Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -7,15 +7,31 @@ import seaborn as sns
|
|
7 |
import matplotlib.pyplot as plt
|
8 |
from datetime import date
|
9 |
from pathlib import Path
|
|
|
|
|
10 |
|
11 |
# -------------------------------------------------
|
12 |
# CONFIG ------------------------------------------
|
13 |
# -------------------------------------------------
|
14 |
CSV_PATH = Path("price_data.csv")
|
15 |
-
PARQUET_PATH = Path("domae-202503.parquet")
|
16 |
MACRO_START, MACRO_END = "1996-01-01", "2030-12-31"
|
17 |
MICRO_START, MICRO_END = "2020-01-01", "2026-12-31"
|
18 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
st.set_page_config(page_title="ํ๋ชฉ๋ณ ๊ฐ๊ฒฉ ์์ธก", page_icon="๐", layout="wide")
|
20 |
|
21 |
# -------------------------------------------------
|
@@ -53,10 +69,11 @@ def _standardize_columns(df: pd.DataFrame) -> pd.DataFrame:
|
|
53 |
df.rename(columns={df.columns[0]: "date"}, inplace=True)
|
54 |
|
55 |
# โโ convert YYYYMM string to datetime โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
|
56 |
-
if "date" in df.columns and pd.api.types.is_object_dtype(df["date"]):
|
57 |
-
|
58 |
-
|
59 |
-
|
|
|
60 |
|
61 |
# โโ build item from pdlt_nm + spcs_nm if needed โโโโโโโโโโโโโโโโโโโโ
|
62 |
if "item" not in df.columns and {"pdlt_nm", "spcs_nm"}.issubset(df.columns):
|
@@ -73,62 +90,105 @@ def _standardize_columns(df: pd.DataFrame) -> pd.DataFrame:
|
|
73 |
@st.cache_data(show_spinner=False)
|
74 |
def load_data() -> pd.DataFrame:
|
75 |
"""Load price data from Parquet if available, else CSV. Handle flexible schema."""
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
st.stop()
|
90 |
|
91 |
-
df["date"] = pd.to_datetime(df["date"], errors="coerce")
|
92 |
-
df = df.dropna(subset=["date", "item", "price"])
|
93 |
-
df.sort_values("date", inplace=True)
|
94 |
-
return df
|
95 |
-
|
96 |
|
97 |
@st.cache_data(show_spinner=False)
|
98 |
def get_items(df: pd.DataFrame):
|
99 |
return sorted(df["item"].unique())
|
100 |
|
101 |
|
102 |
-
@st.cache_data(show_spinner=False)
|
103 |
def fit_prophet(df: pd.DataFrame, horizon_end: str):
|
104 |
# Make a copy and ensure we have data
|
105 |
df = df.copy()
|
106 |
df = df.dropna(subset=["date", "price"])
|
107 |
|
|
|
|
|
|
|
108 |
if len(df) < 2:
|
109 |
-
st.warning("๋ฐ์ดํฐ ํฌ์ธํธ๊ฐ ๋ถ์กฑํฉ๋๋ค. ์์ธก์ ์ํด์๋ ์ต์ 2๊ฐ ์ด์์ ์ ํจ ๋ฐ์ดํฐ๊ฐ ํ์ํฉ๋๋ค.")
