Upload folder using huggingface_hub

app.py

@@ -6,10 +6,13 @@ from fastapi.middleware.cors import CORSMiddleware
 from zoneinfo import ZoneInfo
 from data_updater import update_daily_data, is_trading_day
 from forecaster_engine import generate_predictions
+from signal_generator import generate_signals
 
 IST = ZoneInfo("Asia/Kolkata")
 MARKET_CLOSE_BUFFER = time(15, 45) # Update runs after 3:45 PM
+SIGNAL_TIME = time(9, 30)         # Signal generation at 9:30 AM
 PREDICTIONS_FILE = os.path.join(os.path.dirname(__file__), "predictions.json")
+SIGNALS_FILE = os.path.join(os.path.dirname(__file__), "signals.json")
 
 app = FastAPI(title="HF NIFTY Forecaster Backend")
 
@@ -33,6 +36,16 @@ def run_update_pipeline():
     except Exception as e:
         print(f"Pipeline error: {e}")
 
+def run_signal_pipeline():
+    """Run the 5-ticker signal generator."""
+    try:
+        result = generate_signals()
+        print(f"Signal generation result: {result.get('primary_signal', {}).get('action', 'UNKNOWN')}")
+    except Exception as e:
+        print(f"Signal pipeline error: {e}")
+
+# ── Existing Endpoints ───────────────────────────────────────────────────────
+
 @app.get("/predictions")
 def get_predictions():
     if not os.path.exists(PREDICTIONS_FILE):
@@ -62,6 +75,87 @@ def cron_trigger(background_tasks: BackgroundTasks):
     
     return {"status": "triggered", "message": "Update and forecast pipeline started in the background."}
 
+# ── NEW: Signal Generator Endpoints ──────────────────────────────────────────
+
+@app.get("/signals")
+def get_signals():
+    """Get the latest generated trading signals for the 5-ticker system."""
+    if not os.path.exists(SIGNALS_FILE):
+        raise HTTPException(status_code=404, detail="Signals not yet generated. Trigger /cron/signal first.")
+    
+    with open(SIGNALS_FILE, "r") as f:
+        data = json.load(f)
+        
+    return data
+
+@app.post("/cron/signal")
+def signal_trigger(background_tasks: BackgroundTasks):
+    """
+    Trigger signal generation at 9:30 AM IST.
+    Trains models, fetches live candles, generates BUY/SELL signals.
+    """
+    now = datetime.now(IST)
+    today = now.date()
+    
+    # Check if it's a trading day
+    if not is_trading_day(today):
+        return {"status": "skipped", "reason": f"{today} is a holiday or weekend"}
+    
+    # Run signal generation in background
+    background_tasks.add_task(run_signal_pipeline)
+    
+    return {
+        "status": "triggered",
+        "message": "Signal generation pipeline started. Check /signals for results.",
+        "trigger_time": now.isoformat(),
+    }
+
+@app.post("/signals/generate-now")
+def force_signal_generation(background_tasks: BackgroundTasks):
+    """Force signal generation immediately, bypassing time checks."""
+    background_tasks.add_task(run_signal_pipeline)
+    return {
+        "status": "triggered",
+        "message": "Signal generation forced. Check /signals for results.",
+        "trigger_time": datetime.now(IST).isoformat(),
+    }
+
+@app.get("/portfolio")
+def get_portfolio():
+    """Get current portfolio status from trade journal."""
+    trade_log = os.path.join(os.path.dirname(__file__), "data", "live_trades.json")
+    if not os.path.exists(trade_log):
+        return {
+            "starting_capital": 3692.0,
+            "current_capital": 3692.0,
+            "total_pnl": 0,
+            "trades_count": 0,
+            "win_rate": 0,
+        }
+    
+    with open(trade_log, "r") as f:
+        data = json.load(f)
+    
+    trades = data.get("trades", [])
+    starting_cap = data.get("starting_capital", 3692.0)
+    
+    cap = starting_cap
+    for t in trades:
+        if "net_pnl" in t and t["net_pnl"] is not None:
+            cap += t["net_pnl"]
+    
+    n_closed = len([t for t in trades if t.get("net_pnl") is not None])
+    n_wins = len([t for t in trades if (t.get("net_pnl") or 0) > 0])
+    
+    return {
+        "starting_capital": starting_cap,
+        "current_capital": round(cap, 2),
+        "total_pnl": round(cap - starting_cap, 2),
+        "trades_count": n_closed,
+        "win_rate": round(n_wins / n_closed * 100, 1) if n_closed > 0 else 0,
+        "last_updated": data.get("last_updated"),
+    }
+
 @app.get("/health")
 def health_check():
     return {"status": "alive", "server_time_ist": datetime.now(IST).isoformat()}

