Azmi
/
TIF_E41212325
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

new

This commit is contained in:
Azmikun1 2025-12-18 08:30:49 +07:00
parent bb8fa42585
commit 4c6f66a4da
2 changed files with 233 additions and 209 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

440
app.py
View File

@ -3,10 +3,10 @@ import streamlit as st
import joblib
import pandas as pd
import numpy as np
import os # Tambahan untuk manajemen file
from sklearn.preprocessing import MinMaxScaler
from tensorflow.keras.models import load_model
from datetime import date, timedelta
import yfinance as yf
from pathlib import Path
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
@ -18,9 +18,19 @@ rcParams['font.family'] = 'DejaVu Sans'
# --- Konfigurasi Halaman ---
st.set_page_config(
page_title="Prediksi Harga Ethereum (GRU)",
page_icon="🪙"
page_icon="🪙",
layout="wide" # Opsional: agar tampilan lebih luas
)
# --- KONFIGURASI FILE ADMIN ---
DATA_FOLDER = 'dataset'
DATA_FILE = 'databackup_eth.csv'
DATA_PATH = os.path.join(DATA_FOLDER, DATA_FILE)
# Pastikan folder dataset ada
if not os.path.exists(DATA_FOLDER):
os.makedirs(DATA_FOLDER)
# --- GLOBAL STYLES (UI only) ---
st.markdown("""
<style>
@ -32,6 +42,15 @@ html, body, [class*="css"] { font-family: "Inter", "DejaVu Sans", sans-serif; }
--ring: #e5e7eb;
}
/* --- PERBAIKAN DISINI: MEMBATASI LEBAR KONTEN --- */
/* Ini memaksa konten tetap di tengah dengan lebar maksimal tertentu */
.block-container {
max-width: 1000px; /* Atur angka ini sesuai selera (misal 900px - 1200px) */
padding-top: 2rem;
padding-bottom: 2rem;
margin: auto; /* Posisi otomatis di tengah */
}
/* gradient title */
.app-title {
text-align:center;
@ -81,74 +100,49 @@ html, body, [class*="css"] { font-family: "Inter", "DejaVu Sans", sans-serif; }
""", unsafe_allow_html=True)
# --- Fungsi-fungsi Bantuan ---
# --- Fungsi-fungsi Bantuan (JANGAN UBAH LOGIKA UTAMA) ---
@st.cache_data(ttl="1h")
def load_eth_data():
ticker = "ETH-USD"
df = None
# 1. COBA ONLINE
try:
df = yf.download(
ticker,
start="2024-01-01",
end=date.today() + timedelta(days=1),
progress=False,
auto_adjust=True,
multi_level_index=False
)
if df is not None and not df.empty:
# Bersihkan Index & Kolom
df = df.reset_index()
"""
"""
# Cek apakah file admin ada
if os.path.exists(DATA_PATH):
try:
df = pd.read_csv(DATA_PATH)
new_cols = []
for col in df.columns:
col_name = col[0] if isinstance(col, tuple) else str(col)
new_cols.append(col_name)
df.columns = new_cols
# --- PEMBERSIHAN DATA AGAR SESUAI FORMAT MODEL ---
# 1. Hapus kolom index lama jika ada
if "Unnamed: 0" in df.columns:
df = df.drop(columns=["Unnamed: 0"])
# Pastikan kolom pertama adalah Date
if 'Date' not in df.columns:
df = df.rename(columns={df.columns[0]: 'Date'})
# Hapus Timezone
df['Date'] = pd.to_datetime(df['Date']).dt.tz_localize(None)
# Simpan Backup (Timpa file lama agar fresh)
try:
df.to_csv("eth_backup.csv", index=False)
except:
pass
# 2. Standarisasi nama kolom Date
if "Date" not in df.columns:
# Coba cari kolom yang mirip 'date'
found = False
for col in df.columns:
if col.lower() == "date":
df = df.rename(columns={col: "Date"})
found = True
break
if not found:
# Jika tidak ada header Date, asumsikan kolom pertama adalah Date
df = df.rename(columns={df.columns[0]: 'Date'})
return df, "online"
except Exception as e:
print(f"Gagal Online: {e}")
# 2. COBA BACKUP (JIKA ONLINE GAGAL)
try:
df_backup = pd.read_csv("eth_backup.csv")
# Bersihkan kolom sampah jika ada
if "Unnamed: 0" in df_backup.columns:
df_backup = df_backup.drop(columns=["Unnamed: 0"])
# 3. Konversi ke datetime
df["Date"] = pd.to_datetime(df["Date"])
# Hapus timezone jika ada agar kompatibel dengan matplotlib/numpy
if df["Date"].dt.tz is not None:
df["Date"] = df["Date"].dt.tz_localize(None)
return df, "admin_file"
# Pastikan kolom Date dikenali
if "Date" in df_backup.columns:
df_backup["Date"] = pd.to_datetime(df_backup["Date"])
elif df_backup.columns[0].lower() == "date":
df_backup = df_backup.rename(columns={df_backup.columns[0]: "Date"})
df_backup["Date"] = pd.to_datetime(df_backup["Date"])
return df_backup, "backup"
except FileNotFoundError:
# 3. GAGAL TOTAL
return None, "error"
except Exception as e:
return None, f"error_read: {str(e)}"
else:
return None, "no_file"
def validate_scaler(scaler):
"""Validasi scaler agar konsisten dengan training."""