|
110 |
return None, None
|
111 |
|
112 |
# Convert to Prophet format
|
113 |
prophet_df = df.rename(columns={"date": "ds", "price": "y"})
|
114 |
|
115 |
-
|
116 |
-
|
117 |
-
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
|
125 |
-
|
|
|
|
|
|
|
|
|
126 |
|
127 |
# -------------------------------------------------
|
128 |
# LOAD DATA ---------------------------------------
|
129 |
# -------------------------------------------------
|
130 |
raw_df = load_data()
|
131 |
|
|
|
|
|
|
|
|
|
132 |
st.sidebar.header("๐ ํ๋ชฉ ์ ํ")
|
133 |
selected_item = st.sidebar.selectbox("ํ๋ชฉ", get_items(raw_df))
|
134 |
current_date = date.today()
|
@@ -143,29 +203,50 @@ if item_df.empty:
|
|
143 |
# MACRO FORECAST 1996โ2030 ------------------------
|
144 |
# -------------------------------------------------
|
145 |
st.header(f"๐ {selected_item} ๊ฐ๊ฒฉ ์์ธก ๋์๋ณด๋")
|
146 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
147 |
|
148 |
# Add diagnostic info
|
149 |
with st.expander("๋ฐ์ดํฐ ์ง๋จ"):
|
150 |
st.write(f"- ์ ์ฒด ๋ฐ์ดํฐ ์: {len(item_df)}")
|
151 |
-
st.write(f"-
|
152 |
-
|
153 |
-
|
|
|
|
|
|
|
154 |
|
155 |
if len(macro_df) < 2:
|
156 |
-
st.warning(f"{
|
157 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
158 |
st.plotly_chart(fig, use_container_width=True)
|
159 |
else:
|
160 |
try:
|
161 |
-
|
|
|
|
|
162 |
if m_macro is not None and fc_macro is not None:
|
163 |
-
fig_macro = px.line(fc_macro, x="ds", y="yhat", title="
|
164 |
-
fig_macro.add_scatter(x=macro_df["date"], y=macro_df["price"], mode="lines", name="
|
165 |
st.plotly_chart(fig_macro, use_container_width=True)
|
166 |
|
167 |
latest_price = macro_df.iloc[-1]["price"]
|
168 |
-
|
|
|
|
|
|
|
169 |
macro_pct = (macro_pred - latest_price) / latest_price * 100
|
170 |
st.metric("2030 ์์ธก๊ฐ", f"{macro_pred:,.0f}", f"{macro_pct:+.1f}%")
|
171 |
else:
|
@@ -173,7 +254,7 @@ else:
|
|
173 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
174 |
st.plotly_chart(fig, use_container_width=True)
|
175 |
except Exception as e:
|
176 |
-
st.error(f"์ค๋ฅ ๋ฐ์: {str(e)}")
|
177 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
178 |
st.plotly_chart(fig, use_container_width=True)
|
179 |
|
@@ -182,21 +263,40 @@ else:
|
|
182 |
# -------------------------------------------------
|
183 |
st.subheader("๐ 2024โ2026 ๋จ๊ธฐ ์์ธก")
|
184 |
|
185 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
186 |
if len(micro_df) < 2:
|
187 |
st.warning(f"{MICRO_START} ์ดํ ๋ฐ์ดํฐ๊ฐ ์ถฉ๋ถํ์ง ์์ต๋๋ค.")
|
188 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ์ต๊ทผ ๊ฐ๊ฒฉ")
|
189 |
st.plotly_chart(fig, use_container_width=True)
|
190 |
else:
|
191 |
try:
|
192 |
-
|
|
|
|
|
193 |
if m_micro is not None and fc_micro is not None:
|
194 |
-
fig_micro = px.line(fc_micro, x="ds", y="yhat", title="
|
195 |
-
fig_micro.add_scatter(x=micro_df["date"], y=micro_df["price"], mode="lines", name="
|
196 |
st.plotly_chart(fig_micro, use_container_width=True)
|
197 |
|
198 |
latest_price = micro_df.iloc[-1]["price"]
|
199 |
-
|
|
|
|
|
200 |
micro_pct = (micro_pred - latest_price) / latest_price * 100
|
201 |
st.metric("2026 ์์ธก๊ฐ", f"{micro_pred:,.0f}", f"{micro_pct:+.1f}%")
|
202 |
else:
|
@@ -209,16 +309,19 @@ else:
|
|
209 |
# -------------------------------------------------
|
210 |
with st.expander("๐ ์์ฆ๋๋ฆฌํฐ & ํจํด ์ค๋ช
"):
|
211 |
if 'm_micro' in locals() and m_micro is not None and 'fc_micro' in locals() and fc_micro is not None:
|
212 |
-
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
-
|
218 |
-
|
219 |
-
|
220 |
-
|
221 |
-
|
|
|
|
|
|
|
222 |
else:
|
223 |
st.info("ํจํด ๋ถ์์ ์ํ ์ถฉ๋ถํ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|
