Update app.py

app.py

@@ -18,6 +18,7 @@ def main():
     import eli5
     import shap
     from custom_models import HF_DistilBertBasedModelAppDocs, HF_BertBasedModelAppDocs
+    import docx
 
     # Initialize Spacy
     nlp = spacy.load("en_core_web_sm")
@@ -92,8 +93,24 @@ def main():
         else:
             model = HF_DistilBertBasedModelAppDocs.from_pretrained("ferdmartin/HF_DistilBertBasedModelAppDocs").to(device)
         return model
-    
 
+    def app_model(option, model, text):
+        with st.spinner("Wait for the magic 🪄🔮"):
+            # Use model
+            if option in ("Naive Bayes", "Logistic Regression"):
+                prediction, predict_proba = nb_lr(model, text)
+                st.session_state["sklearn"] = True
+            else:
+                prediction, predict_proba = torch_pred(tokenizer, model, text)
+                st.session_state["torch"] = True
+    
+            # Store the result in session state
+            st.session_state["color_pred"] = "blue" if prediction == 0 else "red"
+            prediction = pred_str(prediction)
+            st.session_state["prediction"] = prediction
+            st.session_state["predict_proba"] = predict_proba
+            st.session_state["text"] = text
+            
     # Streamlit app:
 
     models_available = {"Logistic Regression":"models/baseline_model_lr2.joblib", 
@@ -102,9 +119,9 @@ def main():
                         "BERT-based model": "ferdmartin/HF_BertBasedModelAppDocs"
                         }
 
-    st.set_page_config(page_title="AI/Human GradAppDocs", page_icon="📝🦾", layout="wide")
+    st.set_page_config(page_title="AI/Human GradAppDocs", page_icon="🤖", layout="wide")
     st.title("Academic Application Document Classifier")
-    st.header("Is it human-made 📝 or Generated with AI 🤖 ?  ")
+    st.header("Is it human-made 📝 or Generated with AI 🦾 ?  ")
     
     # Check the model to use
     def restore_prediction_state():
@@ -131,32 +148,19 @@ def main():
             </style>
             """
     st.markdown(hide_st_style, unsafe_allow_html=True)
-
+    col1, col2 = st.columns(2)
     # Use model
-    if st.button("Let's check this text!"):
-        if text.strip() == "":
-            st.error("Please enter some text")
+    with col1:
+        if st.button("Let's check this text!"):
+            if text.strip() == "":
+                st.error("Please enter some text")
         else:
-            with st.spinner("Wait for the magic 🪄🔮"):
-                # Use model
-                if option in ("Naive Bayes", "Logistic Regression"):
-                    prediction, predict_proba = nb_lr(model, text)
-                    st.session_state["sklearn"] = True
-                else:
-                    prediction, predict_proba = torch_pred(tokenizer, model, text)
-                    st.session_state["torch"] = True
-
-            # Store the result in session state
-            st.session_state["color_pred"] = "blue" if prediction == 0 else "red"
-            prediction = pred_str(prediction)
-            st.session_state["prediction"] = prediction
-            st.session_state["predict_proba"] = predict_proba
-            st.session_state["text"] = text
-            
+        
+            app_model(option, model, text)
             # Print result
-            st.markdown(f"I think this text is: **:{st.session_state['color_pred']}[{st.session_state['prediction']}]** (Prediction probability: {st.session_state['predict_proba'] * 100}%)")
-
-    elif "prediction" in st.session_state:
+            #st.markdown(f"I think this text is: **:{st.session_state['color_pred']}[{st.session_state['prediction']}]** (Prediction probability: {st.session_state['predict_proba'] * 100}%)")
+   
+    if "prediction" in st.session_state:
         # Display the stored result if available        
         st.markdown(f"I think this text is: **:{st.session_state['color_pred']}[{st.session_state['prediction']}]** (Prediction probability: {st.session_state['predict_proba'] * 100}%)")