Learn R Programming
RemixAutoML (version 0.5.4)

AutoCatBoostVectorCARMA: AutoCatBoostVectorCARMA

Description

AutoCatBoostVectorCARMA Multiple Regression, Mutlivariate Forecasting with calendar variables, Holiday counts, holiday lags, holiday moving averages, differencing, transformations, interaction-based categorical encoding using target variable and features to generate various time-based aggregated lags, moving averages, moving standard deviations, moving skewness, moving kurtosis, moving quantiles, parallelized interaction-based fourier pairs by grouping variables, and Trend Variables.

Usage

AutoCatBoostVectorCARMA(
  data,
  NonNegativePred = FALSE,
  RoundPreds = FALSE,
  TrainOnFull = FALSE,
  TargetColumnName = "Target",
  DateColumnName = "DateTime",
  HierarchGroups = NULL,
  GroupVariables = NULL,
  TimeWeights = 1,
  FC_Periods = 30,
  TimeUnit = "week",
  TimeGroups = c("weeks", "months"),
  NumOfParDepPlots = 10L,
  TargetTransformation = FALSE,
  Methods = c("BoxCox", "Asinh", "Asin", "Log", "LogPlus1", "Logit", "YeoJohnson"),
  AnomalyDetection = NULL,
  XREGS = NULL,
  Lags = c(1L:5L),
  MA_Periods = c(2L:5L),
  SD_Periods = NULL,
  Skew_Periods = NULL,
  Kurt_Periods = NULL,
  Quantile_Periods = NULL,
  Quantiles_Selected = c("q5", "q95"),
  Difference = TRUE,
  FourierTerms = 6L,
  CalendarVariables = c("second", "minute", "hour", "wday", "mday", "yday", "week",
    "isoweek", "month", "quarter", "year"),
  HolidayVariable = c("USPublicHolidays", "EasterGroup", "ChristmasGroup",
    "OtherEcclesticalFeasts"),
  HolidayLookback = NULL,
  HolidayLags = 1L,
  HolidayMovingAverages = 1L:2L,
  TimeTrendVariable = FALSE,
  ZeroPadSeries = NULL,
  DataTruncate = FALSE,
  SplitRatios = c(0.7, 0.2, 0.1),
  TaskType = "GPU",
  NumGPU = 1,
  PartitionType = "timeseries",
  Timer = TRUE,
  DebugMode = FALSE,
  EvalMetric = "RMSE",
  EvalMetricValue = 1.5,
  LossFunction = "RMSE",
  LossFunctionValue = 1.5,
  GridTune = FALSE,
  PassInGrid = NULL,
  ModelCount = 100,
  MaxRunsWithoutNewWinner = 50,
  MaxRunMinutes = 24L * 60L,
  Langevin = FALSE,
  DiffusionTemperature = 10000,
  NTrees = 1000,
  L2_Leaf_Reg = NULL,
  LearningRate = NULL,
  RandomStrength = 1,
  BorderCount = 254,
  Depth = 6,
  RSM = 1,
  BootStrapType = "Bayesian",
  GrowPolicy = "SymmetricTree",
  ModelSizeReg = 0.5,
  FeatureBorderType = "GreedyLogSum",
  SamplingUnit = "Group",
  SubSample = NULL,
  ScoreFunction = "Cosine",
  MinDataInLeaf = 1
)

Arguments

data

Supply your full series data set here

NonNegativePred

TRUE or FALSE

RoundPreds

Rounding predictions to an integer value. TRUE or FALSE. Defaults to FALSE

TrainOnFull

Set to TRUE to train on full data

TargetColumnName

List the column names of your target variables column. E.g. c('Target1','Target2', ..., 'TargetN')

DateColumnName

List the column name of your date column. E.g. 'DateTime'

HierarchGroups

Vector of hierachy categorical columns.

GroupVariables

Defaults to NULL. Use NULL when you have a single series. Add in GroupVariables when you have a series for every level of a group or multiple groups.

TimeWeights

NULL or a value.

FC_Periods

Set the number of periods you want to have forecasts for. E.g. 52 for weekly data to forecast a year ahead

TimeUnit

List the time unit your data is aggregated by. E.g. '1min', '5min', '10min', '15min', '30min', 'hour', 'day', 'week', 'month', 'quarter', 'year'.

TimeGroups

Select time aggregations for adding various time aggregated GDL features.

NumOfParDepPlots

Supply a number for the number of partial dependence plots you want returned

TargetTransformation

Run AutoTransformationCreate() to find best transformation for the target variable. Tests YeoJohnson, BoxCox, and Asigh (also Asin and Logit for proportion target variables).

Methods

Transformation options to test which include 'BoxCox', 'Asinh', 'Asin', 'Log', 'LogPlus1', 'Logit', 'YeoJohnson'

AnomalyDetection

NULL for not using the service. Other, provide a list, e.g. AnomalyDetection = list('tstat_high' = 4, tstat_low = -4)

XREGS

Additional data to use for model development and forecasting. Data needs to be a complete series which means both the historical and forward looking values over the specified forecast window needs to be supplied.

