Fix readme

README.md

@@ -126,7 +126,7 @@ The dataset is split into `train`, `validation` and `test` sets. In the followin
 | :-----------: | :-------------------: | :-------------------: |
 | `train`       |    11471              |    11090              |
 | `validation`  |    145                |    144                |
-| `test`        |    1000               |    970                |
+| `test`        |    1000               |    976                |
 
 ### Dataset Creation