224 |
|
@@ -226,17 +329,42 @@ with st.expander("๐ ์์ฆ๋๋ฆฌํฐ & ํจํด ์ค๋ช
"):
|
|
226 |
# CORRELATION HEATMAP -----------------------------
|
227 |
# -------------------------------------------------
|
228 |
st.subheader("๐งฎ ํ๋ชฉ ๊ฐ ์๊ด๊ด๊ณ")
|
229 |
-
try:
|
230 |
-
monthly_pivot = (raw_df.assign(month=lambda d: d.date.dt.to_period("M"))
|
231 |
-
.groupby(["month", "item"], as_index=False)["price"].mean()
|
232 |
-
.pivot(index="month", columns="item", values="price"))
|
233 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
234 |
if monthly_pivot.shape[1] > 1: # At least 2 items needed for correlation
|
|
|
|
|
|
|
|
|
235 |
corr = monthly_pivot.corr()
|
|
|
|
|
236 |
fig, ax = plt.subplots(figsize=(12, 10))
|
237 |
mask = np.triu(np.ones_like(corr, dtype=bool))
|
238 |
-
|
|
|
|
|
|
|
|
|
239 |
square=True, linewidths=.5, cbar_kws={"shrink": .5})
|
|
|
|
|
|
|
240 |
|
241 |
# Highlight correlations with selected item
|
242 |
if selected_item in corr.columns:
|
@@ -259,9 +387,10 @@ try:
|
|
259 |
st.info("์๊ด๊ด๊ณ ๋ถ์์ ์ํ ์ถฉ๋ถํ ํ๋ชฉ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|
260 |
except Exception as e:
|
261 |
st.error(f"์๊ด๊ด๊ณ ๋ถ์ ์ค๋ฅ: {str(e)}")
|
|
|
262 |
|
263 |
# -------------------------------------------------
|
264 |
# FOOTER ------------------------------------------
|
265 |
# -------------------------------------------------
|
266 |
st.markdown("---")
|
267 |
-
st.caption("ยฉ
|
|
|
7 |
import matplotlib.pyplot as plt
|
8 |
from datetime import date
|
9 |
from pathlib import Path
|
10 |
+
import matplotlib.font_manager as fm
|
11 |
+
import matplotlib as mpl
|
12 |
|
13 |
# -------------------------------------------------
|
14 |
# CONFIG ------------------------------------------
|
15 |
# -------------------------------------------------
|
16 |
CSV_PATH = Path("price_data.csv")
|
17 |
+
PARQUET_PATH = Path("domae-202503.parquet")
|
18 |
MACRO_START, MACRO_END = "1996-01-01", "2030-12-31"
|
19 |
MICRO_START, MICRO_END = "2020-01-01", "2026-12-31"
|
20 |
|
21 |
+
# ํ๊ธ ํฐํธ ์ค์
|
22 |
+
# 1. ์์คํ
์ ์ค์น๋ ํ๊ธ ํฐํธ ์ฐพ๊ธฐ
|
23 |
+
font_list = [f.name for f in fm.fontManager.ttflist if 'gothic' in f.name.lower() or
|
24 |
+
'gulim' in f.name.lower() or 'malgun' in f.name.lower() or
|
25 |
+
'nanum' in f.name.lower() or 'batang' in f.name.lower()]
|
26 |
+
|
27 |
+
if font_list:
|
28 |
+
font_name = font_list[0]
|
29 |
+
plt.rcParams['font.family'] = font_name
|
30 |
+
mpl.rcParams['axes.unicode_minus'] = False
|
31 |
+
else:
|
32 |
+
# ํฐํธ๊ฐ ์์ ๊ฒฝ์ฐ ๊ธฐ๋ณธ ํฐํธ ์ค์
|
33 |
+
plt.rcParams['font.family'] = 'DejaVu Sans'
|
34 |
+
|
35 |
st.set_page_config(page_title="ํ๋ชฉ๋ณ ๊ฐ๊ฒฉ ์์ธก", page_icon="๐", layout="wide")
|
36 |
|
37 |
# -------------------------------------------------
|
|
|
69 |
df.rename(columns={df.columns[0]: "date"}, inplace=True)
|
70 |
|
71 |
# โโ convert YYYYMM string to datetime โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
|
72 |
+
if "date" in df.columns and pd.api.types.is_object_dtype(df["date"]):
|
73 |
+
if len(df) > 0: # ๋ฐ์ดํฐ๊ฐ ์๋์ง ํ์ธ
|
74 |
+
sample = str(df["date"].iloc[0])
|
75 |
+
if sample.isdigit() and len(sample) in (6, 8):
|
76 |
+
df["date"] = pd.to_datetime(df["date"].astype(str).str[:6], format="%Y%m", errors="coerce")
|
77 |
|
78 |
# โโ build item from pdlt_nm + spcs_nm if needed โโโโโโโโโโโโโโโโโโโโ
|
79 |
if "item" not in df.columns and {"pdlt_nm", "spcs_nm"}.issubset(df.columns):
|
|
|
90 |
@st.cache_data(show_spinner=False)
|
91 |
def load_data() -> pd.DataFrame:
|
92 |
"""Load price data from Parquet if available, else CSV. Handle flexible schema."""