core/__init__.py

@@ -0,0 +1,6 @@
+# core/__init__.py
+from core.config import *
+from core.taxes import calculate_taxes_and_slippage
+from core.features import extract_semantic_features, extract_sequential_features
+from core.models import build_pipeline_map, train_models
+from core.groww import fetch_groww_candles

core/config.py

@@ -0,0 +1,37 @@
+"""
+Central configuration for the intraday trading system.
+All constants, paths, and model architecture definitions live here.
+"""
+
+from pathlib import Path
+
+# ── Paths ──
+PROJECT_DIR = Path(__file__).resolve().parent.parent
+DATA_DIR    = PROJECT_DIR / "data" / "minute_ohlcv"
+TRADE_LOG   = PROJECT_DIR / "data" / "live_trades.json"
+LOG_FILE    = PROJECT_DIR / "logs" / "live_trader.log"
+
+# ── Universe ──
+TICKERS = ["INFY", "ASIANPAINT", "TECHM", "POWERGRID", "ONGC"]
+
+# ── Capital & Risk ──
+STARTING_CAP   = 3692.0
+LEVERAGE       = 5.0
+MIN_CONFIDENCE = 0.50
+
+# ── Schedule (IST) ──
+SIGNAL_HOUR   = 9
+SIGNAL_MINUTE = 31
+
+# ── Pipeline mapping ──
+#    Defines which feature extractor and model architecture each ticker uses.
+#    Format:  { ticker: (feature_type, model_key) }
+#       feature_type : "semantic" | "sequential"
+#       model_key    : "ensemble" | "lr_pipeline"
+PIPELINE_MAP = {
+    "INFY":       ("semantic",   "ensemble"),
+    "TECHM":      ("sequential", "kbest15_lr"),
+    "ASIANPAINT": ("sequential", "ensemble"),
+    "POWERGRID":  ("sequential", "ensemble"),
+    "ONGC":       ("semantic",   "lr_pipeline"),
+}

