saymrwulf/prophet
1
0
Fork 0
mirror of https://github.com/saymrwulf/prophet.git synced 2026-07-05 04:17:56 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix bug about holiday year different from input year and others (#670)

Browse source 
* Allow both both hoidays and append holidays

Match holidays in predict and fit

Add test for append_holiday features; minor fixes

Add column name validation for append_holidays names; allow only one country

Fix bug about holiday year different from input year and also other bugs

Change function description

Add append holiday feature for R

* Add test for R/Pyrhon; fix bugs
This commit is contained in:
ziye666 2018-10-18 17:40:17 -07:00 committed by Ben Letham
parent ad095ac690
commit 95fa4460a9
20 changed files with 67980 additions and 163 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
R/DESCRIPTION
View file

@ -32,7 +32,7 @@ Suggests:
readr
License: BSD_3_clause + file LICENSE
LazyData: true
RoxygenNote: 6.0.1
RoxygenNote: 6.1.0
VignetteBuilder: knitr
SystemRequirements: C++11
Encoding: UTF-8

2
R/NAMESPACE
View file

@ -8,7 +8,9 @@ export(add_seasonality)
export(cross_validation)
export(dyplot.prophet)
export(fit.prophet)
export(get_holiday_names)
export(make_future_dataframe)
export(make_holidays_df)
export(performance_metrics)
export(plot_cross_validation_metric)
export(plot_forecast_component)

12
R/R/data.R Normal file
View file

@ -0,0 +1,12 @@
## Copyright (c) 2017-present, Facebook, Inc.
## All rights reserved.
## This source code is licensed under the BSD-style license found in the
## LICENSE file in the root directory of this source tree. An additional grant
## of patent rights can be found in the PATENTS file in the same directory.
#' holidays table
#'
#' @format A data frame with five variables: ds, holiday, country, year
"generated_holidays"

1
R/R/diagnostics.R
View file

@ -154,6 +154,7 @@ prophet_copy <- function(m, cutoff = NULL) {
weekly.seasonality = FALSE,
daily.seasonality = FALSE,
holidays = m$holidays,
append.holidays = m$append.holidays,
seasonality.mode = m$seasonality.mode,
seasonality.prior.scale = m$seasonality.prior.scale,
changepoint.prior.scale = m$changepoint.prior.scale,

46
R/R/make_holidays.R Normal file
View file

@ -0,0 +1,46 @@
## Copyright (c) 2017-present, Facebook, Inc.
## All rights reserved.
## This source code is licensed under the BSD-style license found in the
## LICENSE file in the root directory of this source tree. An additional grant
## of patent rights can be found in the PATENTS file in the same directory.
#' Return all possible holiday names of given country
#'
#' @param country.name Country name (character).
#'
#' @return A vector of all possible holiday names (unique) of given country.
#' @export
get_holiday_names <- function(country.name){
holidays <- generated_holidays %>%
dplyr::filter(country == country.name) %>%
dplyr::select(holiday) %>%
unique()
return(holidays$holiday)
}
#' Make dataframe of holidays for given years and countries
#'
#' @param country.name Country name (character).
#'
#' @return Dataframe with 'ds' and 'holiday', which can directly feed
#' to 'holidays' params in Prophet
#' @export
make_holidays_df <- function(years, country.name){
country.holidays = generated_holidays %>%
dplyr::filter(country == country.name)
max.year <- max(country.holidays$year)
min.year <- min(country.holidays$year)
if (max(years) > max.year || min(years) < min.year){
warning.msg = paste("Holidays for", country.name, "are only supported from", min.year,
"to", max.year)
warning(warning.msg)
}
holidays.df <- country.holidays %>%
dplyr::filter(year %in% years) %>%
dplyr::select(ds, holiday) %>%
data.frame
return(holidays.df)
}