issues = []
@ -191,44 +185,23 @@ def validate_model_input(model):
@st.cache_resource
def load_gru_assets():
"""Load model & scaler hasil training (wajib pakai scaler yang sama)."""
"""Load model & scaler hasil training."""
model_path = Path("gru_model.h5")
scaler_path = Path("scaler_gru.pkl")
if not model_path.exists() or not scaler_path.exists():
st.warning(
f"File tidak ditemukan.\n"
f"Model: {model_path.resolve().name} ada? {model_path.exists()}\n"
f"Scaler: {scaler_path.resolve().name} ada? {scaler_path.exists()}"
)
return None, None
try:
model = load_model(model_path, compile=False)
scaler = joblib.load(scaler_path)
ok_scaler, scaler_issues, scaler_warnings = validate_scaler(scaler)
for w in scaler_warnings:
st.warning("⚠️ " + w)
if not ok_scaler:
st.error("❌ Scaler tidak kompatibel:\n- " + "\n- ".join(scaler_issues))
return None, None
ok_model, model_issues = validate_model_input(model)
if not ok_model:
st.error("❌ Model input tidak kompatibel:\n- " + "\n- ".join(model_issues))
return None, None
return model, scaler
except Exception as e:
st.error(f"Gagal load aset: {e}")
return None, None
def predict_from_sequence_pure(model, initial_sequence_scaled, n_days):
seq = np.array(initial_sequence_scaled, dtype="float32").reshape(-1) # (time_step,)
preds = []
@ -242,7 +215,7 @@ def predict_from_sequence_pure(model, initial_sequence_scaled, n_days):
def create_combined_chart(df, start_date, future_dates, future_predictions):
"""Membuat grafik gabungan dengan gaya yang sama seperti referensi."""
"""Membuat grafik gabungan."""
context_start_date = start_date - timedelta(days=60)
recent_df = df[(df["Date"] >= context_start_date) & (df["Date"] <= start_date)].copy()
@ -271,7 +244,6 @@ def create_combined_chart(df, start_date, future_dates, future_predictions):
color="black", s=15, alpha=0.6, zorder=5, label="Data Harian Aktual"
)
# garis penghubung trend terakhir ke prediksi pertama
try:
last_trend_date = recent_df["Date"].iloc[-1]
last_trend_price = recent_df["Trend_Aktual"].dropna().iloc[-1]
@ -308,132 +280,184 @@ def create_combined_chart(df, start_date, future_dates, future_predictions):
return fig
# --- Tampilan Antarmuka Aplikasi ---
# --- HALAMAN: DASHBOARD PUBLIK (Landing Page) ---
def show_dashboard():
st.markdown("<div class='app-title'>Prediksi Harga Ethereum (GRU)</div>", unsafe_allow_html=True)
st.markdown("<div class='app-subtitle'>Data historis & prediksi ETH-USD berbasis GRU</div>", unsafe_allow_html=True)
st.markdown("<div class='app-title'>Prediksi Harga Ethereum (GRU)</div>", unsafe_allow_html=True)
st.markdown("<div class='app-subtitle'>Data historis & prediksi ETH-USD berbasis GRU</div>", unsafe_allow_html=True)
df, data_source = load_eth_data()
model, scaler = load_gru_assets()
df, data_source = load_eth_data()
model, scaler = load_gru_assets()
# LOGIC CHECK DATA
if data_source == "no_file":
st.warning("⚠️ Data historis belum tersedia. Silakan hubungi Admin untuk mengupload 'databackup_eth'.")