core/features.py

@@ -0,0 +1,202 @@
+"""
+Feature extraction from 1-min OHLCV DataFrames.
+
+Two pipelines:
+  - semantic:   Hand-crafted morning indicators (gap, VWAP deviation, momentum, etc.)
+  - sequential: Raw normalised return and volume vectors for the 09:15-09:30 window.
+
+Both return (X, y, metadata) where:
+  X        : pd.DataFrame of features, indexed by date
+  y        : pd.Series of binary labels (1 = price up 09:30->15:10)
+  metadata : dict[date] -> { c_0930, h_0930, l_0930, v_0930, c_1510, h_1510, l_1510 }
+"""
+
+import numpy as np
+import pandas as pd
+
+
+# ── Helpers ──────────────────────────────────────────────────────────────────
+
+def _safe_fill(arr):
+    """Forward-fill then back-fill NaNs in a 1-D array."""
+    return pd.Series(arr).ffill().bfill().values
+
+
+# ── Semantic features ────────────────────────────────────────────────────────
+
+def extract_semantic_features(df):
+    """
+    Compute high-level morning indicators from the 09:15-09:30 candle window.
+
+    Features: gap, return_16m, return_first_5m, return_next_11m, std_dev,
+              range_pct, upper_shadow, lower_shadow, total_vol, vwap_dev,
+              price_momentum, vol_momentum, morning_trend.
+    """
+    df = df.copy()
+    df["time"]     = df.index.time
+    df["date_only"] = df.index.date
+    required_times = (
+        pd.date_range("09:15", "09:30", freq="min").time.tolist()
+        + [pd.to_datetime("15:10").time()]
+    )
+    df = df[~df.index.duplicated(keep="first")]
+
+    daily_close      = df.groupby("date_only")["close"].last()
+    prev_daily_close = daily_close.shift(1)
+
+    # Compute morning session dip/peak (09:31 to 12:00) for limit-order entries
+    time_0931 = pd.to_datetime("09:31").time()
+    time_1200 = pd.to_datetime("12:00").time()
+    df_morning = df[(df["time"] >= time_0931) & (df["time"] <= time_1200)]
+    morning_low_per_date  = df_morning.groupby("date_only")["low"].min()
+    morning_high_per_date = df_morning.groupby("date_only")["high"].max()
+
+    df_filtered = df[df["time"].isin(required_times)].copy()
+    pivot_close = df_filtered.pivot(index="date_only", columns="time", values="close")
+    pivot_open  = df_filtered.pivot(index="date_only", columns="time", values="open")
+    pivot_high  = df_filtered.pivot(index="date_only", columns="time", values="high")
+    pivot_low   = df_filtered.pivot(index="date_only", columns="time", values="low")
+    pivot_vol   = df_filtered.pivot(index="date_only", columns="time", values="volume")
+
+    time_0930 = pd.to_datetime("09:30").time()
+    time_1510 = pd.to_datetime("15:10").time()
+
+    if time_0930 not in pivot_close.columns:
+        return None, None, None
+
+    pivot_close = pivot_close.dropna(subset=[time_0930])
+    valid_dates = pivot_close.index
+    times_15m   = pd.date_range("09:15", "09:30", freq="min").time
+
+    feature_dicts = []
+    metadata      = {}
+
+    for date in valid_dates:
+        f = {}
+        c_series = _safe_fill(pivot_close.loc[date, times_15m].values.astype(float))
+        o_series = _safe_fill(pivot_open.loc[date, times_15m].values.astype(float))
+        h_series = _safe_fill(pivot_high.loc[date, times_15m].values.astype(float))
+        l_series = _safe_fill(pivot_low.loc[date, times_15m].values.astype(float))
+        v_series = pd.Series(pivot_vol.loc[date, times_15m].values.astype(float)).fillna(0).values
+
+        o_915 = o_series[0]
+        c_930 = c_series[-1]
+        c_919 = c_series[4] if len(c_series) > 4 else c_series[-1]
+
+        pdc = prev_daily_close.get(date, np.nan)
+        f["gap"]             = 0 if (pd.isna(pdc) or pdc == 0) else (o_915 / pdc) - 1.0
+        f["return_16m"]      = (c_930 / o_915) - 1.0 if o_915 != 0 else 0
+        f["return_first_5m"] = (c_919 / o_915) - 1.0 if o_915 != 0 else 0
+        f["return_next_11m"] = (c_930 / c_919) - 1.0 if c_919 != 0 else 0
+        f["std_dev"]         = np.std(c_series / (o_915 + 1e-8))
+
+        max_h = np.max(h_series)
+        min_l = np.min(l_series)
+        f["range_pct"]       = (max_h - min_l) / (o_915 + 1e-8)
+        f["upper_shadow"]    = (max_h - max(o_915, c_930)) / (o_915 + 1e-8)
+        f["lower_shadow"]    = (min(o_915, c_930) - min_l) / (o_915 + 1e-8)
+        f["total_vol"]       = np.sum(v_series)
+
+        vwap = np.sum(((h_series + l_series + c_series) / 3.0) * v_series) / (np.sum(v_series) + 1e-8)
+        f["vwap_dev"]        = (c_930 / vwap) - 1.0 if vwap != 0 else 0
+        f["price_momentum"]  = (c_series[-1] - c_series[-3]) / (c_series[-3] + 1e-8)
+        f["vol_momentum"]    = (v_series[-1] - v_series[-3]) / (v_series[-3] + 1e-8)
