Update app.py

app.py

@@ -107,13 +107,8 @@ def preprocess_data(df):
 
     return df
 
-def preprocess_csv_data(df):
-    df = df[df['Answer'] == 'Fallback Message shown']
-    df.rename(columns={'Question': 'texts'}, inplace=True)
-    df['texts'] = df['texts'].astype(str)
-    return preprocess_data(df)
-
-def cluster_data(df, num_clusters):
+def cluster_data(df):
+    num_clusters = 15  # Set the number of clusters to 15
     vectorizer = TfidfVectorizer(stop_words='english')
     X = vectorizer.fit_transform(df['texts'])
 
@@ -139,18 +134,26 @@ def visualize_clusters(df):
 
 def main(file, num_clusters_to_display):
     try:
-        file_ext = file.name.split('.')[-1].lower()
-        if file_ext == 'xlsx':
+        # Determine the file type
+        file_extension = file.name.split('.')[-1].lower()
+        
+        # Load the file
+        if file_extension == 'xlsx':
             df = pd.read_excel(file)
-            df = preprocess_data(df)
-        elif file_ext == 'csv':
+        elif file_extension == 'csv':
             df = pd.read_csv(file)
-            df = preprocess_csv_data(df)
         else:
-            return "Unsupported file format. Please upload an Excel (.xlsx) or CSV (.csv) file."
-
-        num_clusters = 10  # Set the number of clusters
-        df = cluster_data(df, num_clusters)
+            return "Unsupported file type. Please upload an Excel or CSV file."
+        
+        # Process CSV file specifically
+        if file_extension == 'csv':
+            # Keep only rows where 'Answer' is 'Fallback Message shown'
+            df = df[df['Answer'] == 'Fallback Message shown']
+            # Focus on 'Query' column for text processing
+            df.rename(columns={'Query': 'texts'}, inplace=True)
+        
+        df = preprocess_data(df)
+        df = cluster_data(df)
         visualize_clusters(df)
 
         cluster_sizes = df['Cluster'].value_counts()
@@ -166,21 +169,31 @@ def main(file, num_clusters_to_display):
         df['Cluster'] = pd.Categorical(df['Cluster'], categories=top_clusters, ordered=True)
         df = df.sort_values('Cluster')
 
+        # Save the resulting DataFrame to a CSV file
         with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as tmpfile:
             df.to_csv(tmpfile.name, index=False)
+            tmpfile.flush()
+
             return tmpfile.name
+
     except Exception as e:
-        return str(e)
+        return f"An error occurred: {str(e)}"
+
+def upload_file(file, num_clusters_to_display):
+    result = main(file, num_clusters_to_display)
+    if result.endswith(".csv"):
+        return result
+    else:
+        return f"Error: {result}"
 
 interface = gr.Interface(
-    fn=main,
+    fn=upload_file,
     inputs=[
-        gr.File(label="Upload Excel or CSV File (.xlsx or .csv)"),
-        gr.Slider(1, 10, step=1, label="Number of Categories to Display")
+        gr.inputs.File(label="Upload Excel or CSV File (.xlsx or .csv)", type="file"),
+        gr.inputs.Slider(minimum=1, maximum=20, step=1, default=5, label="Number of Categories to Display")
     ],
-    outputs=gr.File(label="Clustered Data CSV"),
-    title="Unanswered User Queries Clustering",
-    description="Upload an Excel or CSV file and select the number of largest clusters to display (excluding cluster 0)"
+    outputs=gr.outputs.File(label="Output CSV File"),
+    title="Unanswered User Queries Clustering"
 )
 
-interface.launch()
+interface.launch(debug=True)