PMAT Reference

pmat_analysis.r

module

Precipitable Water Model Analysis Tool: Analysis

synopsis

This module contains analysis functions

exp.regression(t=NULL, mean.out)

Detail

Function includes all of the stuff to generate the exponential regression model with intervals

Parameters

• t (double) – training fraction

• mean.out (list) – the output of mean.filter

Returns

returns the data series and model statistics

Return type

list

lin.regression(x, y)

Detail

Linear regression function

Parameters

• x (double) – the domain of the dataset

• y (double) – the range of the dataset

Returns

returns the data series and model statistics

Return type

list

data.partition(x, y, tr.sz=tr.sz)

Detail

splits the data into a training/testing set

Parameters

• x (double) – domain of the data

• y (double) – range of the data

• tr.sz (double) – fraction of the data in the testing set

Returns

a list containing the training and testing sets

Return type

list

iterative.analysis(out.bool, mean.out)

Detail

computes regression statistics and outputs to a yaml file

Parameters

• out.bool (logical) – determine whether to generate new _output.yml

• mean.out (list) – output of mean.filter

Returns

iterative stats and _output.yml

Return type

list

lsvm(x, y, l, tr.sz=0.7, seed=sample(1:2^15, 1))

Detail

Generates a Linear Support Vector Machine and draws the decision hyperplane and support vectors

Parameters

• x (double) – domain of dataset

• y (double) – range of dataset

• l (double) – labels of the dataset

• tr.sz (double) – fraction of data to be used for model training

• seed (integer) – the random seed

Returns

list of data, labels, and the coefficients

Return type

list

pmat_processing.r

module

Precipitable Water Model Analysis Tool: Pre-processing

synopsis

functions for preprocessing

colscheme(range, set='Set1')

Detail

a function that generates an array of colors based on the number of elements

Parameters

range (list) – a list of data series

Returns

a list of colors

Return type

list

nan.filter(stuff)

Detail

removes nan values from a set of lists

Parameters

stuff (list) – list of arrays

Returns

returns list with filtered data and the indices with nans

Return type

list

mean.filter(nan.out, n)

Detail

filters the data based on the comparison of the daily std and the average std of the dataset

Parameters

• nan.out (list) – the output of nan.filter

• n (integer) – threshold

Returns

an array of indices for PWV values to be analyzed

Return type

list

dna.filter(filter.res)

Detail

removes data labels as Do Not Analyze

Parameters

filter.res (list) – overcast.filter results

Returns

overcast.filter results with DNA points removed

Return type

list

inf.counter(bool, snsr_data, label)

Detail

identifies the -Inf values

Parameters

• bool (logical) – decides if -Inf is not replaced with NaN

• snsr_data (list) – the dataset

• label (character) – the identifier for the dataset (e.g. sky, gro, skyo, groo)

Returns

data set that replaces all -Infs for NaN (If bool == FALSE).

Return type

list

index.norm(x)

Detail

calculates the normalized index of the dataset

Parameters

x (double) – data range

Returns

an array of values between 0 and 1

Return type

double

overcast.filter(col_con, col_date, col_com, pw_name, snsr_name, cloud.bool)

Detail

Filters our data with overcast condition

Parameters

• col_con (integer) – column index for condition labels

• col_date (integer) – column index for date stamp

• col_com (integer) – column index for comments

• pw_name (list) – pw measurement labels

• snsr_name (list) – sensor labels

• cloud_bool (logical) –

Returns

A list of lists containing either clear-sky/overcast data

Return type

list

sky.processing(filter.res)

Detail

Computes average values and weighted averages

Parameters

filter.res (list) – results of the overcast.filter function

Returns

series of arrays including average PWV, RH, etc.

Return type

list

pmat_products.r

module

Precipitable Water Model Analysis Tool: Products

synopsis

plotting functions for PMAT

time.pwindex(datetime)

Detail

Normalized PWV index for both clear sky and overcast data

Parameters

datetime – the datestamp of the data

time.nth_range(y, title, color, leg.lab, ylab, datetime, over.bool)

Detail

Multirange Time Series plot series

Parameters

• y (list) – the range of the plot

• title (character) – the title/description of the plot

• color (character) – the color string for the data range

• leg.lab (list) – a list of

• ylab (character) – the y-axis label

• datetime (double) – the datetime array

• over.bool (logical) – the condition of data (clear sky/overcast)

time.composite(y, title, color, ylab, datetime, over.bool)

Detail

Time Series composite plot series

Parameters

• y (list) – the range of the plot

• title (character) – the title/description of the plot

• color (character) – the color string for the data range

• ylab (character) – the y-axis label

• datetime (double) – the datetime array

• over.bool (logical) – the condition of data (clear sky/overcast)

time.mono_composite(y, title, ylab, datetime, over.bool)

Detail

Time Series composite plot series

Parameters

• y (list) – the range of the plot

• title (character) – the title/description of the plot

• ylab (character) – the y-axis label

• datetime (double) – the datetime array

• overcast (logical) – the condition of data (clear sky/overcast)

time.multiyear(y, title, color, datetime, ylab, over.bool)