|
93 |
+
try:
|
94 |
+
if PARQUET_PATH.exists():
|
95 |
+
st.sidebar.info("Parquet ํ์ผ์์ ๋ฐ์ดํฐ๋ฅผ ๋ถ๋ฌ์ต๋๋ค.")
|
96 |
+
df = pd.read_parquet(PARQUET_PATH)
|
97 |
+
st.sidebar.success(f"Parquet ๋ฐ์ดํฐ ๋ก๋ ์๋ฃ: {len(df)}๊ฐ ํ")
|
98 |
+
elif CSV_PATH.exists():
|
99 |
+
st.sidebar.info("CSV ํ์ผ์์ ๋ฐ์ดํฐ๋ฅผ ๋ถ๋ฌ์ต๋๋ค.")
|
100 |
+
df = pd.read_csv(CSV_PATH)
|
101 |
+
st.sidebar.success(f"CSV ๋ฐ์ดํฐ ๋ก๋ ์๋ฃ: {len(df)}๊ฐ ํ")
|
102 |
+
else:
|
103 |
+
st.error("๐พ price_data.csv ๋๋ domae-202503.parquet ํ์ผ์ ์ฐพ์ ์ ์์ต๋๋ค.")
|
104 |
+
st.stop()
|
105 |
+
|
106 |
+
# ์๋ณธ ๋ฐ์ดํฐ ํํ ํ์ธ
|
107 |
+
st.sidebar.write("์๋ณธ ๋ฐ์ดํฐ ์ปฌ๋ผ:", list(df.columns))
|
108 |
+
|
109 |
+
df = _standardize_columns(df)
|
110 |
+
st.sidebar.write("ํ์คํ ํ ์ปฌ๋ผ:", list(df.columns))
|
111 |
+
|
112 |
+
missing = {c for c in ["date", "item", "price"] if c not in df.columns}
|
113 |
+
if missing:
|
114 |
+
st.error(f"ํ์ ์ปฌ๋ผ ๋๋ฝ: {', '.join(missing)} โ ํ์ผ ์ปฌ๋ผ๋ช
์ ํ์ธํ์ธ์.")
|
115 |
+
st.stop()
|
116 |
+
|
117 |
+
# ๋ ์ง ๋ณํ ์ ํ ๋ฐ์ดํฐ ์ ํ์ธ
|
118 |
+
before_date_convert = len(df)
|
119 |
+
df["date"] = pd.to_datetime(df["date"], errors="coerce")
|
120 |
+
after_date_convert = df.dropna(subset=["date"]).shape[0]
|
121 |
+
if before_date_convert != after_date_convert:
|
122 |
+
st.warning(f"๋ ์ง ๋ณํ ์ค {before_date_convert - after_date_convert}๊ฐ ํ์ด ์ ์ธ๋์์ต๋๋ค.")
|
123 |
+
|
124 |
+
# NA ๋ฐ์ดํฐ ์ฒ๋ฆฌ
|
125 |
+
before_na_drop = len(df)
|
126 |
+
df = df.dropna(subset=["date", "item", "price"])
|
127 |
+
after_na_drop = len(df)
|
128 |
+
if before_na_drop != after_na_drop:
|
129 |
+
st.warning(f"NA ์ ๊ฑฐ ์ค {before_na_drop - after_na_drop}๊ฐ ํ์ด ์ ์ธ๋์์ต๋๋ค.")
|
130 |
+
|
131 |
+
df.sort_values("date", inplace=True)
|
132 |
+
|
133 |
+
# ๏ฟฝ๏ฟฝ์ดํฐ ๋ ์ง ๋ฒ์ ํ์ธ
|
134 |
+
if len(df) > 0:
|
135 |
+
st.sidebar.write(f"๋ฐ์ดํฐ ๋ ์ง ๋ฒ์: {df['date'].min().strftime('%Y-%m-%d')} ~ {df['date'].max().strftime('%Y-%m-%d')}")
|
136 |
+
st.sidebar.write(f"์ด ํ๋ชฉ ์: {df['item'].nunique()}")
|
137 |
+
else:
|
138 |
+
st.error("์ ํจํ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค!")
