Update

README.md

@@ -51,8 +51,10 @@ model = AutoModelForSequenceClassification.from_pretrained("LiYuan/amazon-query-
 
 Download all the raw [dataset](https://www.aicrowd.com/challenges/esci-challenge-for-improving-product-search/dataset_files) from the Amazon KDD Cup website.
 
-The three dataset we used like below:
-
+1. Concatenate the all product attributes from the product dataset
+2. Join it with a training query dataset
+3. Stratified Split the merged data into 571,223-row training, 142,806-row validation, 79,337-row test set
+4. Train on the full training set
 
 
 ## Training procedure