diff --git a/cv/classification/acmix/pytorch/README.md b/cv/classification/acmix/pytorch/README.md
index 6836fc1afcf9ad5a3538bd008b5fd6116c4f3bf9..b7f5e724480dc2817883180c8c401faed58e963c 100644
--- a/cv/classification/acmix/pytorch/README.md
+++ b/cv/classification/acmix/pytorch/README.md
@@ -9,6 +9,7 @@ Convolution and self-attention are two powerful techniques for representation le
 pip install termcolor==1.1.0 yacs==0.1.8 timm==0.4.5
 ```
 
+Sign up and login in [imagenet official website](https://www.image-net.org/index.php), then choose 'Download' to download the whole imagenet dataset. Specify `/path/to/imagenet` to your imagenet path in later training process.
 
 ## Step 2: Training
 
diff --git a/cv/classification/convnext/pytorch/README.md b/cv/classification/convnext/pytorch/README.md
index 16e6232d573fad7770e531a767a508fc5a27d65f..f193f7a4f421754146494896ca9c658e67884d5c 100644
--- a/cv/classification/convnext/pytorch/README.md
+++ b/cv/classification/convnext/pytorch/README.md
@@ -8,6 +8,9 @@ The ConvNeXT model was proposed in [A ConvNet for the 2020s](https://arxiv.org/a
 pip install timm==0.3.2 tensorboardX six torch torchvision
 ```
 
+Sign up and login in [imagenet official website](https://www.image-net.org/index.php), then choose 'Download' to download the whole imagenet dataset. Specify `/path/to/imagenet` to your imagenet path in later training process.
+
+
 ## Step 2: Training
 ### Multiple GPUs on one machine
 ```bash
diff --git a/cv/detection/retinanet/pytorch/README.md b/cv/detection/retinanet/pytorch/README.md
index 0512a21a3ab244f69f4e10c28e00884c37e775af..e4c5838ef050fe76784adb42433a572c4b9a3dd7 100644
--- a/cv/detection/retinanet/pytorch/README.md
+++ b/cv/detection/retinanet/pytorch/README.md
@@ -14,6 +14,8 @@ pip3 install 'scipy' 'matplotlib' 'pycocotools' 'opencv-python' 'easydict' 'tqdm
 
 ## Step 2: Training on COCO dataset
 
+Download the [COCO Dataset](https://cocodataset.org/#home) 
+
 ### Multiple GPUs on one machine
 
 ```shell