|
139 |
+
|
140 |
+
return df
|
141 |
+
except Exception as e:
|
142 |
+
st.error(f"๋ฐ์ดํฐ ๋ก๋ ์ค ์ค๋ฅ ๋ฐ์: {str(e)}")
|
143 |
st.stop()
|
144 |
|
|
|
|
|
|
|
|
|
|
|
145 |
|
146 |
@st.cache_data(show_spinner=False)
|
147 |
def get_items(df: pd.DataFrame):
|
148 |
return sorted(df["item"].unique())
|
149 |
|
150 |
|
151 |
+
@st.cache_data(show_spinner=False, ttl=3600)
|
152 |
def fit_prophet(df: pd.DataFrame, horizon_end: str):
|
153 |
# Make a copy and ensure we have data
|
154 |
df = df.copy()
|
155 |
df = df.dropna(subset=["date", "price"])
|
156 |
|
157 |
+
# ์ค๋ณต ๋ ์ง ์ฒ๋ฆฌ - ๋์ผ ๋ ์ง์ ์ฌ๋ฌ ๊ฐ์ด ์์ผ๋ฉด ํ๊ท ๊ฐ ์ฌ์ฉ
|
158 |
+
df = df.groupby("date")["price"].mean().reset_index()
|
159 |
+
|
160 |
if len(df) < 2:
|
161 |
+
st.warning(f"๋ฐ์ดํฐ ํฌ์ธํธ๊ฐ ๋ถ์กฑํฉ๋๋ค. ์์ธก์ ์ํด์๋ ์ต์ 2๊ฐ ์ด์์ ์ ํจ ๋ฐ์ดํฐ๊ฐ ํ์ํฉ๋๋ค. (ํ์ฌ {len(df)}๊ฐ)")
|
162 |
return None, None
|
163 |
|
164 |
# Convert to Prophet format
|
165 |
prophet_df = df.rename(columns={"date": "ds", "price": "y"})
|
166 |
|
167 |
+
try:
|
168 |
+
# Fit the model
|
169 |
+
m = Prophet(yearly_seasonality=True, weekly_seasonality=False, daily_seasonality=False)
|
170 |
+
m.fit(prophet_df)
|
171 |
+
|
172 |
+
# Generate future dates
|
173 |
+
periods = max((pd.Timestamp(horizon_end) - df["date"].max()).days, 1)
|
174 |
+
future = m.make_future_dataframe(periods=periods, freq="D")
|
175 |
+
|
176 |
+
# Make predictions
|
177 |
+
forecast = m.predict(future)
|
178 |
+
return m, forecast
|
179 |
+
except Exception as e:
|
180 |
+
st.error(f"Prophet ๋ชจ๋ธ ์์ฑ ์ค ์ค๋ฅ: {str(e)}")
|
181 |
+
return None, None
|
182 |
|
183 |
# -------------------------------------------------
|
184 |
# LOAD DATA ---------------------------------------
|
185 |
# -------------------------------------------------
|
186 |
raw_df = load_data()
|
187 |
|
188 |
+
if len(raw_df) == 0:
|
189 |
+
st.error("๋ฐ์ดํฐ๊ฐ ๋น์ด ์์ต๋๋ค. ํ์ผ์ ํ์ธํด์ฃผ์ธ์.")
|
190 |
+
st.stop()
|
191 |
+
|
192 |
st.sidebar.header("๐ ํ๋ชฉ ์ ํ")
|
193 |
selected_item = st.sidebar.selectbox("ํ๋ชฉ", get_items(raw_df))
|
194 |
current_date = date.today()
|
|
|
203 |
# MACRO FORECAST 1996โ2030 ------------------------
|
204 |
# -------------------------------------------------
|
205 |
st.header(f"๐ {selected_item} ๊ฐ๊ฒฉ ์์ธก ๋์๋ณด๋")
|
206 |
+
|
207 |
+
# ๋ฐ์ดํฐ ํํฐ๋ง ๋ก์ง ๊ฐ์ - ์๊ฐ ๋ฒ์๋ฅผ ์กฐ์ ํ์ฌ ๋ ๋ง์ ๋ฐ์ดํฐ ํฌํจ
|
208 |
+
try:
|
209 |
+
macro_start_dt = pd.Timestamp(MACRO_START)
|
210 |
+
# ๋ฐ์ดํฐ๊ฐ ์ถฉ๋ถํ์ง ์์ผ๋ฉด ์์ ๋ ์ง๋ฅผ ์กฐ์
|
211 |
+
if len(item_df[item_df["date"] >= macro_start_dt]) < 10:
|
212 |
+
# ๊ฐ์ฅ ์ค๋๋ ๋ ์ง๋ถํฐ ์์
|
213 |
+
macro_start_dt = item_df["date"].min()
|
214 |
+
st.info(f"์ถฉ๋ถํ ๋ฐ์ดํฐ๊ฐ ์์ด ์์ ๋ ์ง๋ฅผ {macro_start_dt.strftime('%Y-%m-%d')}๋ก ์กฐ์ ํ์ต๋๋ค.")
