hkjc 0.3.23__tar.gz → 0.4.0__tar.gz

{hkjc-0.3.23 → hkjc-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hkjc
-Version: 0.3.23
+Version: 0.4.0
 Summary: Library for scrapping HKJC data and perform basic analysis
 Requires-Python: >=3.11
 Requires-Dist: beautifulsoup4>=4.14.2
@@ -12,4 +12,5 @@ Requires-Dist: numpy>=2.3.3
 Requires-Dist: polars>=1.33.1
 Requires-Dist: pyarrow>=21.0.0
 Requires-Dist: requests>=2.32.5
+Requires-Dist: scipy>=1.16.2
 Requires-Dist: tqdm>=4.67.1

hkjc-0.4.0/dashboard/app.py

@@ -0,0 +1,225 @@
+from flask import Flask, jsonify, render_template, request, make_response
+from flask_caching import Cache
+
+import polars as pl
+import numpy as np
+
+from hkjc.live import live_odds, _fetch_live_races
+from hkjc.harville_model import fit_harville_to_odds
+from hkjc.historical import get_horse_data
+from hkjc.speedpro import speedmap, speedpro_energy
+from hkjc.strategy import qpbanker, place_only
+from hkjc import generate_all_qp_trades, generate_all_pla_trades, pareto_filter
+
+
+def arr_to_dict(arr: np.ndarray, dtype=float):
+    """Convert 0-indexed numpy array into 1-indexed nested dictionary
+
+    Args:
+        arr (np.ndarray): 0-indexed numpy array
+        dtype (type, optional): data type. Defaults to float.
+
+    Returns:
+        dict: 1-indexed nested dictionary
+    """
+    if arr.ndim == 1:
+        return {i+1: dtype(np.round(v, 1)) for i, v in enumerate(arr) if not (np.isnan(v) or np.isinf(v))}
+
+    return {i+1: arr_to_dict(v) for i, v in enumerate(arr)}
+
+
+app = Flask(__name__)
+config = {
+    "CACHE_TYPE": "RedisCache",
+    "CACHE_REDIS_HOST": "localhost",
+    "CACHE_REDIS_PORT": "6379"
+}
+app.config.from_mapping(config)
+cache = Cache(app)
+
+
+@app.route('/')
+def disp_race_info():
+    race_info = _fetch_live_races('', '')
+
+    try:
+        df_speedpro = speedpro_energy(race_info['Date'])
+        for race_num, race in race_info['Races'].items():
+            for i, runner in enumerate(race['Runners']):
+                df = (df_speedpro
+                      .filter(pl.col('RaceNo') == race_num)
+                      .filter(pl.col('RunnerNumber') == int(runner['No']))
+                      )
+                race_info['Races'][race_num]['Runners'][i]['SPEnergy'] = df['SpeedPRO_Energy_Difference'].item(
+                    0)
+                race_info['Races'][race_num]['Runners'][i]['Fitness'] = df['FitnessRatings'].item(
+                    0)
+    except:  # fill with dummy value if SpeedPro not available
+        for race_num, race in race_info['Races'].items():
+            for i, runner in enumerate(race['Runners']):
+                race_info['Races'][race_num]['Runners'][i]['SPEnergy'] = 0
+                race_info['Races'][race_num]['Runners'][i]['Fitness'] = 0
+
+    return render_template('index.html',
+                           race_info=race_info)
+
+
+turf_going_dict = {'FIRM': 'F',
+                   'GOOD TO FIRM': 'GF',
+                   'GOOD': 'G',
+                   'GOOD TO YIELDING': 'GY',
+                   'YIELDING': 'Y',
+                   'YIELDING TO SOFT': 'YS',
+                   'SOFT': 'S',
+                   'HEAVY': 'H'}
+aw_going_dict = {'WET FAST': 'WF',
+                 'FAST': 'FT',
+                 'GOOD': 'GD',
+                 'SLOW': 'SL',
+                 'WET SLOW': 'WS',
+                 'RAIN AFFECTED': 'RA',
+                 'NORMAL WATERING': 'NW'}
+going_dict = {'TURF': turf_going_dict, 'ALL WEATHER TRACK': aw_going_dict}
+
+
+@app.route('/horse_info/<horse_no>', methods=['GET'])
+# cache horse history for 1 day
+@cache.cached(timeout=86400, query_string=True)
+def disp_horse_info(horse_no):
+    # read optional filters
+    dist = request.args.get('dist', type=int)
+    track = request.args.get('track')
+    going = request.args.get('going')
+    venue = request.args.get('venue')
+
+    if track not in going_dict.keys():
+        track = None
+    if venue not in ['HV', 'ST']:
+        venue = None
+    if (going is not None) and (track is not None) and (going in going_dict[track].keys()):
+        going = going_dict[track][going]  # translate going to code
+    else:
+        going = None
+
+    df = get_horse_data(horse_no)
+
+    if df.height > 0:
+        if dist is not None:
+            df = df.filter(pl.col('Dist') == dist)
+        if track and track.upper() == 'TURF':
+            df = df.filter(pl.col('Track') == 'Turf')
+        elif track and track.upper() == 'ALL WEATHER TRACK':
