_large_patch16_384_pretrained.pdparams) | | ViT_large_
patch32_384 | 0.8153 | 0.9608 | - | - | | | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ViT_large_patch32_384_pretrained.pdparams) | -| | | | | | | | | + | 模型 | Top-1 Acc | Top-5 Acc | time(ms)
bs=1 | time(ms)
bs=4 | Flops(G) | Params(M) | 下载地址 | @@ -311,7 +311,7 @@ ViT（Vision Transformer）与DeiT（Data-efficient Image Transformers）系列 | DeiT_small_
distilled_patch16_224 | 0.809 | 0.953 | - | - | | 22 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/DeiT_small_distilled_patch16_224_pretrained.pdparams) | | DeiT_base_
distilled_patch16_224 | 0.831 | 0.964 | - | - | | 87 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/DeiT_base_distilled_patch16_224_pretrained.pdparams) | | DeiT_base_
distilled_patch16_384 | 0.851 | 0.973 | - | - | | 88 | [下载链接](https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/DeiT_base_distilled_patch16_384_pretrained.pdparams) | -| | | | | | | | | + diff --git a/docs/zh_CN_tmp/algorithm_introduction/image_classification.md b/docs/zh_CN_tmp/algorithm_introduction/image_classification.md index 4e9383129801f27663a9e2b909527e2023519445..05b7c88671a5737db27fe5ba487bcf7288921de3 100644 --- a/docs/zh_CN_tmp/algorithm_introduction/image_classification.md +++ b/docs/zh_CN_tmp/algorithm_introduction/image_classification.md @@ -28,7 +28,7 @@ CIFAR-10数据集由10个类的60000个彩色图像组成，图像分辨率为32 ### 2.2 模型准备 -在数据确定后，模型往往决定了最终算法精度的上限，在图像分类领域，经典的模型层出不穷，PaddleClas提供了36个系列共175个ImageNet预训练模型。具体的精度、速度等指标请参考[骨干网络和预训练模型库](../ImageNet_models_cn.md)。 +在数据确定后，模型往往决定了最终算法精度的上限，在图像分类领域，经典的模型层出不穷，PaddleClas提供了36个系列共175个ImageNet预训练模型。具体的精度、速度等指标请参考[骨干网络和预训练模型库](./ImageNet_models.md)。 ### 2.3 模型训练 diff --git a/docs/zh_CN_tmp/data_preparation/.gitkeep b/docs/zh_CN_tmp/data_preparation/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000 diff --git a/docs/zh_CN_tmp/image_recognition_pipeline/.gitkeep b/docs/zh_CN_tmp/image_recognition_pipeline/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000 diff --git a/docs/zh_CN_tmp/others/competition_support.md b/docs/zh_CN_tmp/others/competition_support.md index db702d496db76eee61dfc3ce9209a9e7bac5c198..4a3f24ef956f51aa056facbae823e0914c1b300a 100644 --- a/docs/zh_CN_tmp/others/competition_support.md +++ b/docs/zh_CN_tmp/others/competition_support.md @@ -6,7 +6,7 @@ PaddleClas的建设源于百度实际视觉业务应用的淬炼和视觉前沿 * 2019年Kaggle Open Images V5图像目标检测挑战赛亚军 * 技术报告可以参考：[https://arxiv.org/pdf/1911.07171.pdf](https://arxiv.org/pdf/1911.07171.pdf) - * 详细文档与开源的模型可以参考：[OIDV5目标检测github地址](https://github.com/PaddlePaddle/PaddleDetection/blob/master/docs/featured_model/OIDV5_BASELINE_MODEL.md) + * 详细文档与开源的模型可以参考：[OIDV5目标检测github地址](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.2/static/docs/featured_model/champion_model/OIDV5_BASELINE_MODEL.md) * 2019年Kaggle地标检索挑战赛亚军 * 技术报告可以参考：[https://arxiv.org/abs/1906.03990](https://arxiv.org/abs/1906.03990) diff --git a/docs/zh_CN_tmp/others/more_demo.md b/docs/zh_CN_tmp/others/more_demo.md index 4c3b4fa51e7448969bdc3b034779fc360037f88c..5971e7e3dd4b6065481725dbf953f46fb021b3ba 100644 --- a/docs/zh_CN_tmp/others/more_demo.md +++ b/docs/zh_CN_tmp/others/more_demo.md @@ -1,39 +1,39 @@ ## 识别效果展示 - 商品识别
- +
- +
- +
- +
- +
- +
- +
- +
- 动漫人物识别
- +
- +
- +
- logo识别 @@ -49,13 +49,13 @@ - 车辆识别
- +
- +
- +
[更多效果图](../../images/recognition/more_demo_images) diff --git a/docs/zh_CN_tmp/quick_start/.gitkeep b/docs/zh_CN_tmp/quick_start/.gitkeep deleted file mode 100644 index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000