|
215 |
+
|
216 |
+
macro_df = item_df[item_df["date"] >= macro_start_dt].copy()
|
217 |
+
except Exception as e:
|
218 |
+
st.error(f"๋ ์ง ํํฐ๋ง ์ค๋ฅ: {str(e)}")
|
219 |
+
macro_df = item_df.copy() # ํํฐ๋ง ์์ด ์ ์ฒด ๋ฐ์ดํฐ ์ฌ์ฉ
|
220 |
|
221 |
# Add diagnostic info
|
222 |
with st.expander("๋ฐ์ดํฐ ์ง๋จ"):
|
223 |
st.write(f"- ์ ์ฒด ๋ฐ์ดํฐ ์: {len(item_df)}")
|
224 |
+
st.write(f"- ๋ถ์ ๋ฐ์ดํฐ ์: {len(macro_df)}")
|
225 |
+
if len(macro_df) > 0:
|
226 |
+
st.write(f"- ๊ธฐ๊ฐ: {macro_df['date'].min().strftime('%Y-%m-%d')} ~ {macro_df['date'].max().strftime('%Y-%m-%d')}")
|
227 |
+
st.dataframe(macro_df.head())
|
228 |
+
else:
|
229 |
+
st.write("๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|
230 |
|
231 |
if len(macro_df) < 2:
|
232 |
+
st.warning(f"{selected_item}์ ๋ํ ๋ฐ์ดํฐ๊ฐ ์ถฉ๋ถํ์ง ์์ต๋๋ค. ์ ์ฒด ๊ธฐ๊ฐ ๋ฐ์ดํฐ๋ฅผ ํ์ํฉ๋๋ค.")
|
233 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
234 |
st.plotly_chart(fig, use_container_width=True)
|
235 |
else:
|
236 |
try:
|
237 |
+
with st.spinner("์ฅ๊ธฐ ์์ธก ๋ชจ๋ธ ์์ฑ ์ค..."):
|
238 |
+
m_macro, fc_macro = fit_prophet(macro_df, MACRO_END)
|
239 |
+
|
240 |
if m_macro is not None and fc_macro is not None:
|
241 |
+
fig_macro = px.line(fc_macro, x="ds", y="yhat", title="์ฅ๊ธฐ ์์ธก (1996โ2030)")
|
242 |
+
fig_macro.add_scatter(x=macro_df["date"], y=macro_df["price"], mode="lines", name="์ค์ ๊ฐ๊ฒฉ")
|
243 |
st.plotly_chart(fig_macro, use_container_width=True)
|
244 |
|
245 |
latest_price = macro_df.iloc[-1]["price"]
|
246 |
+
# 2030๋
๋ง์ง๋ง ๋ ์ฐพ๊ธฐ
|
247 |
+
target_date = pd.Timestamp(MACRO_END)
|
248 |
+
close_dates = fc_macro.loc[(fc_macro["ds"] - target_date).abs().argsort()[:1], "ds"].values[0]
|
249 |
+
macro_pred = fc_macro.loc[fc_macro["ds"] == close_dates, "yhat"].iloc[0]
|
250 |
macro_pct = (macro_pred - latest_price) / latest_price * 100
|
251 |
st.metric("2030 ์์ธก๊ฐ", f"{macro_pred:,.0f}", f"{macro_pct:+.1f}%")
|
252 |
else:
|
|
|
254 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
255 |
st.plotly_chart(fig, use_container_width=True)
|
256 |
except Exception as e:
|
257 |
+
st.error(f"์ฅ๊ธฐ ์์ธก ์ค๋ฅ ๋ฐ์: {str(e)}")
|
258 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ๊ณผ๊ฑฐ ๊ฐ๊ฒฉ")
|
259 |
st.plotly_chart(fig, use_container_width=True)
|
260 |
|
|
|
263 |
# -------------------------------------------------
|
264 |
st.subheader("๐ 2024โ2026 ๋จ๊ธฐ ์์ธก")
|
265 |
|
266 |
+
# ๋ฐ์ดํฐ ํํฐ๋ง ๋ก์ง ๊ฐ์
|
267 |
+
try:
|
268 |
+
micro_start_dt = pd.Timestamp(MICRO_START)
|
269 |
+
# ๋ฐ์ดํฐ๊ฐ ์ถฉ๋ถํ์ง ์์ผ๋ฉด ์์ ๋ ์ง๋ฅผ ์กฐ์
|
270 |
+
if len(item_df[item_df["date"] >= micro_start_dt]) < 10:
|
271 |
+
# ์ต๊ทผ 30% ๋ฐ์ดํฐ๋ง ์ฌ์ฉ
|
272 |
+
n = max(2, int(len(item_df) * 0.3))
|
273 |
+
micro_df = item_df.sort_values("date").tail(n).copy()
|
274 |
+
st.info(f"์ถฉ๋ถํ ์ต๊ทผ ๋ฐ์ดํฐ๊ฐ ์์ด ์ต๊ทผ {n}๊ฐ ๋ฐ์ดํฐ ํฌ์ธํธ๋ง ์ฌ์ฉํฉ๋๋ค.")