return # Stop eksekusi jika data tidak ada
elif "error" in data_source:
st.error(f"❌ Terjadi kesalahan membaca data: {data_source}")
return
# LOGIC UI (TOAST/WARNING) DITARUH DI SINI (DILUAR CACHE)
if data_source == "backup":
st.toast("Koneksi Yahoo lambat. Menggunakan data backup lokal.", icon="⚠️")
elif data_source == "error":
st.error("❌ Gagal memuat data (Online gagal & Backup tidak ada).")
if df is not None and model is not None and scaler is not None:
# Menampilkan tabel data historis
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>📚 Data Historis Harga Ethereum </div>", unsafe_allow_html=True)
# --- PERUBAHAN DISINI: HAPUS .tail(100) ---
# Menggunakan height=400 agar tabel lebih tinggi dan bisa di-scroll
st.dataframe(df, height=400, use_container_width=True)
# Menampilkan visualisasi line chart
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>📈 Visualisasi Harga Historis</div>", unsafe_allow_html=True)
# Grafik ini sudah otomatis mengambil full data df
st.line_chart(df.rename(columns={"Date": "index"}).set_index("index")["Close"])
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>🎯 Mulai Prediksi Berdasarkan Tanggal</div>", unsafe_allow_html=True)
time_step = int(model.input_shape[1])
min_selectable_date = (df["Date"].min() + timedelta(days=time_step)).date()
max_selectable_date = df["Date"].max().date()
col1, col2 = st.columns(2)
with col1:
selected_date = st.date_input(
"Pilih tanggal mulai prediksi:",
value=max_selectable_date,
min_value=min_selectable_date,
max_value=max_selectable_date,
help=f"Pilih tanggal antara {min_selectable_date} dan {max_selectable_date}"
)
with col2:
prediction_days = st.slider(
"Pilih jumlah hari untuk prediksi (1-30 hari):",
min_value=1, max_value=30, value=15, step=1
)
if st.button("Buat Prediksi", key="predict_button"):
if prediction_days < 1:
st.warning("Jumlah hari prediksi minimal **1**. Silakan geser slidernya dulu. 🙂")
st.stop()
with st.spinner("Memproses dan menjalankan prediksi..."):
# scaling (mengikuti scaler training)
close_values = df[["Close"]].values.astype(float)
scaled_data = scaler.transform(close_values)
# Cocokkan berdasarkan tanggal (robust)
selected_date_dt = pd.to_datetime(selected_date)
mask = df["Date"].dt.date == selected_date_dt.date()
idxs = np.where(mask.to_numpy())[0]
if len(idxs) == 0:
st.error(f"Tanggal {selected_date_dt.strftime('%Y-%m-%d')} tidak ditemukan dalam dataset.")
st.stop()
start_index = int(idxs[0])
if start_index - time_step < 0:
st.error(
f"Tanggal terlalu awal untuk diprediksi. "
f"Butuh {time_step} hari data sebelumnya."
)
st.stop()
initial_sequence = scaled_data[start_index - time_step: start_index] # (time_step,1)
# ✅ Prediksi MURNI dari model
future_predictions_scaled = predict_from_sequence_pure(model, initial_sequence, prediction_days)
# ✅ Inverse transform MURNI dari scaler
future_predictions = scaler.inverse_transform(future_predictions_scaled)
# sanity-check
if not np.isfinite(future_predictions).all():
st.error("❌ Hasil prediksi mengandung NaN/Inf. Periksa kompatibilitas model/scaler.")
