Update dataset usage

README.md

@@ -54,10 +54,22 @@ While `kn` translations are available in the NLLB dataset, their quality is poor
 ```python
 import pandas as pd
 
-# Load individual splits
-train_df = pd.read_csv('train.csv')
-test_df = pd.read_csv('test.csv')
-valid_df = pd.read_csv('validation.csv')
+splits = {
+   'train': 'data/train-00000-of-00001.parquet',
+   'validation': 'data/validation-00000-of-00001.parquet', 
+   'test': 'data/test-00000-of-00001.parquet'
+}
+
+# Load all splits into DataFrames
+dataframes = {}
+for split, path in splits.items():
+   dataframes[split] = pd.read_parquet(f"hf://datasets/pavan-naik/mt-nllb-en-kn/{path}")
+
+# Access individual splits
+train_data = dataframes['train']
+test_data = dataframes['test']
+validation_data = dataframes['validation']
+
 ```
 
 ### Using HuggingFace 🤗 Datasets
@@ -72,8 +84,6 @@ train_data = dataset["train"]
 test_data = dataset["test"]
 validation_data = dataset["validation"]
 
-# Example usage
-print(train_data[0])  # Print first example
 ```