+        f["morning_trend"]   = np.polyfit(np.arange(len(c_series)), c_series, 1)[0]
+
+        feature_dicts.append(f)
+
+        metadata[date] = {
+            "c_0930": c_930,
+            "h_0930": float(pivot_high.loc[date, time_0930]) if time_0930 in pivot_high.columns else c_930,
+            "l_0930": float(pivot_low.loc[date, time_0930])  if time_0930 in pivot_low.columns  else c_930,
+            "v_0930": float(pivot_vol.loc[date, time_0930])  if time_0930 in pivot_vol.columns  else 0,
+            "c_1510": float(pivot_close.loc[date, time_1510]) if time_1510 in pivot_close.columns else c_930,
+            "h_1510": float(pivot_high.loc[date, time_1510])  if time_1510 in pivot_high.columns else c_930,
+            "l_1510": float(pivot_low.loc[date, time_1510])   if time_1510 in pivot_low.columns  else c_930,
+            "dip_low":   float(morning_low_per_date.get(date, c_930)),
+            "peak_high": float(morning_high_per_date.get(date, c_930)),
+        }
+
+    X = pd.DataFrame(feature_dicts, index=valid_dates).fillna(0)
+
+    if time_1510 in pivot_close.columns:
+        target = (pivot_close[time_1510] > pivot_close[time_0930]).astype(int)
+    else:
+        target = pd.Series(0, index=valid_dates)
+
+    return X, target, metadata
+
+
+# ── Sequential features ──────────────────────────────────────────────────────
+
+def extract_sequential_features(df):
+    """
+    Compute normalised return and raw volume vectors for each minute
+    in the 09:15-09:30 window, relative to the 09:30 close.
+    """
+    df = df.copy()
+    df["time"]      = df.index.time
+    df["date_only"] = df.index.date
+    required_times = (
+        pd.date_range("09:15", "09:30", freq="min").time.tolist()
+        + [pd.to_datetime("15:10").time()]
+    )
+    df = df[~df.index.duplicated(keep="first")]
+
+    # Compute morning session dip/peak (09:31 to 12:00) for limit-order entries
+    time_0931 = pd.to_datetime("09:31").time()
+    time_1200 = pd.to_datetime("12:00").time()
+    df_morning = df[(df["time"] >= time_0931) & (df["time"] <= time_1200)]
+    morning_low_per_date  = df_morning.groupby("date_only")["low"].min()
+    morning_high_per_date = df_morning.groupby("date_only")["high"].max()
+
+    df_filtered = df[df["time"].isin(required_times)].copy()
+    pivot_close = df_filtered.pivot(index="date_only", columns="time", values="close")
+    pivot_high  = df_filtered.pivot(index="date_only", columns="time", values="high")
+    pivot_low   = df_filtered.pivot(index="date_only", columns="time", values="low")
+    pivot_vol   = df_filtered.pivot(index="date_only", columns="time", values="volume")
+
+    time_0930 = pd.to_datetime("09:30").time()
+    time_1510 = pd.to_datetime("15:10").time()
+
+    if time_0930 not in pivot_close.columns:
+        return None, None, None
+
+    pivot_close = pivot_close.dropna(subset=[time_0930])
+    valid_dates = pivot_close.index
+    times_15m   = pd.date_range("09:15", "09:30", freq="min").time
+
+    feature_dicts = []
+    metadata      = {}
+
+    for date in valid_dates:
+        f = {}
+        c_series = _safe_fill(pivot_close.loc[date, times_15m].values.astype(float))
+        v_series = pd.Series(pivot_vol.loc[date, times_15m].values.astype(float)).fillna(0).values
+        c_ref    = c_series[-1]
+
+        for i, t in enumerate(times_15m):
+            f[f"ret_c_{i}"]   = (c_series[i] / (c_ref + 1e-8)) - 1.0
+            f[f"raw_vol_{i}"] = v_series[i]
+        feature_dicts.append(f)
+
+        metadata[date] = {
+            "c_0930": c_ref,
+            "h_0930": float(pivot_high.loc[date, time_0930]) if time_0930 in pivot_high.columns else c_ref,
+            "l_0930": float(pivot_low.loc[date, time_0930])  if time_0930 in pivot_low.columns  else c_ref,
+            "v_0930": float(pivot_vol.loc[date, time_0930])  if time_0930 in pivot_vol.columns  else 0,
+            "c_1510": float(pivot_close.loc[date, time_1510]) if time_1510 in pivot_close.columns else c_ref,
+            "h_1510": float(pivot_high.loc[date, time_1510])  if time_1510 in pivot_high.columns else c_ref,
+            "l_1510": float(pivot_low.loc[date, time_1510])   if time_1510 in pivot_low.columns  else c_ref,
+            "dip_low":   float(morning_low_per_date.get(date, c_ref)),
+            "peak_high": float(morning_high_per_date.get(date, c_ref)),
+        }
+
+    X = pd.DataFrame(feature_dicts, index=valid_dates).fillna(0)
+
+    if time_1510 in pivot_close.columns:
+        target = (pivot_close[time_1510] > pivot_close[time_0930]).astype(int)
+    else:
+        target = pd.Series(0, index=valid_dates)
+
+    return X, target, metadata