Lags

Select the periods for all lag variables you want to create. E.g. c(1:5,52)

MA_Periods

Select the periods for all moving average variables you want to create. E.g. c(1:5,52)

SD_Periods

Select the periods for all moving standard deviation variables you want to create. E.g. c(1:5,52)

Skew_Periods

Select the periods for all moving skewness variables you want to create. E.g. c(1:5,52)

Kurt_Periods

Select the periods for all moving kurtosis variables you want to create. E.g. c(1:5,52)

Quantile_Periods

Select the periods for all moving quantiles variables you want to create. E.g. c(1:5,52)

Quantiles_Selected

Select from the following 'q5', 'q10', 'q15', 'q20', 'q25', 'q30', 'q35', 'q40', 'q45', 'q50', 'q55', 'q60', 'q65', 'q70', 'q75', 'q80', 'q85', 'q90', 'q95'

Difference

Puts the I in ARIMA for single series and grouped series.

FourierTerms

Set to the max number of pairs. E.g. 2 means to generate two pairs for by each group level and interations if hierarchy is enabled.

CalendarVariables

NULL, or select from 'second', 'minute', 'hour', 'wday', 'mday', 'yday', 'week', 'isoweek', 'month', 'quarter', 'year'

HolidayVariable

NULL, or select from 'USPublicHolidays', 'EasterGroup', 'ChristmasGroup', 'OtherEcclesticalFeasts'

HolidayLookback

Number of days in range to compute number of holidays from a given date in the data. If NULL, the number of days are computed for you.

HolidayLags

Number of lags to build off of the holiday count variable.

HolidayMovingAverages

Number of moving averages to build off of the holiday count variable.

TimeTrendVariable

Set to TRUE to have a time trend variable added to the model. Time trend is numeric variable indicating the numeric value of each record in the time series (by group). Time trend starts at 1 for the earliest point in time and increments by one for each success time point.

ZeroPadSeries

Set to 'all', 'inner', or NULL. See TimeSeriesFill for explanation

DataTruncate

Set to TRUE to remove records with missing values from the lags and moving average features created

SplitRatios

E.g c(0.7,0.2,0.1) for train, validation, and test sets

TaskType

Has to CPU for now. If catboost makes GPU available for 'MultiRMSE' then it will be enabled. If you set to GPU the function will coerce it back to CPU.

NumGPU

Defaults to 1. If CPU is set this argument will be ignored.

PartitionType

Select 'random' for random data partitioning 'timeseries' for partitioning by time frames

Timer

Set to FALSE to turn off the updating print statements for progress

DebugMode

Defaults to FALSE. Set to TRUE to get a print statement of each high level comment in function

EvalMetric

'MultiRMSE' only. If catboost updates this I'll add more later

EvalMetricValue

Placeholder for later

LossFunction

'MultiRMSE' only. If catboost updates this I'll add more later

LossFunctionValue

Placeholder for later

GridTune

Set to TRUE to run a grid tune

PassInGrid

Defaults to NULL

ModelCount

Set the number of models to try in the grid tune

MaxRunsWithoutNewWinner

Default is 50

MaxRunMinutes

Default is 60*60

Langevin

Enables the Stochastic Gradient Langevin Boosting mode. If TRUE and TaskType == 'GPU' then TaskType will be converted to 'CPU'

DiffusionTemperature

Default is 10000

NTrees

Select the number of trees you want to have built to train the model

L2_Leaf_Reg

l2 reg parameter

LearningRate

Defaults to NULL. Catboost will dynamically define this if L2_Leaf_Reg is NULL and RMSE is chosen (otherwise catboost will default it to 0.03). Then you can pull it out of the model object and pass it back in should you wish.

RandomStrength

Default is 1

BorderCount

Default is 254

Depth

Depth of catboost model

RSM

CPU only. If TaskType is GPU then RSM will not be used

BootStrapType

If NULL, then if TaskType is GPU then Bayesian will be used. If CPU then MVS will be used. If MVS is selected when TaskType is GPU, then BootStrapType will be switched to Bayesian

GrowPolicy

Default is SymmetricTree. Others include Lossguide and Depthwise

ModelSizeReg

Defaults to 0.5. Set to 0 to allow for bigger models. This is for models with high cardinality categorical features. Valuues greater than 0 will shrink the model and quality will decline but models won't be huge.

FeatureBorderType

Defaults to 'GreedyLogSum'. Other options include: Median, Uniform, UniformAndQuantiles, MaxLogSum, MinEntropy

SamplingUnit

Default is Group. Other option is Object. if GPU is selected, this will be turned off unless the loss_function is YetiRankPairWise

SubSample

Can use if BootStrapType is neither Bayesian nor No. Pass NULL to use Catboost default. Used for bagging.

