Update app.py

app.py

@@ -1,10 +1,3 @@
-# ================================
-# ENVIRONMENT (MUST BE FIRST)
-# ================================
-import os
-os.environ["KERAS_BACKEND"] = "tensorflow"
-os.environ["TF_ENABLE_ONEDNN_OPTS"] = "0"
-
 # ================================
 # IMPORTS
 # ================================
@@ -12,14 +5,14 @@ import gradio as gr
 import tensorflow as tf
 import numpy as np
 import json
-from PIL import Image
 from tensorflow.keras.preprocessing.text import tokenizer_from_json
+from PIL import Image
 
 # ================================
-# LOAD MODEL (.keras ONLY)
+# LOAD MODEL (TF 2.15 SAFE)
 # ================================
-fusion_model = tf.keras.models.load_model(
-    "fusion_ticket_model_final.keras",
+model = tf.keras.models.load_model(
+    "fusion_ticket_model_tf215.h5",
     compile=False
 )
 
@@ -29,12 +22,12 @@ print("✅ Fusion model loaded")
 # LOAD TOKENIZER (JSON – SAFE)
 # ================================
 with open("tokenizer.json", "r") as f:
-    tokenizer = tokenizer_from_json(f.read())
+    tokenizer = tokenizer_from_json(json.load(f))
 
 print("✅ Tokenizer loaded")
 
 # ================================
-# CONSTANTS (MUST MATCH TRAINING)
+# CONSTANTS (MATCH TRAINING)
 # ================================
 IMG_SIZE = (128, 128)
 MAX_LEN = 50
@@ -47,7 +40,8 @@ def preprocess_image(image: Image.Image):
     image = image.convert("RGB")
     image = image.resize(IMG_SIZE)
     img = np.array(image, dtype=np.float32) / 255.0
-    return np.expand_dims(img, axis=0)
+    img = np.expand_dims(img, axis=0)
+    return img
 
 # ================================
 # TEXT PREPROCESSING
@@ -56,9 +50,10 @@ def preprocess_text(text):
     if text is None:
         text = ""
     seq = tokenizer.texts_to_sequences([text])
-    return tf.keras.preprocessing.sequence.pad_sequences(
+    padded = tf.keras.preprocessing.sequence.pad_sequences(
         seq, maxlen=MAX_LEN
     )
+    return padded
 
 # ================================
 # PREDICTION FUNCTION
@@ -75,7 +70,7 @@ def predict_ticket(image, text):
     img = preprocess_image(image)
     txt = preprocess_text(text)
 
-    probs = fusion_model.predict([img, txt], verbose=0)[0]
+    probs = model.predict([img, txt], verbose=0)[0]
 
     return {
         "Critical": float(probs[0]),
@@ -93,15 +88,16 @@ interface = gr.Interface(
         gr.Image(type="pil", label="📤 Upload Ticket Screenshot"),
         gr.Textbox(
             lines=4,
-            placeholder="Describe the issue (recommended)",
+            placeholder="Describe the issue (optional but improves accuracy)",
             label="✍️ Ticket Description"
         )
     ],
     outputs=gr.Label(num_top_classes=4, label="🚨 Predicted Severity"),
     title="🎫 Ticket Severity Classification System",
     description=(
-        "CNN + NLP **Fusion Model** that predicts ticket urgency\n\n"
-        "**Classes:** Critical | High | Medium | Low"
+        "This system uses a **CNN + NLP Fusion Model** trained on "
+        "ticket screenshots and descriptions.\n\n"
+        "**Severity Levels:** Critical | High | Medium | Low"
     )
 )