core/groww.py

@@ -0,0 +1,112 @@
+"""
+Groww charting API client.
+
+Fetches 1-minute OHLCV candle data for NSE CASH segment tickers.
+Handles:
+  - Retry with exponential backoff (3 attempts)
+  - None / missing values in candle arrays
+  - Cumulative volume -> per-candle volume conversion
+"""
+
+import time
+import logging
+import traceback
+from datetime import datetime
+
+import numpy as np
+import pandas as pd
+import requests
+
+logger = logging.getLogger("live_trader")
+
+GROWW_HEADERS = {
+    "User-Agent": (
+        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+        "(KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+    ),
+    "Accept": "application/json",
+}
+
+
+def fetch_groww_candles(ticker, days=5, max_retries=3):
+    """
+    Fetch 1-min OHLCV from Groww for the last *days* calendar days.
+
+    Returns a DataFrame [open, high, low, close, volume] with DateTimeIndex,
+    or None on complete failure.
+    """
+    end_ts   = int(time.time() * 1000)
+    start_ts = end_ts - (days * 24 * 3600 * 1000)
+
+    url = (
+        f"https://groww.in/v1/api/charting_service/v2/chart/exchange/NSE"
+        f"/segment/CASH/{ticker}"
+        f"?endTimeInMillis={end_ts}"
+        f"&intervalInMinutes=1"
+        f"&startTimeInMillis={start_ts}"
+    )
+
+    for attempt in range(1, max_retries + 1):
+        try:
+            resp = requests.get(url, headers=GROWW_HEADERS, timeout=15)
+            resp.raise_for_status()
+            data = resp.json()
+
+            if "candles" not in data or not data["candles"]:
+                logger.warning(f"[{ticker}] No candles in API response (attempt {attempt})")
+                if attempt < max_retries:
+                    time.sleep(2 * attempt)
+                    continue
+                return None
+
+            rows = []
+            for c in data["candles"]:
+                # Skip candles with None / missing OHLCV values
+                if c[1] is None or c[2] is None or c[3] is None or c[4] is None or c[5] is None:
+                    continue
+                try:
+                    dt = datetime.fromtimestamp(c[0])
+                    rows.append({
+                        "date":    dt,
+                        "open":    float(c[1]),
+                        "high":    float(c[2]),
+                        "low":     float(c[3]),
+                        "close":   float(c[4]),
+                        "cum_vol": float(c[5]),
+                    })
+                except (TypeError, ValueError):
+                    continue
+
+            if not rows:
+                logger.warning(f"[{ticker}] All candles had None values (attempt {attempt})")
+                if attempt < max_retries:
+                    time.sleep(2 * attempt)
+                    continue
+                return None
+
+            df = pd.DataFrame(rows)
+            df.set_index("date", inplace=True)
+            df.sort_index(inplace=True)
+
+            # Groww volume is cumulative per day -> difference it
+            df["date_only"] = df.index.date
+            df["volume"]    = df.groupby("date_only")["cum_vol"].diff().fillna(df["cum_vol"])
+            df["volume"]    = np.where(df["volume"] < 0, df["cum_vol"], df["volume"])
+            df.drop(columns=["cum_vol", "date_only"], inplace=True)
+
+            logger.info(f"[{ticker}] Fetched {len(df)} candles "
+                        f"({df.index.min()} -> {df.index.max()})")
+            return df
+
+        except requests.exceptions.RequestException as e:
+            logger.error(f"[{ticker}] API error attempt {attempt}/{max_retries}: {e}")
+            if attempt < max_retries:
+                time.sleep(3 * attempt)
+
+        except Exception as e:
+            logger.error(f"[{ticker}] Unexpected error attempt {attempt}/{max_retries}: {e}")
+            logger.debug(traceback.format_exc())
+            if attempt < max_retries:
+                time.sleep(3 * attempt)
+
+    return None

