Update app.py

app.py

@@ -1,72 +1,88 @@
 import gradio as gr
 import pandas as pd
-from sklearn.pipeline import Pipeline
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.linear_model import LogisticRegression
-import tempfile
 
 df_train = None
 model = None
+vectorizer = None
 
 def load_training_file(file):
     global df_train
     if file is None:
         return "❌ Please upload a file.", gr.update(choices=[], value=None), gr.update(choices=[], value=None)
-    
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".xlsx") as tmp:
-        tmp.write(file.read_bytes())
-        tmp_path = tmp.name
 
-    df_train = pd.read_excel(tmp_path)
+    df_train = pd.read_excel(file.name)
     col_names = list(df_train.columns)
+
     return f"✅ Loaded file with {len(df_train)} rows", gr.update(choices=col_names, value=col_names[0]), gr.update(choices=col_names, value=col_names[-1])
 
 def train_model(text_column, target_column):
-    global model, df_train
+    global model, vectorizer
+
     if df_train is None:
-        return "⚠️ Please load a training file first."
+        return "❌ No training data loaded."
 
     if text_column not in df_train.columns or target_column not in df_train.columns:
-        return "❌ Selected columns not found in the data."
+        return "❌ Invalid column selection."
 
     df_filtered = df_train.dropna(subset=[text_column, target_column])
-    if df_filtered.empty:
-        return "❌ No valid data after dropping missing values."
 
-    X = df_filtered[text_column]
+    vectorizer = TfidfVectorizer()
+    X = vectorizer.fit_transform(df_filtered[text_column])
     y = df_filtered[target_column]
 
-    model = Pipeline([
-        ("tfidf", TfidfVectorizer()),
-        ("clf", LogisticRegression(max_iter=1000))
-    ])
+    model = LogisticRegression(max_iter=1000)
     model.fit(X, y)
-    return f"✅ Model trained with {len(X)} samples."
 
-def predict(text):
-    if model is None:
-        return "⚠️ Please train the model first."
-    return model.predict([text])[0]
+    return f"✅ Model trained on {len(df_filtered)} examples."
+
+def predict_label(text_input):
+    if model is None or vectorizer is None:
+        return "❌ Model is not trained yet."
+
+    X = vectorizer.transform([text_input])
+    prediction = model.predict(X)[0]
+    return f"🔮 Prediction: {prediction}"
 
 with gr.Blocks() as demo:
-    gr.Markdown("## 🧠 Text Classification Trainer")
-    
+    gr.Markdown("# 🧠 Text Classification App")
+
     with gr.Row():
-        training_file = gr.File(label="Upload Excel file (.xlsx)")
-        status = gr.Textbox(label="Status", interactive=False)
+        file_input = gr.File(label="Upload Excel File (.xlsx)", file_types=[".xlsx"])
+        load_button = gr.Button("📂 Load File")
 
+    status_output = gr.Markdown()
     with gr.Row():
-        text_column = gr.Dropdown(choices=[], label="Select Text Column")
-        target_column = gr.Dropdown(choices=[], label="Select Target Column")
-        train_btn = gr.Button("Train Model")
+        text_column_dropdown = gr.Dropdown(label="Text column")
+        target_column_dropdown = gr.Dropdown(label="Target column")
+
+    train_button = gr.Button("🚀 Train Model")
+    training_status = gr.Markdown()
 
     with gr.Row():
-        input_text = gr.Textbox(label="Enter text to predict")
-        output_label = gr.Textbox(label="Predicted Label", interactive=False)
-        predict_btn = gr.Button("Predict")
+        input_text = gr.Textbox(label="Enter text to classify")
+        predict_button = gr.Button("🔍 Predict")
+
+    prediction_output = gr.Markdown()
+
+    load_button.click(
+        fn=load_training_file,
+        inputs=file_input,
+        outputs=[status_output, text_column_dropdown, target_column_dropdown]
+    )
+
+    train_button.click(
+        fn=train_model,
+        inputs=[text_column_dropdown, target_column_dropdown],
+        outputs=training_status
+    )
 
-    training_file.change(fn=load_training_file, inputs=[training_file], outputs=[status, text_column, target_column])
-    train_btn.click(fn=train_model, inputs=[text_column, target_column], outputs=[status])
-    predict_btn.click(fn=predict, inputs=[input_text], outputs=[output_label])
+    predict_button.click(
+        fn=predict_label,
+        inputs=input_text,
+        outputs=prediction_output
+    )
 
-demo.launch()
+if __name__ == "__main__":
+    demo.launch()