+            df = df.filter(pl.col('Track') == 'AWT')
+        if going is not None:
+            df = df.filter(pl.col('G').str.starts_with(going[0]))
+        if venue is not None:
+            df = df.filter(pl.col('Venue') == venue)
+
+    return render_template('horse-info.html', df=df.head(5))
+
+
+@app.route('/live_odds/<int:race_no>')
+def disp_live_odds(race_no=1):
+    odds_dict = live_odds('', '', race_no)
+    fitted_odds = fit_harville_to_odds(odds_dict)['odds_fit']
+    odds_json = {'Raw': {k: arr_to_dict(v) for k, v in odds_dict.items()},
+                 'Fit': {k: arr_to_dict(v) for k, v in fitted_odds.items()}
+                 }
+
+    return jsonify(odds_json)
+
+
+@app.route('/speedmap/<int:race_no>')
+def disp_speedmap(race_no=1):
+    return speedmap(race_no)
+
+
+@app.route('/qp/<int:race_no>/<int:banker>/<cover>', methods=['GET'])
+def disp_qp_metrics(race_no=1, banker=1, cover='2'):
+    use_filter = request.args.get('filter')
+    use_filter = use_filter and (use_filter.lower() == 'true')
+
+    odds = live_odds('', '', race_number=race_no)
+
+    if use_filter:
+        res = fit_harville_to_odds(odds)
+        if res['success']:
+            odds = res['odds_fit']
+    covered = [int(v) for v in cover.split(',')]
+    ev = qpbanker.expected_value(odds['PLA'], odds['QPL'], banker, covered)
+    win = qpbanker.win_probability(odds['PLA'], banker, covered)
+    avg_odds = qpbanker.average_odds(odds['QPL'], banker, covered)
+
+    return {'Banker': banker, 'Covered': covered, 'ExpValue': round(ev, 2), 'WinProb': round(win, 2), 'AvgOdds': round(avg_odds, 2)}
+
+@app.route('/pla/<int:race_no>/<cover>', methods=['GET'])
+def disp_pla_metrics(race_no=1, cover=[]):
+    odds = live_odds('', '', race_number=race_no)
+
+    res = fit_harville_to_odds(odds)
+    if not res['success']:
+        raise RuntimeError(
+            f"[ERROR] Harville model fitting failed: {res.get('message','')}")
+    
+    covered = [int(v) for v in cover.split(',')]
+    ev = place_only.expected_value(odds['PLA'], res['prob_fit']['P'], covered)
+    win = place_only.win_probability(res['prob_fit']['P'], covered)
+    avg_odds = place_only.average_odds(odds['PLA'], covered)
+
+    return {'Covered': covered, 'ExpValue': round(ev, 2), 'WinProb': round(win, 2), 'AvgOdds': round(avg_odds, 2)}
+
+def elimination(lst):
+    cond = [~pl.col('Covered').list.contains(l) for l in lst]
+    return pl.all_horizontal(cond)
+
+
+def format_trade(trade):
+    return {'Banker': trade.get('Banker',None),
+            'Covered': trade['Covered'],
+            'WinProb': round(trade['WinProb'], 2),
+            'ExpValue': round(trade['ExpValue'], 2),
+            'AvgOdds': round(trade['AvgOdds'], 2)}
+
+
+@app.route('/qprec/<int:race_no>/<int:banker>', methods=['GET'])
+def disp_qp_recs(race_no=1, banker=1, exclude=[], maxC=None):
+    use_filter = request.args.get('filter')
+    use_filter = use_filter and (use_filter.lower() == 'true')
+    exclude = request.args.get('exclude')
+    if exclude:
+        excluded = [int(v) for v in exclude.split(',')]
+    maxC = request.args.get('maxC', type=int)
+
+    df_trades = generate_all_qp_trades(
+        '', '', race_no, fit_harville=use_filter)
+    df_trades = df_trades.filter(pl.col('Banker') == banker)
+    df_trades = df_trades.filter(pl.col('ExpValue') >= 0.05)
+    if exclude:
+        df_trades = df_trades.filter(elimination(excluded))
+    if maxC:
+        df_trades = df_trades.filter(pl.col('NumCovered') <= maxC)
+    pareto_trades = pareto_filter(df_trades, [], ['WinProb', 'ExpValue']).sort(
+        'WinProb', descending=True)
+
+    return [format_trade(t) for t in pareto_trades.iter_rows(named=True)]
+
+
+
+
+@app.route('/plarec/<int:race_no>', methods=['GET'])
+def disp_pla_recs(race_no=1, exclude=[], maxC=None):
+    exclude = request.args.get('exclude')
+    if exclude:
+        excluded = [int(v) for v in exclude.split(',')]
+    maxC = request.args.get('maxC', type=int)
+
+    df_trades = generate_all_pla_trades('', '', race_no)
+    df_trades = df_trades.filter(pl.col('ExpValue') >= 0.05)
+    if exclude:
+        df_trades = df_trades.filter(elimination(excluded))
+    if maxC:
+        df_trades = df_trades.filter(pl.col('NumCovered') <= maxC)
+    pareto_trades = pareto_filter(df_trades, [], ['WinProb', 'ExpValue']).sort(
+        'WinProb', descending=True)
+
+    return [format_trade(t) for t in pareto_trades.iter_rows(named=True)]