updateed dataset features schema in README.md to match parquet files (2nd attempt)

README.md

@@ -29,11 +29,11 @@ dataset_info:
       dtype: string
     - name: text
       dtype: string
-    - name: source_type  
-      dtype: string     
+    - name: source_type
+      dtype: string
     - name: spans
       sequence:
-        feature: 
+        feature:
           annotator:
             dtype: string
           end:
@@ -44,15 +44,15 @@ dataset_info:
             dtype: int64
           text:
             dtype: string
-splits:
+  splits:
     - name: train
       num_examples: 19783
     - name: validation
       num_examples: 1795
     - name: test
       num_examples: 1844
-  download_size: TBD          
-  dataset_size:  TBD
+  download_size: TBD
+  dataset_size: TBD
 ---
 
 # ShamNER – Spoken Arabic Named‑Entity Recognition Corpus (Levantine v1.1)