Update README.md

README.md

@@ -27,12 +27,7 @@ Each example includes: the NLQ, database identifier, a canonical dataset id, the
 
 ---
 
-## What’s inside
-
-### Labels
-- **4-class:** `Table`, `Column`, `Value`, `O`  
-
-### Fields per example
+## Fields
 - `question_id` *(int)* — Example id
 - `db_id` *(str)* — Database identifier
 - `dber_id` *(str)* — Canonical id linking back to the source file/split (BIRD, SPIDER)
@@ -52,14 +47,6 @@ Each example includes: the NLQ, database identifier, a canonical dataset id, the
 
 ## Splits
 
-### Split groups
-
-- Human
-  - Human_train (`human_train`)
-  - Human_test (`human_test`)
-- Synthetic
-  - Synthetic_train (`synthetic_train`)
-
 **Entity token prevalence is consistent across splits: ~29% entity vs. ~71% `O`.**
 
 | Split             | # Examples |
@@ -110,10 +97,16 @@ Each example includes: the NLQ, database identifier, a canonical dataset id, the
 }
 ```
 
----
+<!-- ---
 
 ## Usage
 
+### Load from Hub
+```python
+from datasets import load_dataset
+ds = load_dataset("Voice49/dber")
+```
+
 ### Load JSONL files
 ```python
 from datasets import load_dataset
@@ -125,14 +118,7 @@ data_files = {
 }
 ds = load_dataset("json", data_files=data_files)
 print(ds)
-print(ds["human_train"][0])
-```
-
-### Load from the Hub
-```python
-from datasets import load_dataset
-ds = load_dataset("Voice49/dber")
-```
+``` -->
 
 ---