Update app.py

app.py

@@ -1,19 +1,34 @@
 from transformers import AutoModel, AutoTokenizer
 from flask import Flask, request, jsonify
+import tensorflow as tf
 
 app = Flask(__name__)
 
-# Load model and tokenizer from Hugging Face Hub
+# Load Hugging Face model and tokenizer
 tokenizer = AutoTokenizer.from_pretrained("Erfan11/Neuracraft", use_auth_token="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")
-model = AutoModel.from_pretrained("Erfan11/Neuracraft", use_auth_token="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")
+hf_model = AutoModel.from_pretrained("Erfan11/Neuracraft", use_auth_token="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")
+
+# Load TensorFlow model
+tf_model = tf.keras.models.load_model('path_to_your_tf_model.h5')
 
 @app.route('/predict', methods=['POST'])
 def predict():
     data = request.get_json()
+    
+    # Tokenize the input using Hugging Face's tokenizer
     inputs = tokenizer(data["text"], return_tensors="pt")
-    outputs = model(**inputs)
-    # Process your model's output as needed
-    return jsonify(outputs)
+    
+    # Make prediction with Hugging Face model
+    hf_outputs = hf_model(**inputs)
+    
+    # Optionally: You can also add TensorFlow model predictions here, depending on what it’s used for.
+    # Assuming the TensorFlow model is used for something else like feature extraction
+    tf_outputs = tf_model.predict([data["text"]])  # Modify based on your input processing
+
+    return jsonify({
+        "hf_outputs": hf_outputs[0].tolist(),  # Convert Hugging Face output to JSON serializable format
+        "tf_outputs": tf_outputs.tolist()  # Convert TensorFlow output to JSON serializable format
+    })
 
 if __name__ == '__main__':
     app.run(debug=True)