app.py

import streamlit as st
import pandas as pd
import joblib
import plotly.express as px

# -------------------------------
# 1️⃣ Dosyaları yükle
# -------------------------------
DATA_PATH = "spotify_clustered.csv"
MODEL_PATH = "kmeans_music_model.pkl"
SCALER_PATH = "scaler_music.pkl"

@st.cache_data
def load_data():
    df = pd.read_csv(DATA_PATH)
    unnecessary_cols = ['Unnamed: 0', 'track_id', 'album_name', 'explicit']
    df_display = df.drop(columns=[c for c in unnecessary_cols if c in df.columns])
    return df, df_display

@st.cache_data
def load_model_scaler():
    model = joblib.load(MODEL_PATH)
    scaler = joblib.load(SCALER_PATH)
    return model, scaler

df, df_display = load_data()
model, scaler = load_model_scaler()

# -------------------------------
# 2️⃣ Sayfa ayarları
# -------------------------------
st.set_page_config(page_title="Spotify Clusters", layout="wide")
st.title("🎵 Spotify Music Clustering / Spotify Müzik Kümeleme")
st.markdown("---")

# -------------------------------
# 3️⃣ Veri önizleme
# -------------------------------
st.subheader("📄 Dataset Preview / Veri Önizleme")
st.dataframe(df_display.head(10), use_container_width=True)

# -------------------------------
# 4️⃣ Titremesiz grafik – Plotly
# -------------------------------
st.subheader("🎯 Feature Analysis / Özellik Analizi")
fig = px.scatter(
    df,
    x='danceability',
    y='energy',
    color='cluster',
    labels={'danceability': 'Danceability / Dans Edilebilirlik', 
            'energy': 'Energy / Enerji', 
            'cluster': 'Cluster / Küme'},
    opacity=0.6
)
st.plotly_chart(fig, use_container_width=True)

# -------------------------------
# 5️⃣ Prediction Section / Tahmin
# -------------------------------
st.divider()
st.subheader("🤖 Predict New Song Cluster / Yeni Şarkı Tahmini")
c1, c2, c3 = st.columns(3)
with c1:
    pop = st.slider("Popularity", 0, 100, 50)
    dur = st.slider("Duration (ms)", 0, 600000, 200000)
    dance = st.slider("Danceability", 0.0, 1.0, 0.5)
with c2:
    energy = st.slider("Energy", 0.0, 1.0, 0.5)
    loud = st.slider("Loudness", -60.0, 0.0, -10.0)
    tempo = st.slider("Tempo", 0.0, 250.0, 120.0)
with c3:
    speech = st.slider("Speechiness", 0.0, 1.0, 0.1)

if st.button("Predict Cluster"):
    new_data = [[pop, dur, dance, energy, loud, tempo, speech]]
    new_data_scaled = scaler.transform(new_data)
    res = model.predict(new_data_scaled)[0]
    st.success(f"Predicted Cluster: {res}")
    st.write(df[df['cluster']==res][['track_name','artists']].head(5))

# -------------------------------
# 6️⃣ Cluster means
# -------------------------------
st.divider()
st.subheader("Cluster Characteristics / Küme Ortalamaları")
numeric_only = df.select_dtypes(include=['float64','int64'])
means = numeric_only.groupby(df['cluster']).mean()
st.dataframe(means, use_container_width=True)