ScoreFunction

Default is Cosine. CPU options are Cosine and L2. GPU options are Cosine, L2, NewtonL2, and NewtomCosine (not available for Lossguide)

MinDataInLeaf

Defaults to 1. Used if GrowPolicy is not SymmetricTree

Value

Returns a data.table of original series and forecasts, the catboost model objects (everything returned from AutoCatBoostRegression()), a time series forecast plot, and transformation info if you set TargetTransformation to TRUE. The time series forecast plot will plot your single series or aggregate your data to a single series and create a plot from that.

See Also

Other Automated Panel Data Forecasting: AutoCatBoostCARMA(), AutoCatBoostHurdleCARMA(), AutoH2OCARMA(), AutoLightGBMCARMA(), AutoXGBoostCARMA()

Examples

Run this code
# NOT RUN {
# Two group variables and xregs

# Load Walmart Data from Dropbox
data <- data.table::fread(
 'https://www.dropbox.com/s/2str3ek4f4cheqi/walmart_train.csv?dl=1')

# Filter out zeros
data <- data[Weekly_Sales != 0]

# Subset for Stores / Departments With Full Series
data <- data[, Counts := .N, by = c('Store','Dept')][Counts == 143][
 , Counts := NULL]

# Subset Columns (remove IsHoliday column)----
keep <- c('Store','Dept','Date','Weekly_Sales')
data <- data[, ..keep]
data <- data[Store %in% c(1,2)]
xregs <- data.table::copy(data)
xregs[, GroupVar := do.call(paste, c(.SD, sep = ' ')), .SDcols = c('Store','Dept')]
xregs[, c('Store','Dept') := NULL]
data.table::setnames(xregs, 'Weekly_Sales', 'Other')
xregs[, Other := jitter(Other, factor = 25)]
data <- data[as.Date(Date) < as.Date('2012-09-28')]

# Vector CARMA testing
data[, Weekly_Profit := Weekly_Sales * 0.75]

# Build forecast
CatBoostResults <- RemixAutoML::AutoCatBoostVectorCARMA(

  # data args
  data = data, # TwoGroup_Data,
  TargetColumnName = c('Weekly_Sales','Weekly_Profit'),
  DateColumnName = 'Date',
  HierarchGroups = NULL,
  GroupVariables = c('Store','Dept'),
  TimeWeights = 1,
  TimeUnit = 'weeks',
  TimeGroups = c('weeks','months'),

  # Production args
  TaskType = 'GPU',
  NumGPU = 1,
  TrainOnFull = TRUE,
  SplitRatios = c(1 - 10 / 138, 10 / 138),
  PartitionType = 'random',
  FC_Periods = 4,
  Timer = TRUE,
  DebugMode = TRUE,

  # Target transformations
  TargetTransformation = TRUE,
  Methods = c('BoxCox', 'Asinh', 'Asin', 'Log',
              'LogPlus1', 'Logit', 'YeoJohnson'),
  Difference = FALSE,
  NonNegativePred = FALSE,
  RoundPreds = FALSE,

  # Date features
  CalendarVariables = c('week', 'month', 'quarter'),
  HolidayVariable = c('USPublicHolidays',
                      'EasterGroup',
                      'ChristmasGroup','OtherEcclesticalFeasts'),
  HolidayLookback = NULL,
  HolidayLags = 1,
  HolidayMovingAverages = 1:2,

  # Time series features
  Lags = list('weeks' = seq(2L, 10L, 2L),
              'months' = c(1:3)),
  MA_Periods = list('weeks' = seq(2L, 10L, 2L),
                    'months' = c(2,3)),
  SD_Periods = NULL,
  Skew_Periods = NULL,
  Kurt_Periods = NULL,
  Quantile_Periods = NULL,
  Quantiles_Selected = c('q5','q95'),

  # Bonus features
  AnomalyDetection = NULL,
  XREGS = xregs,
  FourierTerms = 2,
  TimeTrendVariable = TRUE,
  ZeroPadSeries = NULL,
  DataTruncate = FALSE,

  # Eval args
  NumOfParDepPlots = 100L,
  EvalMetric = 'MultiRMSE',
  EvalMetricValue = 1.5,
  LossFunction = 'MultiRMSE',
  LossFunctionValue = 1.5,

  # Grid args
  GridTune = FALSE,
  PassInGrid = NULL,
  ModelCount = 5,
  MaxRunsWithoutNewWinner = 50,
  MaxRunMinutes = 60*60,

  # ML Args
  NTrees = 1000,
  Depth = 6,
  LearningRate = NULL,
  L2_Leaf_Reg = NULL,
  RandomStrength = 1,
  BorderCount = 254,
  RSM = 1,
  BootStrapType = 'Bayesian',
  GrowPolicy = 'SymmetricTree',
  Langevin = FALSE,
  DiffusionTemperature = 10000,
  ModelSizeReg = 0.5,
  FeatureBorderType = 'GreedyLogSum',
  SamplingUnit = 'Group',
  SubSample = NULL,
  ScoreFunction = 'Cosine',
  MinDataInLeaf = 1)
# }

Run the code above in your browser using DataLab