core/models.py

@@ -0,0 +1,117 @@
+"""
+Model construction and training.
+
+build_pipeline_map()  -> dict of {ticker: (feature_type, fresh_model_clone)}
+train_models()        -> dict of {ticker: (feature_type, fitted_model)}
+"""
+
+import pandas as pd
+from sklearn.ensemble import (
+    RandomForestClassifier,
+    HistGradientBoostingClassifier,
+    VotingClassifier,
+)
+from sklearn.linear_model import LogisticRegression
+from sklearn.preprocessing import StandardScaler
+from sklearn.pipeline import Pipeline
+from sklearn.feature_selection import SelectKBest, f_classif
+from sklearn.base import clone
+
+from core.config import TICKERS, DATA_DIR, PIPELINE_MAP
+from core.features import extract_semantic_features, extract_sequential_features
+
+
+# ── Base model templates ─────────────────────────────────────────────────────
+
+_RF  = RandomForestClassifier(
+    random_state=42, n_estimators=300, max_depth=8,
+    min_samples_leaf=5, n_jobs=-1,
+)
+_GBM = HistGradientBoostingClassifier(
+    random_state=42, max_iter=300, l2_regularization=1.0, max_depth=8,
+)
+_ENSEMBLE = VotingClassifier(
+    estimators=[("rf", _RF), ("gbm", _GBM)], voting="soft",
+)
+_LR_PIPELINE = Pipeline([
+    ("scaler", StandardScaler()),
+    ("lr", LogisticRegression(C=0.1, max_iter=1000)),
+])
+_KBEST15_LR = Pipeline([
+    ("scaler", StandardScaler()),
+    ("kbest", SelectKBest(f_classif, k=15)),
+    ("lr", LogisticRegression(C=0.1, max_iter=1000)),
+])
+
+_MODEL_TEMPLATES = {
+    "ensemble":    _ENSEMBLE,
+    "lr_pipeline": _LR_PIPELINE,
+    "kbest15_lr":  _KBEST15_LR,
+}
+
+
+# ── Public API ───────────────────────────────────────────────────────────────
+
+def build_pipeline_map():
+    """
+    Return a dict of {ticker: (feature_type, fresh_model_clone)}.
+    Uses PIPELINE_MAP from config to look up the architecture per ticker.
+    """
+    result = {}
+    for ticker in TICKERS:
+        feat_type, model_key = PIPELINE_MAP[ticker]
+        result[ticker] = (feat_type, clone(_MODEL_TEMPLATES[model_key]))
+    return result
+
+
+def train_models(log_fn=None):
+    """
+    Load parquet data, extract features, and train all ticker models.
+
+    Parameters
+    ----------
+    log_fn : callable(str), optional
+        Logging function (e.g. logger.info).  Falls back to print.
+
+    Returns
+    -------
+    dict  { ticker: (feature_type, fitted_model) }
+    """
+    if log_fn is None:
+        log_fn = print
+
+    pipeline_map = build_pipeline_map()
+    models = {}
+
+    for ticker in TICKERS:
+        fpath = DATA_DIR / f"{ticker}_minute.parquet"
+        if not fpath.exists():
+            log_fn(f"[{ticker}] Parquet file not found: {fpath}")
+            continue
+
+        df = pd.read_parquet(fpath)
+        df["date"] = pd.to_datetime(df["date"])
+        df.set_index("date", inplace=True)
+        df.sort_index(inplace=True)
+
+        # Keep at most 300 trading days
+        unique_days = df.index.normalize().unique()
+        if len(unique_days) > 300:
+            df = df[df.index.normalize().isin(unique_days[-300:])]
+
+        feat_type, clf = pipeline_map[ticker]
+
+        if feat_type == "semantic":
+            X, y, _ = extract_semantic_features(df)
+        else:
+            X, y, _ = extract_sequential_features(df)
+
+        if X is None or X.empty:
+            log_fn(f"[{ticker}] Feature extraction returned empty!")
+            continue
+
+        clf.fit(X, y)
+        models[ticker] = (feat_type, clf)
+        log_fn(f"[{ticker}] Model trained  ({feat_type})  |  {len(X)} samples")
+
+    return models

