Update src/streamlit_app.py
Browse files- src/streamlit_app.py +16 -44
src/streamlit_app.py
CHANGED
|
@@ -2,40 +2,24 @@ import streamlit as st
|
|
| 2 |
import pandas as pd
|
| 3 |
import joblib
|
| 4 |
import matplotlib.pyplot as plt
|
| 5 |
-
import os
|
| 6 |
|
| 7 |
# --------------------------------------
|
| 8 |
# PAGE CONFIG
|
| 9 |
# --------------------------------------
|
| 10 |
st.set_page_config(page_title="Spotify Clustering", layout="wide")
|
|
|
|
|
|
|
| 11 |
|
| 12 |
# --------------------------------------
|
| 13 |
-
#
|
| 14 |
-
#
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
"..", # bir üst klasör
|
| 20 |
-
os.path.dirname(os.path.abspath(__file__)), # script klasörü
|
| 21 |
-
os.path.join(os.path.dirname(os.path.abspath(__file__)), ".."),
|
| 22 |
-
os.getcwd(), # working dir
|
| 23 |
-
os.path.join(os.getcwd(), "..")
|
| 24 |
-
]
|
| 25 |
-
for d in base_dirs:
|
| 26 |
-
path = os.path.join(d, filename)
|
| 27 |
-
if os.path.exists(path):
|
| 28 |
-
return path
|
| 29 |
-
# Eğer bulunamazsa
|
| 30 |
-
st.error(f"{filename} bulunamadı! Mevcut dosyalar: {os.listdir()}")
|
| 31 |
-
st.stop()
|
| 32 |
-
|
| 33 |
-
DATA_PATH = get_path("spotify_clustered.csv")
|
| 34 |
-
MODEL_PATH = get_path("kmeans_music_model.pkl")
|
| 35 |
-
SCALER_PATH = get_path("scaler_music.pkl")
|
| 36 |
|
| 37 |
# --------------------------------------
|
| 38 |
-
# LOAD DATA
|
| 39 |
# --------------------------------------
|
| 40 |
@st.cache_data
|
| 41 |
def load_data():
|
|
@@ -60,12 +44,6 @@ if "prediction" not in st.session_state:
|
|
| 60 |
st.session_state.prediction = None
|
| 61 |
st.session_state.samples = None
|
| 62 |
|
| 63 |
-
# --------------------------------------
|
| 64 |
-
# TITLE
|
| 65 |
-
# --------------------------------------
|
| 66 |
-
st.title("🎵 Spotify Music Clustering / Spotify Müzik Kümeleme")
|
| 67 |
-
st.markdown("---")
|
| 68 |
-
|
| 69 |
# --------------------------------------
|
| 70 |
# DATA PREVIEW
|
| 71 |
# --------------------------------------
|
|
@@ -73,7 +51,7 @@ st.subheader("📄 Dataset Preview / Veri Önizleme")
|
|
| 73 |
st.dataframe(df_display.head(10), use_container_width=True)
|
| 74 |
|
| 75 |
# --------------------------------------
|
| 76 |
-
#
|
| 77 |
# --------------------------------------
|
| 78 |
col1, col2 = st.columns(2)
|
| 79 |
|
|
@@ -83,7 +61,7 @@ with col1:
|
|
| 83 |
|
| 84 |
with col2:
|
| 85 |
st.subheader("🎯 Feature Analysis / Özellik Analizi")
|
| 86 |
-
fig, ax = plt.subplots(figsize=(6,
|
| 87 |
scatter = ax.scatter(df['danceability'], df['energy'], c=df['cluster'], cmap='viridis', alpha=0.6)
|
| 88 |
ax.set_xlabel("Danceability / Dans Edilebilirlik")
|
| 89 |
ax.set_ylabel("Energy / Enerji")
|
|
@@ -115,9 +93,9 @@ if st.button("Predict Cluster / Kümeyi Tahmin Et ✨"):
|
|
| 115 |
new_data_scaled = scaler.transform(new_data)
|
| 116 |
res = model.predict(new_data_scaled)[0]
|
| 117 |
st.session_state.prediction = res
|
| 118 |
-
st.session_state.samples = df[df['cluster']
|
| 119 |
except Exception as e:
|
| 120 |
-
st.error(f"Error /
|
| 121 |
|
| 122 |
if st.session_state.prediction is not None:
|
| 123 |
st.success(f"### Predicted Cluster / Tahmin Edilen Küme: {st.session_state.prediction}")
|
|
@@ -125,17 +103,11 @@ if st.session_state.prediction is not None:
|
|
| 125 |
st.table(st.session_state.samples)
|
| 126 |
|
| 127 |
# --------------------------------------
|
| 128 |
-
# CLUSTER
|
| 129 |
# --------------------------------------
|
| 130 |