Detail

Climatology plot

Parameters

• y (list) – the range of the plot

• title (character) – the title/description of the plot

• color (character) – the color string for the data range

• datetime (double) – the datetime array

• ylab (character) – the y-axis label

• overcast (logical) – the condition of data (clear sky/overcast)

analysis.nth_range(over.bool, x, y, title, label, color, leg.lab)

Detail

Comparative analysis plot

Parameters

• overcast (logical) – the condition of data (clear sky/overcast)

• x – the domain of the plot

• y – the range of the plot

• title – the title/description of the plot

• label –

• color – the color string for the data

• leg.lab –

analysis.regression(over.bool, x, y, title, label, iter)

Detail

Super Average Plot with Exponential Fit

Parameters

• overcast (bool) – the condition of data (clear sky/overcast)

• x (double) – the domain of the plot

• y (double) – the range of the plot

• title (character) – the title/description of the plot

• label (list) – list containing the x-axis and y-axis labels

• iter (list) – the output of iter.analysis

analysis.svm(model)

Detail

plots SVM

Parameters

model (list) – output of lsvm

pac.compare(over.bool, title, x, y, angular, radial)

Detail

Pac-Man plot of Super Average Plot

Parameters

• overcast (logical) – the condition of data (clear sky/overcast)

• title (character) – the title/description of the plot

• x (double) – the domain of the plot

• y (double) – the range of the plot

• angular (character) – the angular axis label and units

• radial (character) – the radial axis label and units

pac.regression(over.bool)

Detail

Pac-Man residual plot

Parameters

over.bool (bool) – the condition of data (clear sky/overcast)

chart.histogram(range, xlabel, title)

Detail

Histograms of defined quantities

Parameters

• range (double) – the range of the plot

• xlabel (list) – the x-axis label of the plot

• title (list) – the title/description of the histogram

poster.plots(over.bool, iter, mean.out)

Detail

The set of all poster

Parameters

• overcast (bool) – the condition of data (clear sky/overcast)

• iter (list) – output of iterative.analysis

• mean.out (list) –

Returns

All available poster plots

poster1(...)

poster2(over.bool, iter, mean.out)

Detail

The analytics poster plot

Parameters

• overcast (logical) – the condition of data (clear sky/overcast)

• iter – the output of iterative.analysis

• mean.out –

sensor.chart(...)

Detail

sensor specific data type distribution charts

sensor.time(over.bool)

Detail

Instrumentation time series plots

Parameters

over.bool (logical) – the condition of data (clear sky/overcast)

data.gen(over.bool, dir)

Detail

creates a datafile containing the date, avg temp, and avg pwv for a defined condition

Parameters

• over.bool (bool) – the condition of the data (clear sky/overcast)

• dir – directory path

data.ml(out.dir)

Detail

creates a datafile containing the machine learning relavant information

Parameters

out.dir (character) – directory path

data.step(seed, i, coef, r, S)

Detail

writes a yaml object that contains the analysis results of each step in the iterative analysis

Parameters

• seed (integer) – the generated seed

• i (integer) – the step number

• coef (list) – the coefficients of the best-fit

• r (double) – the RSME value

• S (double) – the standard deviation

Returns

a yaml object

Return type

list

data.final(out.dir, lengths, frac.kept, coef, std, rmse, over.bool=args$overcast)

Detail

writes the final results of iterative.analysi

Parameters

• out.dir (character) – the output directory

• lengths (list) – list of lengths

• coef (list) – the average coefficients of the best-fit

• std (double) – the average standard deviation

• rmse (double) – the average rsme values

• over.bool (logical) – the condition of data (clear sky/overcast)

visual.products(set, mean.out, datetime=datetime, over.bool=args$overcast)

Detail

saves plot sets

Parameters

• set (character) – the –set parameter

• mean.out – output of mean.filter

• datetime – the datetime range of the data

• over.bool (logical) – the condition of data (clear sky/overcast)

pmat_run.r

module

Precipitable-Water Model Analysis Tool

synopsis

The main file for PMAT. Documentation available at <https://docs.pmat.app>.

pmat_utility.r

module

Precipitable Water Model Analysis Tool: Utility

synopsis

general functions for PMAT

logg(msglevel, msg, dir=out.dir, lev='INFO')

Detail

creates log entries for _log.txt

Parameters

• msglevel (character) – log level of the message

• msg (character) – the message itself

• dir (character) – directory that stores the log file

• lev (character) – the log level of the user

aloha.first()

Detail

shows first time user information

aloha.startup()

Detail

shows title banner for program

aloha.closing()

Detail

cleans up files and ends the program

reset_time(datetime)

Detail

A function that sets the time to 00:00:00

Parameters

datetime (character) – a Date or datetime object

Returns

A datetime object with time 00:00:00

Return type

double

time_axis_init(date)

Detail

A function that calculates the min, max, and position of the tick marks for

Parameters

date (double) – A date or datetime object

Returns

The max, min, and tick mark positions

Return type

list

time_axis(datetime)

Detail

A function that sets the x-axis format for time series plots

Parameters

datetime (double) – A date or datetime object

stnd_title(des, overcast)

Detail

A function that generates the title based on

Parameters

• des (character) – the description of the plot

• overcast (logical) – the sky condition

Returns

a title string

Return type

character