fix: csv reading

data/spam-text-messages-dataset.csv

@@ -12,7 +12,7 @@ image,text
 10.jpg,"Discount of Rs 100 is now YOURS! Get 126 Channels, worth Rs 319 in just Rs 219 on your Airtel Digital TV account 3037252019 ! Limited Period offer, avail now!"
 11.jpg,"Ride & save with 25% off 5 GO or GO Awfar rides up to 20 EGP off valid for a limited time. Use code JN25 https://bit.ly/3F9vuMi"
 12.jpg,"Study in UK, USA, Canada, Australia, Malaysia, China, Europe & South Korea  World top ten Univ. offers: MBBS, all type of Engineering & Business related subjects   Service charge free till 20 April'23  Our expert team will support for your admission & Visa processing  01720557119  01720557126  01720557134"
-13.jpg,"Ramzan Mubarak! Pakistan Day Gala ends 26th March. Hurry up! shop FLAT 23% & 40% OFF before stock runs out. In-stores & online: https://bit.ly/3Yt8PDI T&Cs apply  To block promotional SMS from the J . Type """"UNSUB"""" and send it to 2223623. To block all promotional SMS: Type """"REG"""" and send it to """"3627"""""
+13.jpg,"Ramzan Mubarak! Pakistan Day Gala ends 26th March. Hurry up! shop FLAT 23% & 40% OFF before stock runs out. In-stores & online: https://bit.ly/3Yt8PDI T&Cs apply  To block promotional SMS from the J . Type """"UNSUB"""" and send it to 2223623. To block all promotional SMS: Type """"REG"""" and send it to """"3627""""""
 14.png,"Get a SPECIAL DATA DEAL of 2GB for sh. 100 valid 24hrs by clicking on https://bit.ly/Safaricomapp or Dial *544*21# and keep the connections going . "
 15.jpg,"JAZA POCHI SAA HII!  SHINDA THAO TANO EVERY HOUR  PLAY NOW> https://aviator254.ke/  STEPS NI TATU TU JOIN, BECOME THE BIG CRASHER, WIN!  STOP *456*9*5#"
 16.jpg,"RAILWAY FURNISHERS*Don't Laybuy It - Railway IT! Open your 24 month account and get your goods upon approval. NO interest! Visit a branch TODAY! STOP2OPTOUT"

spam-text-messages-dataset.py

@@ -53,7 +53,7 @@ class SpamTextMessagesDataset(datasets.GeneratorBasedBuilder):
         ]
 
     def _generate_examples(self, images, annotations):
-        annotations_df = pd.read_csv(annotations)
+        annotations_df = pd.read_csv(annotations, engine='python')
 
         for idx, (image_path, image) in enumerate(images):
             yield idx, {