Skip to content

Latest commit

 

History

History
325 lines (239 loc) · 5.03 KB

TideData.md

File metadata and controls

325 lines (239 loc) · 5.03 KB


Data from https://tidesandcurrents.noaa.gov/waterlevels.html?id=9414290&units=standard&bdate=20190701&edate=20190801&timezone=GMT&datum=MLLW&interval=6&action=data


import math
import datetime
import pytz
import glob
import functools
import operator
import numpy
import pandas
import matplotlib.pyplot
import matplotlib.pylab
import seaborn
import sklearn.linear_model
import sklearn.metrics
import vtreat.cross_plan

from pandas.plotting import register_matplotlib_converters
register_matplotlib_converters()


date_fmt = '%Y/%m/%d %H:%M'
tz = pytz.utc

def parse_date(dtstr):
    d0 = datetime.datetime.strptime(dtstr, date_fmt)
    return d0.replace(tzinfo=tz)
    
base_date_time = datetime.datetime(2001, 1, 1, tzinfo=tz)


print("TZ NAME: {tz}".format(tz=base_date_time.tzname()))


TZ NAME: UTC



print(base_date_time)


2001-01-01 00:00:00+00:00



na_values = [ '', '-' ]
files = [f for f in glob.glob("tide_data/*.csv", recursive=False)]
files.sort()
tides = [pandas.read_csv(f, na_values=na_values) for f in files]
tides = pandas.concat(tides, axis=0)
tides.reset_index(inplace=True, drop=True)
tides.head()



<style scoped>
    .dataframe tbody tr th:only-of-type {
        vertical-align: middle;
    }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}


</style>


  
    

      
      Date
      Time (GMT)
      Predicted (ft)
      Preliminary (ft)
      Verified (ft)
    

  
  
    

      0
      2017/01/01
      00:00
      1.849
      NaN
      2.12
    

    

      1
      2017/01/01
      00:06
      1.695
      NaN
      1.97
    

    

      2
      2017/01/01
      00:12
      1.543
      NaN
      1.88
    

    

      3
      2017/01/01
      00:18
      1.393
      NaN
      1.78
    

    

      4
      2017/01/01
      00:24
      1.247
      NaN
      1.66
    

  





d0 = parse_date('2001/01/01 00:00')
(d0 - base_date_time).total_seconds()


0.0



print("TZ NAME: {tz}".format(tz=d0.tzname()))


TZ NAME: UTC



tides['dt'] = [parse_date(tides['Date'][i] + ' ' + tides['Time (GMT)'][i]) for i in range(tides.shape[0])]
tides['dts'] = [(t - base_date_time).total_seconds() for t in tides['dt']]


tides['tide feet'] = tides['Verified (ft)'].copy()
null_posns = pandas.isnull(tides['tide feet'])
tides.loc[null_posns, 'tide feet'] = tides.loc[null_posns, 'Preliminary (ft)']


tides.head()



<style scoped>
    .dataframe tbody tr th:only-of-type {
        vertical-align: middle;
    }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}


</style>


  
    

      
      Date
      Time (GMT)
      Predicted (ft)
      Preliminary (ft)
      Verified (ft)
      dt
      dts
      tide feet
    

  
  
    

      0
      2017/01/01
      00:00
      1.849
      NaN
      2.12
      2017-01-01 00:00:00+00:00
      504921600.0
      2.12
    

    

      1
      2017/01/01
      00:06
      1.695
      NaN
      1.97
      2017-01-01 00:06:00+00:00
      504921960.0
      1.97
    

    

      2
      2017/01/01
      00:12
      1.543
      NaN
      1.88
      2017-01-01 00:12:00+00:00
      504922320.0
      1.88
    

    

      3
      2017/01/01
      00:18
      1.393
      NaN
      1.78
      2017-01-01 00:18:00+00:00
      504922680.0
      1.78
    

    

      4
      2017/01/01
      00:24
      1.247
      NaN
      1.66
      2017-01-01 00:24:00+00:00
      504923040.0
      1.66
    

  





numpy.mean(tides['tide feet'])


3.354944709837226



deltas = [tides['dts'][i+1] - tides['dts'][i] for i in range(tides.shape[0]-1)]


max(deltas)


360.0



min(deltas)


360.0



tides.to_pickle('tides.pickle.gz')