st.divider()
|
| 131 |
st.subheader("🔍 Cluster Characteristics / Küme Özellikleri")
|
| 132 |
-
numeric_only = df.select_dtypes(include=['float64',
|
| 133 |
if 'cluster' in df.columns:
|
| 134 |
means = numeric_only.groupby(df['cluster']).mean()
|
| 135 |
-
st.dataframe(means, use_container_width=True)
|
| 136 |
-
|
| 137 |
-
# --------------------------------------
|
| 138 |
-
# DEBUG (Opsiyonel)
|
| 139 |
-
# --------------------------------------
|
| 140 |
-
# st.write("Current working dir:", os.getcwd())
|
| 141 |
-
# st.write("Files here:", os.listdir())
|
|
|
|
| 2 |
import pandas as pd
|
| 3 |
import joblib
|
| 4 |
import matplotlib.pyplot as plt
|
|
|
|
| 5 |
|
| 6 |
# --------------------------------------
|
| 7 |
# PAGE CONFIG
|
| 8 |
# --------------------------------------
|
| 9 |
st.set_page_config(page_title="Spotify Clustering", layout="wide")
|
| 10 |
+
st.title("🎵 Spotify Music Clustering / Spotify Müzik Kümeleme")
|
| 11 |
+
st.markdown("---")
|
| 12 |
|
| 13 |
# --------------------------------------
|
| 14 |
+
# FILE PATHS
|
| 15 |
+
# Dosyalar aynı klasörde olduğu için sadece isim yeter
|
| 16 |
+
# --------------------------------------
|
| 17 |
+
DATA_PATH = "spotify_clustered.csv"
|
| 18 |
+
MODEL_PATH = "kmeans_music_model.pkl"
|
| 19 |
+
SCALER_PATH = "scaler_music.pkl"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
# --------------------------------------
|
| 22 |
+
# LOAD DATA & MODEL
|
| 23 |
# --------------------------------------
|
| 24 |
@st.cache_data
|
| 25 |
def load_data():
|
|
|
|
| 44 |
st.session_state.prediction = None
|
| 45 |
st.session_state.samples = None
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
# --------------------------------------
|
| 48 |
# DATA PREVIEW
|
| 49 |
# --------------------------------------
|
|
|
|
| 51 |
st.dataframe(df_display.head(10), use_container_width=True)
|
| 52 |
|
| 53 |
# --------------------------------------
|
| 54 |
+
# VISUALIZATIONS
|
| 55 |
# --------------------------------------
|
| 56 |
col1, col2 = st.columns(2)
|
| 57 |
|
|
|
|
| 61 |
|
| 62 |
with col2:
|
| 63 |
st.subheader("🎯 Feature Analysis / Özellik Analizi")
|
| 64 |
+
fig, ax = plt.subplots(figsize=(6,4))
|
| 65 |
scatter = ax.scatter(df['danceability'], df['energy'], c=df['cluster'], cmap='viridis', alpha=0.6)
|
| 66 |
ax.set_xlabel("Danceability / Dans Edilebilirlik")
|
| 67 |
ax.set_ylabel("Energy / Enerji")
|
|
|
|
| 93 |
new_data_scaled = scaler.transform(new_data)
|
| 94 |
res = model.predict(new_data_scaled)[0]
|
| 95 |
st.session_state.prediction = res
|
| 96 |
+
st.session_state.samples = df[df['cluster']==res][['track_name','artists']].head(5)
|
| 97 |
except Exception as e:
|
| 98 |
+
st.error(f"Prediction Error / Tahmin Hatası: {e}")
|
| 99 |
|
| 100 |
if st.session_state.prediction is not None:
|
| 101 |
st.success(f"### Predicted Cluster / Tahmin Edilen Küme: {st.session_state.prediction}")
|
|
|
|
| 103 |
st.table(st.session_state.samples)
|
| 104 |
|
| 105 |
# --------------------------------------
|
| 106 |
+
# CLUSTER CHARACTERISTICS
|
| 107 |
# --------------------------------------
|
| 108 |
st.divider()
|
| 109 |
st.subheader("🔍 Cluster Characteristics / Küme Özellikleri")
|
| 110 |
+
numeric_only = df.select_dtypes(include=['float64','int64'])
|
| 111 |
if 'cluster' in df.columns:
|
| 112 |
means = numeric_only.groupby(df['cluster']).mean()
|
| 113 |
+
st.dataframe(means, use_container_width=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|