Variables
list	analysisStatistics = ['Count','Mean','RMS','STD']
	Select the statistics to be analyzed options: see su.allFileStats 'analysisStatistics' from individual diagnostics will override this setting. More...

dictionary	diagnosticGroupings = {}

dictionary	dbConf = {}
	Configure the StatisticsDatabase.StatsDB objects. More...

string	user = 'guerrett'
	hasFCLenDir whether directory structure includes forecast length note: overridden when fcTDeltaLast > fcTDeltaFirst More...

	hours
	cntrlExpIndex is the index of the control experiment (used for DiffCI analyses) in the expNames list More...

	days

string	commonFCDirFormat = "%hhr"
	fcDirFormats is used to declare the directory string format for forecast lengths. More...

string	commonStatsFileSubDir = 'diagnostic_stats/obs'
	statsFileSubDirs is the subdirectory within the date directory(ies) that contains the statstics files for constructing the StatsDB object examples: TODO: make StatsDB search for the correct subdirectory for each experiment 'diagnostic_stats' 'diagnostic_stats/obs' 'diagnostic_stats/model' More...

list	analysisTypes = []
	Configure the analysisTypes to apply to the statistics. More...

Variable Documentation

◆ analysisStatistics

list analyze_config.analysisStatistics = ['Count','Mean','RMS','STD']

Select the statistics to be analyzed options: see su.allFileStats 'analysisStatistics' from individual diagnostics will override this setting.

Definition at line 70 of file analyze_config.py.

◆ analysisTypes

list analyze_config.analysisTypes = []

Configure the analysisTypes to apply to the statistics.

below are recommendations for single/multiple forecast lengths
analysisTypes can be mixed and matched as desired, however some of them require nCY, nFC, or nExp > 1
see the individual classes for more details (Analyses.py)

Definition at line 201 of file analyze_config.py.

◆ commonFCDirFormat

string analyze_config.commonFCDirFormat = "%hhr"

fcDirFormats is used to declare the directory string format for forecast lengths.

Can include any combination of substrings from the following examples: "%D" (only number of days) "%D_%HH:%MM:%SS" "%MIN:%SEC" "%s" (total seconds) "%m", "%MIN" (total full minutes) "%h" (total full hours)

By default, all experiments use the same format string. Override that behavior by redefining the fcDirFormats list below.

Definition at line 181 of file analyze_config.py.

◆ commonStatsFileSubDir

string analyze_config.commonStatsFileSubDir = 'diagnostic_stats/obs'

statsFileSubDirs is the subdirectory within the date directory(ies) that contains the statstics files for constructing the StatsDB object examples: TODO: make StatsDB search for the correct subdirectory for each experiment 'diagnostic_stats' 'diagnostic_stats/obs' 'diagnostic_stats/model'

Definition at line 191 of file analyze_config.py.

◆ days

analyze_config.days

Definition at line 165 of file analyze_config.py.

◆ dbConf

dictionary analyze_config.dbConf = {}

Configure the StatisticsDatabase.StatsDB objects.

Definition at line 87 of file analyze_config.py.

◆ diagnosticGroupings

dictionary analyze_config.diagnosticGroupings = {}

Definition at line 82 of file analyze_config.py.

◆ hours

analyze_config.hours

cntrlExpIndex is the index of the control experiment (used for DiffCI analyses) in the expNames list

expLongNames is a list of directories within expDirectory that contains data from individual experiments

expNames is a list of experiment names used for database lookups and analysis labels, e.g., on figures. Make these names concise and exclude spaces.

DAMethods is a list of secondary labels that are within the database intermediate file names. Typical values in MPAS-Workflow applications are 'hofx' for OMF, 'variational' for OMB/OMA, and '' (empty) for model-space verification. It is optional for the user to use whatever string they choose as part of the file name to distinguish the information contained within from other files, e.g., an experiment characteristic. DAMethods is only important for file naming and is not used in the analyses

Note: refer to the StatisticsDatabase.StatsDB class for more details

Append to expLongNames, expNames, DAMethods

EDA w/ baseline config nEnsDAMembers = 20 for mem in list(range(1,nEnsDAMembers+1)): member = '{:03d}'.format(mem) dbConf['expLongNames'].append( 'guerrett_eda_3denvar_conv_clramsua_NMEM20_120km/Verification/bg/mem'+member) dbConf['expNames'].append('EDA'+member) dbConf['DAMethods'].append('omm') dbConf['cntrlExpIndex'] = nEnsDAMembers

3denvar benchmark (conventional + clear-sky AMSUA) APRIL 2021 dbConf['expLongNames'].append('guerrett_3denvar_OIE120km_unstructured/Verification/bg')

Cycle times and forecast length

First and Last CYCLE dates and increment

Definition at line 162 of file analyze_config.py.

◆ user

string analyze_config.user = 'guerrett'

hasFCLenDir whether directory structure includes forecast length note: overridden when fcTDeltaLast > fcTDeltaFirst

expDirectory is the top-level directory that contains data from all experiments. The environment variable, EXP_DIR, can be used to specify its value externally if desired.

Definition at line 96 of file analyze_config.py.

Variables