st.stop()
future_dates = [selected_date_dt + timedelta(days=i) for i in range(1, prediction_days + 1)]
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.subheader("📊 Grafik Prediksi (Historis + Prediksi)")
fig = create_combined_chart(df, selected_date_dt, future_dates, future_predictions)
st.pyplot(fig)
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.subheader("🧾 Tabel Detail Prediksi")
prediction_table_df = pd.DataFrame({
"Tanggal": [d.strftime("%Y-%m-%d") for d in future_dates],
"Harga Prediksi (USD)": [f"USD {price[0]:,.0f}" for price in future_predictions]
})
prediction_table_df.index = prediction_table_df.index + 1
st.dataframe(prediction_table_df, use_container_width=True)
st.markdown("</div>", unsafe_allow_html=True)
last_price = float(df.loc[start_index, "Close"])
max_prediction = float(np.max(future_predictions))
min_prediction = float(np.min(future_predictions))
st.info(f"""
📊 **Informasi Prediksi:**
- Harga terakhir pada tanggal {selected_date.strftime('%Y-%m-%d')}: USD {last_price:,.0f}
- Harga prediksi tertinggi: USD {max_prediction:,.0f}
- Harga prediksi terendah: USD {min_prediction:,.0f}
""")
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='footer'>Dibuat oleh Muhammad Gilman Nadhif Azmi</div>", unsafe_allow_html=True)
elif model is None:
st.error("Gagal memuat model. Pastikan file 'gru_model.h5' ada di folder yang sama dengan aplikasi.")
elif scaler is None:
st.error("Gagal memuat scaler. Pastikan file 'scaler_gru.pkl' ada di folder yang sama dengan aplikasi.")
# --- HALAMAN: ADMIN LOGIN & UPLOAD ---
def show_admin_page():
st.title("🔐 Admin Panel")
st.write("Area khusus admin untuk memperbarui dataset harga Ethereum.")
# Simple Password check
password = st.text_input("Masukkan Password Admin:", type="password")
if df is not None and model is not None and scaler is not None:
# Menampilkan tabel data historis
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>📚 Data Historis Harga Ethereum</div>", unsafe_allow_html=True)
st.dataframe(df, height=300, use_container_width=True)
st.markdown("</div>", unsafe_allow_html=True)
if password == "admin123": # Ganti password sesuai keinginan
st.success("Akses Diterima.")
st.divider()
st.subheader("📂 Upload Dataset Baru")
st.info(f"File akan disimpan sebagai: `{DATA_FILE}` di dalam folder sistem.")
uploaded_file = st.file_uploader("Upload file CSV (Format: Date, Close, dll)", type=['csv'])
if uploaded_file is not None:
# Baca preview
try:
df_preview = pd.read_csv(uploaded_file)
st.write("Preview Data:", df_preview.head())
if st.button("💾 Simpan & Update Sistem"):
# Simpan file ke path yang ditentukan
df_preview.to_csv(DATA_PATH, index=False)
# Clear cache agar halaman public langsung berubah
st.cache_data.clear()
st.success(f"Berhasil! Data telah diperbarui. Pengguna publik sekarang melihat data baru.")
except Exception as e:
st.error(f"File error: {e}")
elif password:
st.error("Password Salah.")
# Menampilkan visualisasi line chart
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>📈 Visualisasi Harga Historis</div>", unsafe_allow_html=True)
st.line_chart(df.rename(columns={"Date": "index"}).set_index("index")["Close"])
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.markdown("<div class='h-section'>🎯 Mulai Prediksi Berdasarkan Tanggal</div>", unsafe_allow_html=True)
# --- MAIN NAVIGATION (SIDEBAR) ---
# Ini adalah logika utama yang memisahkan tampilan Admin dan User Biasa
time_step = int(model.input_shape[1])
min_selectable_date = (df["Date"].min() + timedelta(days=time_step)).date()
max_selectable_date = df["Date"].max().date()
st.sidebar.title("Navigasi")
menu = st.sidebar.radio("Pilih Halaman:", ["Dashboard (Public)", "Admin Panel"])
col1, col2 = st.columns(2)
with col1:
selected_date = st.date_input(
"Pilih tanggal mulai prediksi:",
value=max_selectable_date,
min_value=min_selectable_date,
max_value=max_selectable_date,
help=f"Pilih tanggal antara {min_selectable_date} dan {max_selectable_date}"
)
with col2:
prediction_days = st.slider(
"Pilih jumlah hari untuk prediksi (1-30 hari):",
min_value=1, max_value=30, value=15, step=1
)
if st.button("Buat Prediksi", key="predict_button"):
if prediction_days < 1:
st.warning("Jumlah hari prediksi minimal **1**. Silakan geser slidernya dulu. 🙂")
st.stop()
with st.spinner("Memproses dan menjalankan prediksi..."):
# scaling (mengikuti scaler training)
close_values = df[["Close"]].values.astype(float)
scaled_data = scaler.transform(close_values)
# Cocokkan berdasarkan tanggal (robust)
selected_date_dt = pd.to_datetime(selected_date)
mask = df["Date"].dt.date == selected_date_dt.date()
idxs = np.where(mask.to_numpy())[0]
if len(idxs) == 0:
st.error(f"Tanggal {selected_date_dt.strftime('%Y-%m-%d')} tidak ditemukan dalam dataset.")
st.stop()
start_index = int(idxs[0])
if start_index - time_step < 0:
st.error(
f"Tanggal terlalu awal untuk diprediksi. "
f"Butuh {time_step} hari data sebelumnya. Pilih tanggal setelah "
f"{(df['Date'].min() + timedelta(days=time_step)).strftime('%Y-%m-%d')}."