88
R/R/prophet.R
View file

@ -8,8 +8,8 @@
## Makes R CMD CHECK happy due to dplyr syntax below
globalVariables(c(
"ds", "y", "cap", ".",
"component", "dow", "doy", "holiday", "holidays", "holidays_lower", "holidays_upper", "ix",
"lower", "n", "stat", "trend", "row_number", "extra_regressors", "col",
"component", "dow", "doy", "holiday", "holidays", "append.holidays", "holidays_lower",
"holidays_upper", "ix", "lower", "n", "stat", "trend", "row_number", "extra_regressors", "col",
"trend_lower", "trend_upper", "upper", "value", "weekly", "weekly_lower", "weekly_upper",
"x", "yearly", "yearly_lower", "yearly_upper", "yhat", "yhat_lower", "yhat_upper"))
@ -43,6 +43,7 @@ globalVariables(c(
#' range of days around the date to be included as holidays. lower_window=-2
#' will include 2 days prior to the date as holidays. Also optionally can have
#' a column prior_scale specifying the prior scale for each holiday.
#' @param append.holidays country name or abbreviation (character).
#' @param seasonality.mode 'additive' (default) or 'multiplicative'.
#' @param seasonality.prior.scale Parameter modulating the strength of the
#' seasonality model. Larger values allow the model to fit larger seasonal
@ -87,6 +88,7 @@ prophet <- function(df = NULL,
weekly.seasonality = 'auto',
daily.seasonality = 'auto',
holidays = NULL,
append.holidays = NULL,
seasonality.mode = 'additive',
seasonality.prior.scale = 10,
holidays.prior.scale = 10,
@ -110,6 +112,7 @@ prophet <- function(df = NULL,
weekly.seasonality = weekly.seasonality,
daily.seasonality = daily.seasonality,
holidays = holidays,
append.holidays = append.holidays,
seasonality.mode = seasonality.mode,
seasonality.prior.scale = seasonality.prior.scale,
changepoint.prior.scale = changepoint.prior.scale,
@ -129,6 +132,7 @@ prophet <- function(df = NULL,
params = list(),
history = NULL,
history.dates = NULL,
train.holiday.names = NULL,
train.component.cols = NULL,
component.modes = NULL
)
@ -177,6 +181,11 @@ validate_inputs <- function(m) {
validate_column_name(m, h, check_holidays = FALSE)
}
}
if (!is.null(m$append.holidays)) {
if (!(m$append.holidays %in% generated_holidays$country)){
stop("Holidays in ", m$append.holidays," are not currently supported!")
}
}
if (!(m$seasonality.mode %in% c('additive', 'multiplicative'))) {
stop("seasonality.mode must be 'additive' or 'multiplicative'")
}
@ -214,6 +223,11 @@ validate_column_name <- function(
(name %in% unique(m$holidays$holiday))){
stop("Name ", name, " already used for a holiday.")
}
if(check_holidays & !is.null(m$append.holidays)){
if(name %in% get_holiday_names(m$append.holidays)){
stop("Name ", name, " is a holiday name in ", m$append.holidays, ".")
}
}
if(check_seasonalities & (!is.null(m$seasonalities[[name]]))){
stop("Name ", name, " already used for a seasonality.")
}
@ -534,10 +548,28 @@ make_seasonality_features <- function(dates, period, series.order, prefix) {
make_holiday_features <- function(m, dates) {
# Strip dates to be just days, for joining on holidays
dates <- set_date(format(dates, "%Y-%m-%d"))
wide <- m$holidays %>%
all.holidays <- m$holidays
if (!is.null(m$append.holidays)){
years <- as.numeric(unique(format(dates, "%Y")))
append.holidays.df <- make_holidays_df(years, m$append.holidays) %>%
dplyr::mutate(ds=as.character(ds), holiday=as.character(holiday))
all.holidays <- suppressWarnings(dplyr::bind_rows(all.holidays, append.holidays.df))
}
# Make fit.prophet and predict.prophet holidays components match
if (!is.null(m$append.holidays) && !is.null(m$train.holiday.names)){
row.to.keep <- which(all.holidays$holiday %in% m$train.holiday.names)
all.holidays <- all.holidays[row.to.keep,]
holidays.to.add <- data.frame(holiday=setdiff(m$train.holiday.names,
all.holidays$holiday))
all.holidays <- suppressWarnings(dplyr::bind_rows(all.holidays, holidays.to.add))
}
if (nrow(all.holidays)==0){
return(NULL)
}
wide <- all.holidays %>%
dplyr::mutate(ds = set_date(ds)) %>%
dplyr::group_by(holiday, ds) %>%
dplyr::filter(row_number() == 1) %>%
dplyr::filter(dplyr::row_number() == 1) %>%
dplyr::do({