core/taxes.py

@@ -0,0 +1,50 @@
+"""
+Tax, brokerage, and slippage calculator for Indian equity intraday trades.
+Covers: brokerage (flat Rs 20), STT, exchange txn charge, GST, SEBI fee, stamp duty.
+"""
+
+
+def calculate_taxes_and_slippage(
+    price_0930, price_1510, qty,
+    high_0930, low_0930, high_1510, low_1510,
+    is_short,
+):
+    """
+    Compute net PnL after realistic slippage and all Indian regulatory charges.
+
+    Slippage model:  10% of the 1-min candle's (high - low) range,
+    applied as an adverse fill on both entry and exit.
+
+    Returns
+    -------
+    (net_pnl, total_taxes, exec_buy_price, exec_sell_price)
+    """
+    slip_0930 = (high_0930 - low_0930) * 0.10
+    slip_1510 = (high_1510 - low_1510) * 0.10
+
+    if not is_short:
+        # LONG: Buy at 09:30 (ask penalty), Sell at 15:10 (bid penalty)
+        actual_buy_price  = price_0930 + slip_0930
+        actual_sell_price = price_1510 - slip_1510
+    else:
+        # SHORT: Sell at 09:30 (bid penalty), Buy-to-cover at 15:10 (ask penalty)
+        actual_sell_price = price_0930 - slip_0930
+        actual_buy_price  = price_1510 + slip_1510
+
+    buy_turnover  = actual_buy_price * qty
+    sell_turnover = actual_sell_price * qty
+    total_turnover = buy_turnover + sell_turnover
+
+    brokerage      = 20.0
+    stt            = sell_turnover * 0.00025
+    exc_txn_charge = total_turnover * 0.0000325
+    gst            = (brokerage + exc_txn_charge) * 0.18
+    sebi_fee       = total_turnover * 0.000001
+    stamp_duty     = buy_turnover * 0.00003
+
+    total_taxes = brokerage + stt + exc_txn_charge + gst + sebi_fee + stamp_duty
+
+    gross_pnl = (actual_sell_price - actual_buy_price) * qty
+    net_pnl   = gross_pnl - total_taxes
+
+    return net_pnl, total_taxes, actual_buy_price, actual_sell_price

data/live_trades.json

@@ -0,0 +1,45 @@
+{
+  "starting_capital": 3692.0,
+  "leverage": 5.0,
+  "last_updated": "2026-06-19T17:26:58.267583+05:30",
+  "trades": [
+    {
+      "date": "2026-06-19",
+      "ticker": "POWERGRID",
+      "direction": "LONG",
+      "confidence": 0.7291,
+      "entry_price": 289.6,
+      "shares": 63,
+      "capital_before": 3692.0,
+      "buying_power": 18460.0,
+      "liquidity_capped": false,
+      "candle_volume": 7916,
+      "signal_time": "2026-06-19T17:26:58.267583+05:30",
+      "net_pnl": null,
+      "exit_price": null,
+      "status": "OPEN",
+      "all_predictions": {
+        "INFY": {
+          "prob_up": 0.5328,
+          "prob_down": 0.4672
+        },
+        "ASIANPAINT": {
+          "prob_up": 0.4417,
+          "prob_down": 0.5583
+        },
+        "TECHM": {
+          "prob_up": 0.529,
+          "prob_down": 0.471
+        },
+        "POWERGRID": {
+          "prob_up": 0.7291,
+          "prob_down": 0.2709
+        },
+        "ONGC": {
+          "prob_up": 0.4045,
+          "prob_down": 0.5955
+        }
+      }
+    }
+  ]
+}

data/minute_ohlcv/ADANIENT_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eebc57112ba3b638ac0eb0bfa0afb8ba20fc761f6f45aa91c488e383912a0810
+size 14773599

data/minute_ohlcv/ADANIPORTS_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b67f6879f390678a08e49699c291d9d5f571919efbac93d698e36a33fb9cc18a
+size 15238934

data/minute_ohlcv/APOLLOHOSP_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:383b2cae56de9e5148107517334adbbd2a75135e6338e1253722054945aee6db
+size 16979850

data/minute_ohlcv/ASIANPAINT_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091db8991ee9823b39280e0f498de5bee5da5fb6f870766c4f669538b43dc2a3
+size 15802023

data/minute_ohlcv/AXISBANK_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15f276861577a02701e486b4f3e67203158a45d2187a39f374656ba17b911388
+size 15441163

data/minute_ohlcv/BAJAJ-AUTO_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:601e67d2b7a51e965abd48b74d4eb358492f5e5a78d4667039658af70e5a8220
+size 14106851

data/minute_ohlcv/BAJAJFINSV_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03a29d080333706109ebfc5cbe4c443b768000a52e70357aaa74439e54fb7b52
+size 15227892

data/minute_ohlcv/BAJFINANCE_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a396f007b0d7cea66ad3c70e1c82056d51b4ce9551ebe3055003833eaa1c837c
+size 10715033

data/minute_ohlcv/BHARTIARTL_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7d35c8f229c9dc1d7962c45161a57ef748e4aa3546652c1d33f2c92be9d90fc
+size 14982789

data/minute_ohlcv/BPCL_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cacbf2ddd085066cf589e1e06afa5396830d7372e08a4a0531a884246317df9c
+size 13187593

data/minute_ohlcv/BRITANNIA_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e5d0892ed6492f785ad3fd45b46c5588f7e621894cd99dfd7440c4fe26c7359
+size 17360107