)
st.stop()
initial_sequence = scaled_data[start_index - time_step: start_index] # (time_step,1)
# ✅ Prediksi MURNI dari model
future_predictions_scaled = predict_from_sequence_pure(model, initial_sequence, prediction_days)
# ✅ Inverse transform MURNI dari scaler
future_predictions = scaler.inverse_transform(future_predictions_scaled)
# sanity-check
if not np.isfinite(future_predictions).all():
st.error("❌ Hasil prediksi mengandung NaN/Inf. Periksa kompatibilitas model/scaler.")
st.stop()
future_dates = [selected_date_dt + timedelta(days=i) for i in range(1, prediction_days + 1)]
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.subheader("📊 Grafik Prediksi (Historis + Prediksi)")
fig = create_combined_chart(df, selected_date_dt, future_dates, future_predictions)
st.pyplot(fig)
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='card'>", unsafe_allow_html=True)
st.subheader("🧾 Tabel Detail Prediksi")
prediction_table_df = pd.DataFrame({
"Tanggal": [d.strftime("%Y-%m-%d") for d in future_dates],
"Harga Prediksi (USD)": [f"USD {price[0]:,.0f}" for price in future_predictions]
})
prediction_table_df.index = prediction_table_df.index + 1
st.dataframe(prediction_table_df, use_container_width=True)
st.markdown("</div>", unsafe_allow_html=True)
last_price = float(df.loc[start_index, "Close"])
max_prediction = float(np.max(future_predictions))
min_prediction = float(np.min(future_predictions))
st.info(f"""
📊 **Informasi Prediksi:**
- Harga terakhir pada tanggal {selected_date.strftime('%Y-%m-%d')}: USD {last_price:,.0f}
- Harga prediksi tertinggi: USD {max_prediction:,.0f}
- Harga prediksi terendah: USD {min_prediction:,.0f}
""")
st.markdown("</div>", unsafe_allow_html=True)
st.markdown("<div class='footer'>Dibuat oleh Muhammad Gilman Nadhif Azmi</div>", unsafe_allow_html=True)
elif model is None:
st.error("Gagal memuat model. Pastikan file 'gru_model.h5' ada di folder yang sama dengan aplikasi.")
elif scaler is None:
st.error("Gagal memuat scaler. Pastikan file 'scaler_gru.pkl' ada di folder yang sama dengan aplikasi.")
else:
st.error("Gagal memuat data. Coba cek koneksi internet atau sumber data yfinance.")
if menu == "Dashboard (Public)":
show_dashboard()
elif menu == "Admin Panel":
show_admin_page()

2
eth_backup.csv
View File

@ -715,4 +715,4 @@ Date,Close,High,Low,Open,Volume
2025-12-14,3060.5947265625,3128.622802734375,3034.692626953125,3116.743896484375,15619543350
2025-12-15,2964.18310546875,3175.1181640625,2899.685791015625,3060.4814453125,28765976892
2025-12-16,2964.180908203125,2978.92138671875,2890.01171875,2964.379638671875,22709189818
2025-12-17,2933.59375,2969.88525390625,2920.556884765625,2962.631103515625,19631171584
2025-12-18,2829.30078125,2836.117431640625,2822.574951171875,2832.8203125,26052483072

1 Date Close High Low Open Volume
715 2025-12-14 3060.5947265625 3128.622802734375 3034.692626953125 3116.743896484375 15619543350
716 2025-12-15 2964.18310546875 3175.1181640625 2899.685791015625 3060.4814453125 28765976892
717 2025-12-16 2964.180908203125 2978.92138671875 2890.01171875 2964.379638671875 22709189818
718 2025-12-17 2025-12-18 2933.59375 2829.30078125 2969.88525390625 2836.117431640625 2920.556884765625 2822.574951171875 2962.631103515625 2832.8203125 19631171584 26052483072