Update app.py

app.py

@@ -4,10 +4,11 @@ import seaborn as sns
 import matplotlib.pyplot as plt
 from dataset_import import load_data
 from data_cleaning import clean_data
+from data_labelling import label_dataset
 
 st.title("AnalyzeYT Dataset Analysis")
 
-# Load dataset
+# Load and clean dataset
 data = load_data()
 
 if data is not None:
@@ -20,19 +21,12 @@ if data is not None:
     st.write("Cleaned Data Preview:")
     st.write(data.head())
     
-    # Show data description of cleaned data
-    st.write("Cleaned Data Description:")
+    # Show data description
+    st.write("Data Description:")
     st.write(data.describe())
 
-    # Add correlation matrix for cleaned data
-    st.write("Correlation Matrix of Cleaned Data:")
-    corr = data.corr()
-    st.write(corr)
-    sns.heatmap(corr, annot=True, cmap='coolwarm')
-    st.pyplot()
-
-    # Data visualization options for cleaned data
-    st.write("Data Visualization on Cleaned Data:")
+    # Data visualization options
+    st.write("Data Visualization:")
     chart_type = st.selectbox("Select Chart Type", ['Line Chart', 'Bar Chart', 'Histogram'])
 
     if chart_type == 'Line Chart':
@@ -58,3 +52,24 @@ if data is not None:
         plt.xlabel(selected_column)
         plt.ylabel('Frequency')
         st.pyplot()
+
+    # Add option for labeling the dataset
+    st.write("Labeling Options:")
+    label_option = st.radio("Do you want to label your dataset?", ('No', 'Yes'))
+
+    if label_option == 'Yes':
+        # Select columns for labeling
+        columns_to_label = st.multiselect("Select Columns to Label", data.columns)
+        
+        # Input new column names for labeled data
+        new_column_names = []
+        for col in columns_to_label:
+            new_name = st.text_input(f"Enter new name for labeled column '{col}':", f"{col}_labeled")
+            new_column_names.append(new_name)
+        
+        output_name = st.text_input("Enter Output File Name", "labeled_dataset.csv")
+        if st.button("Run Labeling"):
+            labeled_data = label_dataset(data, columns_to_label, new_column_names)
+            labeled_data.to_csv(output_name, index=False)
+            st.write("Labeling Completed. Download your file below:")
+            st.download_button(label="Download Labeled Dataset", data=labeled_data.to_csv(), file_name=output_name, mime='text/csv')