if (exists('lower_window', where = .) && !is.na(.$lower_window)
&& !is.na(.$upper_window)) {
@ -555,16 +587,17 @@ make_holiday_features <- function(m, dates) {
holiday.features <- data.frame(ds = set_date(dates)) %>%
dplyr::left_join(wide, by = 'ds') %>%
dplyr::select(-ds)
# Make sure fit.prophet and predict.prophet component.cols perfectly equal
holiday.features <- holiday.features %>% dplyr::select(sort(names(.)))
holiday.features[is.na(holiday.features)] <- 0
# Prior scales
if (!('prior_scale' %in% colnames(m$holidays))) {
m$holidays$prior_scale <- m$holidays.prior.scale
if (!('prior_scale' %in% colnames(all.holidays))) {
all.holidays$prior_scale <- m$holidays.prior.scale
}
prior.scales.list <- list()
for (name in unique(m$holidays$holiday)) {
df.h <- m$holidays[m$holidays$holiday == name, ]
for (name in unique(all.holidays$holiday)) {
df.h <- all.holidays[all.holidays$holiday == name, ]
ps <- unique(df.h$prior_scale)
if (length(ps) > 1) {
stop('Holiday ', name, ' does not have a consistent prior scale ',
@ -584,9 +617,14 @@ make_holiday_features <- function(m, dates) {
sn <- strsplit(name, '_delim_', fixed = TRUE)[[1]][1]
prior.scales <- c(prior.scales, prior.scales.list[[sn]])
}
return(list(holiday.features = holiday.features,
holiday.names <- names(prior.scales.list)
if (is.null(m$train.holiday.names)){
m$train.holiday.names <- holiday.names
}
return(list(m = m,
holiday.features = holiday.features,
prior.scales = prior.scales,
holiday.names = names(prior.scales.list)))
holiday.names = holiday.names))
}
#' Add an additional regressor to be used for fitting and predicting.
@ -738,12 +776,15 @@ make_all_seasonality_features <- function(m, df) {
}
# Holiday features
if (!is.null(m$holidays)) {
hf <- make_holiday_features(m, df$ds)
seasonal.features <- cbind(seasonal.features, hf$holiday.features)
prior.scales <- c(prior.scales, hf$prior.scales)
modes[[m$seasonality.mode]] <- c(
modes[[m$seasonality.mode]], hf$holiday.names)
if (!is.null(m$holidays) || !is.null(m$append.holidays)) {
out <- make_holiday_features(m, df$ds)
if (!is.null(out)){
m <- out$m
seasonal.features <- cbind(seasonal.features, out$holiday.features)
prior.scales <- c(prior.scales, out$prior.scales)
modes[[m$seasonality.mode]] <- c(
modes[[m$seasonality.mode]], out$holiday.names)
}
}
# Additional regressors
@ -761,7 +802,8 @@ make_all_seasonality_features <- function(m, df) {
}
components.list <- regressor_column_matrix(m, seasonal.features, modes)
return(list(seasonal.features = seasonal.features,
return(list(m = m,
seasonal.features = seasonal.features,
prior.scales = prior.scales,
component.cols = components.list$component.cols,
modes = components.list$modes))
@ -792,9 +834,9 @@ regressor_column_matrix <- function(m, seasonal.features, modes) {
extra = "merge", fill = "right") %>%
dplyr::select(col, component)
# Add total for holidays
if(!is.null(m$holidays)){
if(!is.null(m$train.holiday.names)){
components <- add_group_component(
components, 'holidays', unique(m$holidays$holiday))
components, 'holidays', unique(m$train.holiday.names))
}
# Add totals for additive and multiplicative components, and regressors
for (mode in c('additive', 'multiplicative')) {
@ -1061,6 +1103,7 @@ fit.prophet <- function(m, df, ...) {
m$history <- history
m <- set_auto_seasonalities(m)
out2 <- make_all_seasonality_features(m, history)
m <- out2$m
seasonal.features <- out2$seasonal.features
prior.scales <- out2$prior.scales
component.cols <- out2$component.cols
@ -1136,7 +1179,7 @@ fit.prophet <- function(m, df, ...) {
m$params <- stan.fit$par
n.iteration <- 1
}
# Cast the parameters to have consistent form, whether full bayes or MAP
for (name in c('delta', 'beta')){
m$params[[name]] <- matrix(m$params[[name]], nrow = n.iteration)
@ -1296,6 +1339,7 @@ predict_trend <- function(model, df) {
#' @keywords internal
predict_seasonal_components <- function(m, df) {
out <- make_all_seasonality_features(m, df)
m <- out$m
seasonal.features <- out$seasonal.features
component.cols <- out$component.cols
lower.p <- (1 - m$interval.width)/2

