Update app.py

app.py

@@ -9,7 +9,7 @@ dataset_names = {
     'ag_news': None,
     'squad': None,
     'cnn_dailymail': '1.0.0',  # Specify configuration for cnn_dailymail
-    'wiki40b': 'ru'  # Specify language for wiki40b
+    'wiki40b': 'en'  # Specify language for wiki40b
 }
 
 # Global variables for model and tokenizer
@@ -32,6 +32,7 @@ def load_and_prepare_datasets():
     
     for ds in datasets:
         if 'train' in ds:
+            # Extract text field based on available keys
             if 'text' in ds['train'].features:
                 train_datasets.append(ds['train'].map(lambda x: {'text': x['text']}))
             elif 'content' in ds['train'].features:  # Example for CNN/DailyMail
@@ -40,6 +41,7 @@ def load_and_prepare_datasets():
                 print(f"Warning: No suitable text field found in {ds['train'].features}")
         
         if 'test' in ds:
+            # Extract text field based on available keys
             if 'text' in ds['test'].features:
                 eval_datasets.append(ds['test'].map(lambda x: {'text': x['text']}))
             elif 'content' in ds['test'].features:  # Example for CNN/DailyMail