Update app.py

app.py

@@ -1,23 +1,22 @@
 import gradio as gr
-from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+from keras.models import load_model
+from keras.preprocessing.sequence import pad_sequences
 
-# Load your custom Hugging Face model
-model_name = "models/Bajiyo/Named_entity_transliteration_malayalam"
-model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+# Load your custom Keras model
+model = load_model('/content/drive/MyDrive/best_model.h5')
 
 # Function for transliteration
 def transliterate_malayalam_to_english(malayalam_text):
-    # Tokenize the input
-    input_ids = tokenizer.encode(malayalam_text, return_tensors="pt", max_length=50, truncation=True)
-
+    # Tokenize and preprocess the input (adjust this based on your preprocessing logic)
+    input_sequence = pad_sequences(tokenizer.texts_to_sequences([malayalam_text]), maxlen=max_seq_length, padding='post')
+    
     # Use the model for prediction
-    output_ids = model.generate(input_ids, max_length=49, num_beams=5)
+    output_sequence = model.predict(input_sequence)
     
-    # Decode the output sequence
-    output_sequence = tokenizer.decode(output_ids[0], skip_special_tokens=True)
+    # Decode the output sequence to get the transliterated text (adjust this based on your decoding logic)
+    predicted_text = decode_output_sequence(output_sequence)
     
-    return output_sequence
+    return predicted_text
 
 # Create a Gradio interface
 iface = gr.Interface(
@@ -28,4 +27,4 @@ iface = gr.Interface(
 )
 
 # Launch the Gradio interface
-iface.launch()
+iface.launch()