Spaces:
Running
Running
Update src/streamlit_app.py
Browse files- src/streamlit_app.py +51 -29
src/streamlit_app.py
CHANGED
|
@@ -130,13 +130,14 @@ def get_avg_daily_volumes(tickers_tuple):
|
|
| 130 |
"""Bulk-fetch ~1 month of daily bars to compute each ticker's typical daily volume."""
|
| 131 |
avg_vols = {}
|
| 132 |
batch_size = 150
|
| 133 |
-
|
|
|
|
| 134 |
progress = st.progress(0, text="Fetching volume baselines...")
|
| 135 |
|
| 136 |
-
for b, i in enumerate(range(0, len(
|
| 137 |
-
batch =
|
| 138 |
try:
|
| 139 |
-
data = yf.download(
|
| 140 |
progress=False, threads=True)
|
| 141 |
except Exception:
|
| 142 |
progress.progress((b + 1) / total_batches)
|
|
@@ -144,11 +145,18 @@ def get_avg_daily_volumes(tickers_tuple):
|
|
| 144 |
|
| 145 |
for ticker in batch:
|
| 146 |
try:
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
|
| 150 |
-
|
| 151 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
if len(vol) >= 5:
|
| 153 |
avg_vols[ticker] = float(vol.mean())
|
| 154 |
except Exception:
|
|
@@ -166,13 +174,14 @@ def run_premarket_scan(universe_label, tickers_tuple, min_pct_avg_vol, min_abs_v
|
|
| 166 |
|
| 167 |
results = []
|
| 168 |
batch_size = 150
|
| 169 |
-
|
|
|
|
| 170 |
progress = st.progress(0, text="Scanning premarket activity...")
|
| 171 |
|
| 172 |
-
for b, i in enumerate(range(0, len(
|
| 173 |
-
batch =
|
| 174 |
try:
|
| 175 |
-
data = yf.download(
|
| 176 |
group_by="ticker", progress=False, threads=True)
|
| 177 |
except Exception:
|
| 178 |
progress.progress((b + 1) / total_batches, text=f"Batch {b + 1}/{total_batches} failed, continuing...")
|
|
@@ -180,11 +189,18 @@ def run_premarket_scan(universe_label, tickers_tuple, min_pct_avg_vol, min_abs_v
|
|
| 180 |
|
| 181 |
for ticker in batch:
|
| 182 |
try:
|
| 183 |
-
|
| 184 |
-
|
| 185 |
-
|
| 186 |
-
|
| 187 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 188 |
if df.empty:
|
| 189 |
continue
|
| 190 |
|
|
@@ -193,7 +209,7 @@ def run_premarket_scan(universe_label, tickers_tuple, min_pct_avg_vol, min_abs_v
|
|
| 193 |
df.index = df.index.tz_convert("America/New_York")
|
| 194 |
else:
|
| 195 |
df = df.copy()
|
| 196 |
-
df.index = df.index.tz_localize("America/New_York")
|
| 197 |
|
| 198 |
premarket = df[df.index.time < dt_time(9, 30)]
|
| 199 |
if premarket.empty:
|
|
@@ -208,7 +224,7 @@ def run_premarket_scan(universe_label, tickers_tuple, min_pct_avg_vol, min_abs_v
|
|
| 208 |
if pm_volume < min_abs_volume or pct_of_avg < min_pct_avg_vol:
|
| 209 |
continue
|
| 210 |
|
| 211 |
-
open_price = premarket["
|
| 212 |
last_price = premarket["Close"].iloc[-1]
|
| 213 |
pct_change = ((last_price - open_price) / open_price * 100) if open_price else np.nan
|
| 214 |
|
|
@@ -553,7 +569,7 @@ def render_peak_detector():
|
|
| 553 |
|
| 554 |
|
| 555 |
# =========================================================
|
| 556 |
-
# TAB 3: SNIPER SCANNER
|
| 557 |
# =========================================================
|
| 558 |
|
| 559 |