|
275 |
+
else:
|
276 |
+
micro_df = item_df[item_df["date"] >= micro_start_dt].copy()
|
277 |
+
except Exception as e:
|
278 |
+
st.error(f"๋จ๊ธฐ ์์ธก ๋ฐ์ดํฐ ํํฐ๋ง ์ค๋ฅ: {str(e)}")
|
279 |
+
# ์ต๊ทผ 10๊ฐ ๋ฐ์ดํฐ ํฌ์ธํธ ์ฌ์ฉ
|
280 |
+
micro_df = item_df.sort_values("date").tail(10).copy()
|
281 |
+
|
282 |
if len(micro_df) < 2:
|
283 |
st.warning(f"{MICRO_START} ์ดํ ๋ฐ์ดํฐ๊ฐ ์ถฉ๋ถํ์ง ์์ต๋๋ค.")
|
284 |
fig = px.line(item_df, x="date", y="price", title=f"{selected_item} ์ต๊ทผ ๊ฐ๊ฒฉ")
|
285 |
st.plotly_chart(fig, use_container_width=True)
|
286 |
else:
|
287 |
try:
|
288 |
+
with st.spinner("๋จ๊ธฐ ์์ธก ๋ชจ๋ธ ์์ฑ ์ค..."):
|
289 |
+
m_micro, fc_micro = fit_prophet(micro_df, MICRO_END)
|
290 |
+
|
291 |
if m_micro is not None and fc_micro is not None:
|
292 |
+
fig_micro = px.line(fc_micro, x="ds", y="yhat", title="๋จ๊ธฐ ์์ธก (2024โ2026)")
|
293 |
+
fig_micro.add_scatter(x=micro_df["date"], y=micro_df["price"], mode="lines", name="์ค์ ๊ฐ๊ฒฉ")
|
294 |
st.plotly_chart(fig_micro, use_container_width=True)
|
295 |
|
296 |
latest_price = micro_df.iloc[-1]["price"]
|
297 |
+
target_date = pd.Timestamp(MICRO_END)
|
298 |
+
close_dates = fc_micro.loc[(fc_micro["ds"] - target_date).abs().argsort()[:1], "ds"].values[0]
|
299 |
+
micro_pred = fc_micro.loc[fc_micro["ds"] == close_dates, "yhat"].iloc[0]
|
300 |
micro_pct = (micro_pred - latest_price) / latest_price * 100
|
301 |
st.metric("2026 ์์ธก๊ฐ", f"{micro_pred:,.0f}", f"{micro_pct:+.1f}%")
|
302 |
else:
|
|
|
309 |
# -------------------------------------------------
|
310 |
with st.expander("๐ ์์ฆ๋๋ฆฌํฐ & ํจํด ์ค๋ช
"):
|
311 |
if 'm_micro' in locals() and m_micro is not None and 'fc_micro' in locals() and fc_micro is not None:
|
312 |
+
try:
|
313 |
+
comp_fig = m_micro.plot_components(fc_micro)
|
314 |
+
st.pyplot(comp_fig)
|
315 |
+
|
316 |
+
month_season = (fc_micro[["ds", "yearly"]]
|
317 |
+
.assign(month=lambda d: d.ds.dt.month)
|
318 |
+
.groupby("month")["yearly"].mean())
|
319 |
+
st.markdown(
|
320 |
+
f"**์ฐ๊ฐ ํผํฌ ์:** {int(month_season.idxmax())}์ \n"
|
321 |
+
f"**์ฐ๊ฐ ์ ์ ์:** {int(month_season.idxmin())}์ \n"
|
322 |
+
f"**์ฐ๊ฐ ๋ณ๋ํญ:** {month_season.max() - month_season.min():.1f}")
|
323 |
+
except Exception as e:
|
324 |
+
st.error(f"์์ฆ๋๋ฆฌํฐ ๋ถ์ ์ค๋ฅ: {str(e)}")
|
325 |
else:
|
326 |
st.info("ํจํด ๋ถ์์ ์ํ ์ถฉ๋ถํ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|
327 |