data/minute_ohlcv/CIPLA_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f27a7153429f341b26463c1b87d01360f7d5d39e8f829dc485bbe450d64c2910
+size 14792094

data/minute_ohlcv/COALINDIA_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bde38dda44b53363d369c08c6e0e786dc6f1c86ef9f9eca5305b6c20689aebba
+size 14020728

data/minute_ohlcv/DIVISLAB_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dc67e4779c2518d6b785d512c3c3e712fc68983c4e94735e651d9e5ba467788
+size 16927541

data/minute_ohlcv/DRREDDY_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:829acfff4c1ff2e33b080182da331bd94ff5b926bc7cf332ca6cc18f03bdf36e
+size 14544884

data/minute_ohlcv/EICHERMOT_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14ec70cd80a92c4951630a4333ca552da7098a6b8ba7e695cbdd6b6a058eccc4
+size 16372533

data/minute_ohlcv/GRASIM_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4968f3b0d7aa3540178fe1cfa3b730dd0f29f517dc8ccd043fb618e637633600
+size 15487201

data/minute_ohlcv/HCLTECH_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ffb95b06a486264d6aa3c194f74848bd337e4ed1db5a3ca13fa4c3697896cb6
+size 14860521

data/minute_ohlcv/HDFCBANK_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23a13cd204544eacde2f66350dc0c757e5bba16c2614863e3f5a600c5621d1ca
+size 14126028

data/minute_ohlcv/HDFCLIFE_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e1cce52af098c3a52ab43dfdf0ffc6a35d51d2a8582e2b76438ce2128dfe55b
+size 10940288

data/minute_ohlcv/HEROMOTOCO_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e85aba15ae24c84ecab70d7c95d951137321ae46653ea59e2aebd7fb552f7ae
+size 16496977

data/minute_ohlcv/HINDALCO_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2ef7911d15de19463ab781a8d99ef8bb8cffca5ea84966a74bfa4f6cd0e5217
+size 14811915

data/minute_ohlcv/HINDUNILVR_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d23ce684444fa3fb77f0728322f095333046cfeb41ce04d24fe3e11004bfb04a
+size 15418103

data/minute_ohlcv/ICICIBANK_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76586d7192f781323d5f2247be3138b8f68b6836f59b1e250d8c54ad4fb772cc
+size 16701861

data/minute_ohlcv/INDUSINDBK_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1b276308d798525dc310eb1c0761b72492feecf199ff1cc0ee2c655efe6a70f
+size 15676926

data/minute_ohlcv/INFY_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3ff8af67ff82f0fb8d0feebee6e9ad69198112623e0017bb4f526e5761fbc3f
+size 15787029

data/minute_ohlcv/ITC_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4549e21d5b011b39303c1deaebc41d2095f6a54b789de59d3019cb7dafe18c81
+size 14482090

data/minute_ohlcv/JSWSTEEL_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abb7ac6787da7e99e28acb1d0e3ac73987cc4ddd2b62f34381354f23f0411cff
+size 15132654

data/minute_ohlcv/KOTAKBANK_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e8714261774b11205a22bfa52c5eea47a15dbf43989e8ddaa172356e96e4187
+size 12401933

data/minute_ohlcv/LTIM_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c58a67f3d9e2aff5aff191fed63f05f405e01a85290b9ba6371d13eda18c4bf3
+size 14677815

data/minute_ohlcv/LT_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1479bc457abdcd5ae385f15ea0139b88a11e89026f6d89ff3bb467049df6dca
+size 16306243

data/minute_ohlcv/MARUTI_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72d247e8eb660cd8807ca5bf123d5cb29405811d973eda9595981032c15bf9d5
+size 18866904

data/minute_ohlcv/MM_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cbfa7c3ba868bbff6ef6b7143f9f3c2573c8e25b742cf899121095be9829c8a
+size 15397098

data/minute_ohlcv/NESTLEIND_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7fa10b93a3d177f38784b59744ef0c610dc4de681f0818424fecb7a73722a66
+size 13168657

data/minute_ohlcv/NTPC_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:269f3bcf9e9cecfc7455b72d082639d57cad7d2b9c5c02c8a36f5c03aa53b85f
+size 13715104

data/minute_ohlcv/ONGC_minute.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e55660ed66362851e0350f4bd3b217eabd5fc31d092a2bd22515c077b3cd7f99
+size 13962049