diff --git a/docs/DATASET.md b/docs/DATASET.md
index 7c5a84a1d0eaa66ffee8600a373d54d54b86b462..1ae8bac8b2a3b20f5be8ea6c9f1f0abd70453245 100644
--- a/docs/DATASET.md
+++ b/docs/DATASET.md
@@ -15,8 +15,9 @@ Airborne objects usually appear quite small at the distances which are relevant
 ## Accessing Training Dataset
 
 The complete training dataset size is ~>11TB.
+You can also download partial dataset (500G) using `partial=True` flag in `Dataset`. It includes all the frames with valid encounter of planned airborne object.
 
-You can access the same in public S3 bucket hosted at `s3://airborne-obj-detection-challenge-training/`.
+You can access the dataset in public S3 bucket hosted at `s3://airborne-obj-detection-challenge-training/`.
 
 **In order to ease access to you as participant, we have added some helper scripts which will help you download the dataset on need basis.
 You can simply load all the ground_truth files and download dataset flight by flight.**
@@ -121,9 +122,17 @@ This library will provide you quick access and download only the files you requi
 
 ```python
 from core.dataset import Dataset
-dataset = Dataset(local_path='data/part1', s3_path='s3://airborne-obj-detection-challenge-training/part1/')
-dataset.add(local_path='data/part2', s3_path='s3://airborne-obj-detection-challenge-training/part2/')
-dataset.add(local_path='data/part3', s3_path='s3://airborne-obj-detection-challenge-training/part3/')
+dataset = Dataset(local_path='data/part1', s3_path='s3://airborne-obj-detection-challenge-training/part1/', prefix='part1')
+dataset.add(local_path='data/part2', s3_path='s3://airborne-obj-detection-challenge-training/part2/', prefix='part2')
+dataset.add(local_path='data/part3', s3_path='s3://airborne-obj-detection-challenge-training/part3/', prefix='part3')
+```
+
+Example for partial dataset:
+```python
+from core.dataset import Dataset
+dataset = Dataset(local_path='data/part1', s3_path='s3://airborne-obj-detection-challenge-training/part1/', prefix='part1', partial=True)
+dataset.add(local_path='data/part2', s3_path='s3://airborne-obj-detection-challenge-training/part2/', prefix='part2', partial=True)
+dataset.add(local_path='data/part3', s3_path='s3://airborne-obj-detection-challenge-training/part3/', prefix='part3', partial=True)
 ```
 
 NOTE: You don't need to have `groundtruth.json` pre-downloaded, it will automatically download, save and load them for you.