|
|
|
329 |
# CORRELATION HEATMAP -----------------------------
|
330 |
# -------------------------------------------------
|
331 |
st.subheader("๐งฎ ํ๋ชฉ ๊ฐ ์๊ด๊ด๊ณ")
|
|
|
|
|
|
|
|
|
332 |
|
333 |
+
try:
|
334 |
+
# ๋๋ฌด ๋ง์ ํ๋ชฉ์ด ์์ผ๋ฉด ์์ N๊ฐ๋ง ์ ํ
|
335 |
+
items_to_corr = raw_df['item'].value_counts().head(30).index.tolist()
|
336 |
+
if selected_item not in items_to_corr and selected_item in raw_df['item'].unique():
|
337 |
+
items_to_corr.append(selected_item)
|
338 |
+
|
339 |
+
filtered_df = raw_df[raw_df['item'].isin(items_to_corr)]
|
340 |
+
|
341 |
+
monthly_pivot = (filtered_df.assign(month=lambda d: d.date.dt.to_period("M"))
|
342 |
+
.groupby(["month", "item"], as_index=False)["price"].mean()
|
343 |
+
.pivot(index="month", columns="item", values="price"))
|
344 |
+
|
345 |
+
# ๊ฒฐ์ธก์น๊ฐ ๋๋ฌด ๋ง์ ์ด ์ ๊ฑฐ
|
346 |
+
threshold = 0.5 # 50% ์ด์ ๊ฒฐ์ธก์น๊ฐ ์๋ ์ด ์ ๊ฑฐ
|
347 |
+
monthly_pivot = monthly_pivot.loc[:, monthly_pivot.isnull().mean() < threshold]
|
348 |
+
|
349 |
if monthly_pivot.shape[1] > 1: # At least 2 items needed for correlation
|
350 |
+
# ๊ฒฐ์ธก์น ์ฒ๋ฆฌ
|
351 |
+
monthly_pivot = monthly_pivot.fillna(method='ffill').fillna(method='bfill')
|
352 |
+
|
353 |
+
# ์๊ด๊ด๊ณ ๊ณ์ฐ
|
354 |
corr = monthly_pivot.corr()
|
355 |
+
|
356 |
+
# ์๊ฐํ
|
357 |
fig, ax = plt.subplots(figsize=(12, 10))
|
358 |
mask = np.triu(np.ones_like(corr, dtype=bool))
|
359 |
+
|
360 |
+
# ์ฌ๊ธฐ์ ํฐํธ ์ค์ ๋ค์ ํ์ธ
|
361 |
+
plt.title(f"{selected_item} ๊ด๋ จ ์๊ด๊ด๊ณ", fontsize=15)
|
362 |
+
|
363 |
+
sns.heatmap(corr, mask=mask, annot=False, cmap="coolwarm", center=0,
|
364 |
square=True, linewidths=.5, cbar_kws={"shrink": .5})
|
365 |
+
|
366 |
+
plt.xticks(rotation=45, ha='right', fontsize=8)
|
367 |
+
plt.yticks(fontsize=8)
|
368 |
|
369 |
# Highlight correlations with selected item
|
370 |
if selected_item in corr.columns:
|
|
|
387 |
st.info("์๊ด๊ด๊ณ ๋ถ์์ ์ํ ์ถฉ๋ถํ ํ๋ชฉ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|
388 |
except Exception as e:
|
389 |
st.error(f"์๊ด๊ด๊ณ ๋ถ์ ์ค๋ฅ: {str(e)}")
|
390 |
+
st.write("์ค๋ฅ ์์ธ ์ ๋ณด:", str(e))
|
391 |
|
392 |
# -------------------------------------------------
|
393 |
# FOOTER ------------------------------------------
|
394 |
# -------------------------------------------------
|
395 |
st.markdown("---")
|
396 |
+
st.caption("ยฉ 2025 ํ๋ชฉ๋ณ ๊ฐ๊ฒฉ ์์ธก ์์คํ
| ๋ฐ์ดํฐ ๋ถ์ ์๋ํ")
|