timeseries class

class csi.timeseries(name, utmzone=None, verbose=True, lon0=None, lat0=None, ellps='WGS84')

A class that handles generic time series

Args:

  • name : Name of the dataset.

Kwargs:

  • utmzone : UTM zone (optional, default=None)

  • lon0 : Longitude of the center of the UTM zone

  • lat0 : Latitude of the center of the UTM zone

  • ellps : ellipsoid (optional, default=’WGS84’)

  • verbose : Talk to me

SortInTime()

Sort ascending in time.

Returns:

  • None

addPointInTime(time, value=0.0, std=0.0)

Augments the time series by one point.

Args:

  • time : datetime.datetime object

Kwargs:

  • value : Value of the time series at time {time}

  • std : Uncertainty at time {time}

adddata(time, values=None, std=None)

Augments the time series

Args:

  • time : list of datetime objects

Kwargs:

  • values : list array or None

  • std : list array or None

Returns:

  • None

checkNaNs()

Returns the index of NaNs

Returns:

  • numpy array of integers

computeDoubleDifference()

Compute the derivative of the TS with a central difference scheme.

Returns:

  • None. Results is stored in self.derivative

findZeroIntersect(data='data')

Returns all the points just before the function crosses 0.

Kwargs:

  • data : Can be ‘data’, ‘synth’ or ‘derivative’.

Returns:

  • None

fitFunction(function, m0, solver='L-BFGS-B', iteration=1000, tol=1e-08)

Fits a function to the timeseries

Args:

  • function : Prediction function,

  • m0 : Initial model

Kwargs:

  • solver : Solver type (see list of solver in scipy.optimize.minimize)

  • iteration : Number of iteration for the solver

  • tol : Tolerance

Returns:

  • None. Model vector is stored in the {m} attribute

fitTidalConstituents(steps=None, linear=False, tZero=datetime.datetime(2000, 1, 1, 0, 0), chunks=None, cossin=False, constituents='all')

Fits tidal constituents on the time series.

Kwargs:

  • steps : list of datetime instances to add step functions in the estimation process.

  • linear : estimate a linear trend.

  • tZero : origin time (datetime instance).

  • chunks : List [ [start1, end1], [start2, end2]] where the fit is performed.

  • cossin : Add a cosine+sine term in the procedure.

  • constituents : list of tidal constituents to include (default is all). For a list, go check tidalfit class

Returns:

  • None

getOffset(date1, date2, nodate=nan, data='data')

Get the offset between date1 and date2.

Args:

  • date1 : datetime object

  • date2 : datetime object

Kwargs:

  • nodate : Value to be returned in case no value is available

  • data : can be ‘data’ or ‘std’

Returns:

  • float

initialize(time=None, start=None, end=None, increment=1)

Initialize the time series.

Kwargs:

  • time : list of datetime instances

  • start : datetime instance of the first period

  • end : datetime instance of the ending period

  • increment : increment of time between periods

Returns:

  • None

plot(figure=1, styles=['.r'], show=True, data='data', subplot=None)

Plots the time series.

Args:

  • figure : Figure id number (default=1)

  • styles : List of styles (default=[‘.r’])

  • show : Show to me (default=True)

  • data : can be ‘data’, ‘derivative’, ‘synth’ or a list of those

  • subplot : axes instance to be used for plotting. If None, creates a new one

Returns:

  • None

readAscii(infile, header=0)

Reads from an ascii file. Format of the file is

year

month

day

hour

min

second

value

err (optional)
Args:

  • infile : Input file (ascii)

Kwargs:

  • header : length of the file header

Returns:

  • None

reference2timeseries(timeseries)

Removes to another gps timeseries the difference between self and timeseries

Args:

  • timeseries : Another timeseries

Returns:

  • float

removeMean(start=None, end=None)

Removes the mean between start and end.

Kwargs:

  • start : datetime.datetime object. If None, takes the first point of the time series

  • end : datetime.datetime object. If None, takes the last point of the time series

Returns:

  • None. Attribute {value} is directly modified.

removePoints(indexes)

Removes the points from the time series

Args:

  • indexes: Indexes of the poitns to remove

Returns:

  • None

smoothGlitches(biggerThan=999999.0, smallerThan=-999999.0, interpNum=5, interpolation='linear')

Removes the glitches and replace them by a value interpolated on interpNum points.

Kwargs:

  • biggerThan : Values higher than biggerThan are glitches.

  • smallerThan : Values smaller than smallerThan are glitches.

  • interpNum : Number of points to take before and after the glicth to predict its values.

  • interpolation : Interpolation method.

Returns:

  • None

trimTime(start, end=datetime.datetime(2100, 1, 1, 0, 0))

Keeps the data between start and end. start and end are 2 datetime.datetime objects.

Args:

  • start : datetime.datetime object

Kwargs:

  • end : datetime.datetime object

Returns:

  • None

write2file(outfile, steplike=False)

Writes the time series to a file.

Args:

  • outfile : output file.

Kwargs:

  • steplike : doubles the output each time so that the plot looks like steps.

Returns:

  • None