2.3.13. vacumm.misc.stats – Statistiques spéciales

Functions:

corr_proba() ensrank()

qtmax() qtmin()

qtminmax()

Classes:

StatAccum all_stats default_restart_file dual_stats single_stats append() dump() get_savail() get_sbias() get_scorr() get_scount() get_scrms() get_shist() get_smean() get_srms() get_sstats() get_sstd() get_stats() get_tavail() get_tbias() get_tcorr() get_tcount() get_tcrms() get_thist() get_tmean() get_trms() get_tstats() get_tstd() isempty() load() StatAccumError

Statistical tools

class StatAccum(tall=False, tavail=None, tmean=None, tstd=None, smax=None, smin=None, tbias=None, trms=None, tcrms=None, tcorr=None, shist=None, tcount=None, sall=False, savail=None, smean=None, sstd=None, tmax=None, tmin=None, sbias=None, srms=None, scrms=None, scorr=None, thist=None, scount=None, bins=None, restart_file=None, restart=False, withtime=None)

Bases: object

Statistics accumulator

Generic params:

• t/sall: Perform all the statistics by default.
• withtime, optional: Input does not contain a time dimension.

Single:

• tcount/scount: Number of observations taken into account
• tavail/savail: Percentage of available observations
• tmean/smean: Temporal (t) / Spatial (s) average
• tstd/sstd: Temporal (t) / Spatial (s) std

Dual:

• tall: Perform all the following statistics by default.
• tbias/sbias: Temporal (t) / Spatial (s) bias
• trms/srms: Temporal (t) / Spatial (s) RMS
• tcrms/scrms: Temporal (t) / Spatial (s) centered RMS
• tcorr/scorr: Temporal (t) / Spatial (s) Correlation

Example:

>>> sa = StatAccum(tbias=True, srms=True)
>>> sa += ssta1, sstb1
>>> sa += ssta2, sstb2
>>> tbias = sa.get_tbias()
>>> srms = sa.get_srms()

all_stats = ('mean', 'std', 'hist', 'min', 'max', 'bias', 'rms', 'crms', 'corr', 'avail', 'count')

append(*items)

Append data to the accumulator

default_restart_file = 'stataccum_restart.nc'

dual_stats = ('bias', 'rms', 'crms', 'corr', 'avail', 'count')

dump(restart_file=None)

Dump the current instance to a netcdf file

get_savail(**kwargs)

Get the spatial availability

get_sbias(**kwargs)

Get the spatial bias

get_scorr(**kwargs)

Get the spatial correlation

get_scount(**kwargs)

Get the spatial count

get_scrms(**kwargs)

Get the spatial centered RMS difference

get_shist(**kwargs)

Get the spatial histogram

get_smean(**kwargs)

Get the spatial standard deviation

get_srms(**kwargs)

Get the spatial absolute RMS difference

get_sstats(**kwargs)

Get all spatial statistics as a dictionary

get_sstd(**kwargs)

Get the spatial mean

get_stats()

Get all statistics as dict

get_tavail(**kwargs)

Get the temporal availability

get_tbias(**kwargs)

Get the temporal bias

get_tcorr(**kwargs)

Get the temporal correlation

get_tcount(**kwargs)

Get the temporal count

get_tcrms(**kwargs)

Get the temporal centered RMS difference

get_thist(**kwargs)

Get the temporal histogram

get_tmean(**kwargs)

Get the temporal standard deviation

get_trms(**kwargs)

Get the temporal absolute RMS difference

get_tstats(**kwargs)

Get all temporal statistics as a dictionary

get_tstd(**kwargs)

Get the temporal mean

isempty()

load(restart_file=None, iterindex=None, nowtime=None)

Load the current instance from a netcdf file

Params:

• restart_file, optional: Netcdf restart file.
• iterindex, optional: If given, the restart file is not loaded if iterindex is greater or equal to the file’s iterindex attribute.
• nowtime, optional: If given, the restart file is not loaded if nowtime is greater or equal to the file’s lasttime attribute.

single_stats = ('mean', 'std', 'hist', 'min', 'max')

exception StatAccumError

Bases: vacumm.VACUMMError

args

message

corr_proba(r, ndata, ndataset=2, dof=False)

Probability of rejecting correlations

• r: Correlation coefficient
• ndata: Number of records use for correlations
• ndataset, optional: Number of datasets (1 for autocorrelations, else 2) [default: 2]

Todo

This must be rewritten using scipy.stats

ensrank(obs, ens, gethist=False, getnrz=False, centered=False)

Compute the rank of a reference (observations) in an ensemble

If nens is the size of the ensemble, the rank may go from 0 to nens.

• obs, (…): Observation array
• ens, (nens,…): Ensemble array
• getrank, optional: If True, return also the rank histogram
• getnrz, optional: If True, return also non recovery zones
• centered, optional: Center the ensemble data on the observations before computing the rank

qtmax(var, qt=95.0)

Get max using quantiles

This is useful for very asymmetic distributions

Params:

• var: A numeric array.
• qt, optional: Percentile used to define max

qtmin(var, qt=5.0)

Get min max using quantiles

This is useful for very asymmetic distributions

Params:

• var: A numeric array.
• qt, optional: Percentile used to define min

qtminmax(var, qt=5.0)

Get min and max using quantiles

This is useful for very asymmetic distributions

Params:

• var: A numeric array.

• qt, optional: Percentile used to define min and max

  • Single number: qt -> [qt, 100-qt]
  • Two elements: used directly.