Update README.md

README.md

@@ -147,35 +147,6 @@ val = ds_multi["validation"]
 test_engu = ds_multi["test_engu"]  # per-language test splits
 ```
 
-If remote code is disallowed, use the bundled parquet files instead of the TSVs:
-```python
-from datasets import load_dataset
-# single pair example
-ds = load_dataset(
-    "parquet",
-    data_files={
-        "train": "parquet/engu/train.parquet",
-        "validation": "parquet/engu/dev.parquet",
-        "test": "parquet/engu/test.parquet",
-    },
-)
-# multilingual example
-ds_multi = load_dataset(
-    "parquet",
-    data_files={
-        "train": "parquet/multilingual/train.parquet",
-        "validation": "parquet/multilingual/dev.parquet",
-        "test_engu": "parquet/multilingual/test_engu.parquet",
-        "test_enhi": "parquet/multilingual/test_enhi.parquet",
-        "test_enmr": "parquet/multilingual/test_enmr.parquet",
-        "test_enta": "parquet/multilingual/test_enta.parquet",
-        "test_ente": "parquet/multilingual/test_ente.parquet",
-        "test_eten": "parquet/multilingual/test_eten.parquet",
-        "test_neen": "parquet/multilingual/test_neen.parquet",
-        "test_sien": "parquet/multilingual/test_sien.parquet",
-    },
-)
-```
 
 ## Data Statistics (DA mean, z_mean)