diff --git a/README.md b/README.md
index 6fdd06f..071c056 100644
--- a/README.md
+++ b/README.md
@@ -121,15 +121,17 @@ If you want to fine tune this operator, make sure your timm version is 0.4.12, a
 ! python -m pip install tqdm augly timm==0.4.12 pytorch-metric-learning==0.9.99
 ```
 
+### Download dataset
+Your need to download dataset from [Facebook AI Image Similarity Challenge: Descriptor Track](https://www.drivendata.org/competitions/79/competition-image-similarity-1-dev/), and use the `./training_images` folder as your training images root. It requires about 165G space.
 
-### Get start to fine-tune
+### Get started to fine-tune
 Just call method op.train() and pass in your args.
 ```python
 import towhee
 
 op = towhee.ops.image_embedding.isc().get_op()
 op.train(training_args={
-        'train_data_dir': './your_training_images',
+        'train_data_dir': './training_images',
         'distributed': False,
         'output_dir': './output',
         'gpu': 0,