def sniper_rsi(series, period=14):
|
|
@@ -693,11 +709,17 @@ def run_full_scan(universe_label, extra_tickers_tuple, rsi_threshold, stretch_th
|
|
| 693 |
|
| 694 |
for ticker in batch:
|
| 695 |
try:
|
| 696 |
-
|
| 697 |
-
|
| 698 |
-
|
| 699 |
-
|
| 700 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 701 |
if len(df) < 200:
|
| 702 |
continue
|
| 703 |
result = analyze_ticker(df, rsi_threshold, stretch_threshold, strict_window, lookback, win_rate_min)
|
|
@@ -906,7 +928,7 @@ def render_sniper_tab():
|
|
| 906 |
|
| 907 |
|
| 908 |
# =========================================================
|
| 909 |
-
# ANALOG PATH PROJECTION
|
| 910 |
# =========================================================
|
| 911 |
|
| 912 |
ANALOG_LOOKBACK = 60
|
|
@@ -945,7 +967,7 @@ def compute_analog_projection(ticker: str, flag_date_str: str, lookback=ANALOG_L
|
|
| 945 |
end_idx = min(end_idx, len(close) - 1)
|
| 946 |
|
| 947 |
if end_idx - lookback - forward < 5:
|
| 948 |
-
return None
|
| 949 |
|
| 950 |
matches = _similarity_matches(close, end_idx, lookback, forward)
|
| 951 |
if not matches:
|
|
@@ -991,7 +1013,7 @@ def compute_analog_projection(ticker: str, flag_date_str: str, lookback=ANALOG_L
|
|
| 991 |
|
| 992 |
|
| 993 |
# =========================================================
|
| 994 |
-
# TAB 4: SIGNAL TRACKER
|
| 995 |
# =========================================================
|
| 996 |
|
| 997 |
def render_signal_tracker():
|
|
|
|
| 130 |
"""Bulk-fetch ~1 month of daily bars to compute each ticker's typical daily volume."""
|
| 131 |
avg_vols = {}
|
| 132 |
batch_size = 150
|
| 133 |
+
tickers_list = list(tickers_tuple)
|
| 134 |
+
total_batches = max(1, (len(tickers_list) + batch_size - 1) // batch_size)
|
| 135 |
progress = st.progress(0, text="Fetching volume baselines...")
|
| 136 |
|
| 137 |
+
for b, i in enumerate(range(0, len(tickers_list), batch_size)):
|
| 138 |
+
batch = tickers_list[i:i + batch_size]
|
| 139 |
try:
|
| 140 |
+
data = yf.download(batch, period="1mo", interval="1d", group_by="ticker",
|
| 141 |
progress=False, threads=True)
|
| 142 |
except Exception:
|
| 143 |
progress.progress((b + 1) / total_batches)
|
|
|
|
| 145 |
|
| 146 |
for ticker in batch:
|
| 147 |
try:
|
| 148 |
+
# Safe MultiIndex column extraction checking
|
| 149 |
+
if isinstance(data.columns, pd.MultiIndex):
|
| 150 |
+
if ticker in data.columns.levels[0]:
|
| 151 |
+
vol = data[ticker]["Volume"].dropna()
|
| 152 |
+
else:
|
| 153 |
+
continue
|
| 154 |
+
else:
|
| 155 |
+
if len(batch) == 1:
|
| 156 |
+
vol = data["Volume"].dropna()
|
| 157 |
+
else:
|
| 158 |
+
continue
|
| 159 |
+
|
| 160 |
if len(vol) >= 5:
|
| 161 |
avg_vols[ticker] = float(vol.mean())
|
| 162 |
except Exception:
|
|
|
|
| 174 |
|
| 175 |
results = []
|
| 176 |
batch_size = 150
|
| 177 |
+
tickers_list = list(tickers_tuple)
|
| 178 |
+
total_batches = max(1, (len(tickers_list) + batch_size - 1) // batch_size)
|
| 179 |
progress = st.progress(0, text="Scanning premarket activity...")
