Update README.md

README.md

@@ -1,5 +1,7 @@
 ---
 dataset_info:
+
+- config_name: passages (subset)
   features:
   - name: pid
     dtype: int64
@@ -11,6 +13,29 @@ dataset_info:
     num_examples: 718217
   download_size: 154088206
   dataset_size: 249167997
+  
+- config_name: queries
+  features:
+  - name: qid
+    dtype: int64
+  - name: text
+    dtype: string
+  splits:
+  - name: train
+  
+- config_name: qrels
+  features:
+  - name: qid
+    dtype: int64
+  - name: qrank
+    dtype: int64
+  - name: pid
+    dtype: int64
+  - name: prank
+    dtype: int64
+  splits:
+  - name: train
+
 configs:
 # - config_name: default
 #   data_files:
@@ -20,24 +45,17 @@ configs:
   data_files:
   - split: train
     path: data/collection.subset.tr.tsv
-    features:
-    - name: pid
-      dtype: int64
-    - name: text
-      dtype: string
+    
 - config_name: queries
   data_files:
   - split: train
     path: data/queries.train.tr.tsv
-    features:
-    - name: qid
-      dtype: int64
-    - name: text
-      dtype: string
+    
 - config_name: qrels
   data_files:
   - split: train
     path: data/qrels.train.tsv
+    
 license: apache-2.0
 task_categories:
 - text-retrieval
@@ -51,6 +69,8 @@ tags:
 - passage-ranking
 - colbert
 pretty_name: MS Marco - TR
+size_categories:
+- 100K<n<1M
 ---
 # Dataset Card for "msmarco-tr"