UPDATE FILES

app.py

@@ -4,6 +4,7 @@ import torch.nn.functional as F
 from transformers import BertTokenizer
 import gradio as gr
 import math
+from transformers import MarianTokenizer
 
 # Define your model architecture here (copied from your previous definition)
 class MultiHeadAttention(nn.Module):
@@ -155,7 +156,7 @@ class Transformer(nn.Module):
 
 # Load tokenizer and model
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-tokenizer = BertTokenizer.from_pretrained("tokenizer")
+tokenizer = MarianTokenizer.from_pretrained("tokenizer")
 
 encoder = Encoder(
     vocab_size=tokenizer.vocab_size + 1,