|
| 180 |
|
| 181 |
+
for b, i in enumerate(range(0, len(tickers_list), batch_size)):
|
| 182 |
+
batch = tickers_list[i:i + batch_size]
|
| 183 |
try:
|
| 184 |
+
data = yf.download(batch, period="1d", interval="1m", prepost=True,
|
| 185 |
group_by="ticker", progress=False, threads=True)
|
| 186 |
except Exception:
|
| 187 |
progress.progress((b + 1) / total_batches, text=f"Batch {b + 1}/{total_batches} failed, continuing...")
|
|
|
|
| 189 |
|
| 190 |
for ticker in batch:
|
| 191 |
try:
|
| 192 |
+
# Bulletproof structural check for MultiIndex extraction layers
|
| 193 |
+
if isinstance(data.columns, pd.MultiIndex):
|
| 194 |
+
if ticker in data.columns.levels[0]:
|
| 195 |
+
df = data[ticker].dropna(subset=["Close"])
|
| 196 |
+
else:
|
| 197 |
+
continue
|
| 198 |
+
else:
|
| 199 |
+
if len(batch) == 1:
|
| 200 |
+
df = data.dropna(subset=["Close"])
|
| 201 |
+
else:
|
| 202 |
+
continue
|
| 203 |
+
|
| 204 |
if df.empty:
|
| 205 |
continue
|
| 206 |
|
|
|
|
| 209 |
df.index = df.index.tz_convert("America/New_York")
|
| 210 |
else:
|
| 211 |
df = df.copy()
|
| 212 |
+
df.index = df.index.tz_localize("UTC").tz_convert("America/New_York")
|
| 213 |
|
| 214 |
premarket = df[df.index.time < dt_time(9, 30)]
|
| 215 |
if premarket.empty:
|
|
|
|
| 224 |
if pm_volume < min_abs_volume or pct_of_avg < min_pct_avg_vol:
|
| 225 |
continue
|
| 226 |
|
| 227 |
+
open_price = premarket["Open"].iloc[0]
|
| 228 |
last_price = premarket["Close"].iloc[-1]
|
| 229 |
pct_change = ((last_price - open_price) / open_price * 100) if open_price else np.nan
|
| 230 |
|
|
|
|
| 569 |
|
| 570 |
|
| 571 |
# =========================================================
|
| 572 |
+
# TAB 3: SNIPER SCANNER
|
| 573 |
# =========================================================
|
| 574 |
|
| 575 |
def sniper_rsi(series, period=14):
|
|
|
|
| 709 |
|
| 710 |
for ticker in batch:
|
| 711 |
try:
|
| 712 |
+
if isinstance(data.columns, pd.MultiIndex):
|
| 713 |
+
if ticker in data.columns.levels[0]:
|
| 714 |
+
df = data[ticker].dropna()
|
| 715 |
+
else:
|
| 716 |
+
continue
|
| 717 |
+
else:
|
| 718 |
+
if len(batch) == 1:
|
| 719 |
+
df = data.dropna()
|
| 720 |
+
else:
|
| 721 |
+
continue
|
| 722 |
+
|
| 723 |
if len(df) < 200:
|
| 724 |
continue
|
| 725 |
result = analyze_ticker(df, rsi_threshold, stretch_threshold, strict_window, lookback, win_rate_min)
|
|
|
|
| 928 |
|
| 929 |
|
| 930 |
# =========================================================
|
| 931 |
+
# ANALOG PATH PROJECTION
|
| 932 |
# =========================================================
|
| 933 |
|
| 934 |
ANALOG_LOOKBACK = 60
|
|
|
|
| 967 |
end_idx = min(end_idx, len(close) - 1)
|
| 968 |
|
| 969 |
if end_idx - lookback - forward < 5:
|
| 970 |
+
return None
|
| 971 |
|
| 972 |
matches = _similarity_matches(close, end_idx, lookback, forward)
|
| 973 |
if not matches:
|
|
|
|
| 1013 |
|
| 1014 |
|
| 1015 |
# =========================================================
|
| 1016 |
+
# TAB 4: SIGNAL TRACKER
|
| 1017 |
# =========================================================
|
| 1018 |
|
| 1019 |
def render_signal_tracker():
|