crowdedness_interpolate_trip

crowdedness_interpolate_trip.py

def get_trip_data(number, date):
  trip = trips[(trips.trip_short_name==number) & (trips.date==date)].iloc[0]
  tripid = trip['trip_id']
  shapeid = trip['shape_id']
  tripshape = shapes[shapes.shape_id == shapeid]
  tripstops = stoptimes[stoptimes.trip_id==tripid].sort_values('stop_sequence')

  lastdep = -1
  lastdist= -1
  seq = 0
  dl = []

  for position in tripstops.iterrows():
    arr = position[1][3]
    dep = position[1][4]
    dist = position[1][5]
    dist = dist + 0.1 if dist == lastdist else dist
    arr = arr + 1 if arr == lastdep else arr
    stop = position[1][2]
    if (dist == dist) & (arr == arr):
      if lastdep > 0:
        seq = seq+1
        stepsize = ((dist-lastdist) / (arr-lastdep))
        for t in range(lastdep+1, arr):
          dst = int(lastdist + ( (t - lastdep) * stepsize))
          lat, lon = interpolate_lat_lon(tripshape, dst)
          dl.append({'trip_id': tripid, 'time': int(t), 'lat': lat, 'lon': lon, 
                     'stop_id': stop, 'ritnumber' : number, 'sequence' : seq})
      seq = seq+1
      for t in range(arr, dep+1):
        dst = int(dist)
        lat, lon = interpolate_lat_lon(tripshape, dst)
        dl.append({'trip_id': tripid, 'time': int(t), 'lat': lat, 'lon': lon, 
                   'stop_id': stop, 'ritnumber' : number, 'sequence' : seq})
      lastdep = dep
      lastdist = dist
  df = pd.DataFrame(dl)
  df.trip_id = df.trip_id.astype(int)
  df.sequence = df.sequence.astype(int)
  df.time = df.time.astype(int)
  df.ritnumber = df.ritnumber.astype(int)
  df.stop_id = df.stop_id.astype(int).astype(str)
  df = df[['trip_id', 'ritnumber', 'sequence', 'time', 'stop_id', 'lat', 'lon']]
  return df