final

app.py

@@ -0,0 +1,72 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+from sklearn.preprocessing import StandardScaler
+from sklearn.cluster import DBSCAN
+
+st.title("Music Genre Clustering with DBSCAN")
+
+# Load dataset directly
+file_path = "top_10000_1950-now.csv"
+df = pd.read_csv(file_path)
+
+# Remove non-numeric columns
+df_numeric = df.select_dtypes(include=[np.number])
+
+# Create tabs
+tab1, tab2, tab3 = st.tabs(["Overview", "Visualization Matrix", "User Input"])
+
+with tab1:
+    st.write("### Dataset Overview")
+    st.dataframe(df.head())
+    st.write("### Dataset Information")
+    st.write(df_numeric.describe())
+
+with tab2:
+    st.write("### Correlation Matrix")
+    plt.figure(figsize=(10, 6))
+    sns.heatmap(df_numeric.corr(), annot=True, cmap="coolwarm", fmt=".2f")
+    st.pyplot(plt)
+    
+    st.write("### Pairplot Visualization")
+    pairplot_features = st.multiselect("Select Features for Pairplot", df_numeric.columns.tolist(),
+                                       default=["Danceability", "Energy", "Tempo", "Loudness", "Valence"])
+    if pairplot_features:
+        sns.pairplot(df[pairplot_features])
+        st.pyplot(plt)
+
+with tab3:
+    st.write("### Clustering Settings")
+    num_features = st.slider("Select Number of Features", 2, len(df_numeric.columns), 5)
+    features = st.multiselect("Select Features for Clustering", 
+                              df_numeric.columns.tolist(),
+                              default=df_numeric.columns[:num_features])
+    
+    if st.button("Run Clustering"):
+        if len(features) >= 2:
+            df_filtered = df_numeric[features].dropna()
+            X_scaled = StandardScaler().fit_transform(df_filtered)
+            
+            eps = 1.0  # Default value, can be modified as needed
+            min_samples = 10  # Default value, can be modified as needed
+            
+            dbscan = DBSCAN(eps=eps, min_samples=min_samples)
+            labels = dbscan.fit_predict(X_scaled)
+            
+            df_filtered["Cluster"] = labels
+            df["Cluster"] = np.nan
+            df.loc[df_filtered.index, "Cluster"] = labels
+            
+            st.write("### Clustered Data:")
+            st.dataframe(df[["Track Name", "Artist Name(s)", "Cluster"]].dropna().head(20))
+            
+            st.write("### Cluster Visualization:")
+            fig, ax = plt.subplots()
+            scatter = ax.scatter(X_scaled[:, 0], X_scaled[:, 1], c=labels, cmap="viridis", alpha=0.7)
+            legend1 = ax.legend(*scatter.legend_elements(), title="Clusters")
+            ax.add_artist(legend1)
+            st.pyplot(fig)
+        else:
+            st.warning("Please select at least two features for clustering.")

requirements.txt

@@ -0,0 +1,6 @@
+streamlit
+panda
+numpy
+matplotlib
+seaborn
+scikit-learn