10
R/data-raw/generated_holidays.R Normal file
View file

@ -0,0 +1,10 @@
## Copyright (c) 2017-present, Facebook, Inc.
## All rights reserved.
## This source code is licensed under the BSD-style license found in the
## LICENSE file in the root directory of this source tree. An additional grant
## of patent rights can be found in the PATENTS file in the same directory.
generated_holidays <- read.csv("data-raw/generated_holidays.csv")
devtools::use_data(generated_holidays, overwrite = TRUE)

67401
R/data-raw/generated_holidays.csv Normal file
View file

File diff suppressed because it is too large Load diff

BIN
R/data/generated_holidays.rda Normal file
View file

Binary file not shown.

14
R/man/generated_holidays.Rd Normal file
View file

@ -0,0 +1,14 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/data.R
\docType{data}
\name{generated_holidays}
\alias{generated_holidays}
\title{holidays table}
\format{A data frame with five variables: ds, holiday, country, year}
\usage{
generated_holidays
}
\description{
holidays table
}
\keyword{datasets}

17
R/man/get_holiday_names.Rd Normal file
View file

@ -0,0 +1,17 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/make_holidays.R
\name{get_holiday_names}
\alias{get_holiday_names}
\title{Return all possible holiday names of given country}
\usage{
get_holiday_names(country.name)
}
\arguments{
\item{country.name}{Country name (character).}
}
\value{
A vector of all possible holiday names (unique) of given country.
}
\description{
Return all possible holiday names of given country
}

18
R/man/make_holidays_df.Rd Normal file
View file

@ -0,0 +1,18 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/make_holidays.R
\name{make_holidays_df}
\alias{make_holidays_df}
\title{Make dataframe of holidays for given years and countries}
\usage{
make_holidays_df(years, country.name)
}
\arguments{
\item{country.name}{Country name (character).}
}
\value{
Dataframe with 'ds' and 'holiday', which can directly feed
to 'holidays' params in Prophet
}
\description{
Make dataframe of holidays for given years and countries
}

3
R/man/plot_forecast_component.Rd
View file

@ -4,7 +4,8 @@
\alias{plot_forecast_component}
\title{Plot a particular component of the forecast.}
\usage{
plot_forecast_component(m, fcst, name, uncertainty = TRUE, plot_cap = FALSE)
plot_forecast_component(m, fcst, name, uncertainty = TRUE,
plot_cap = FALSE)
}
\arguments{
\item{m}{Prophet model}

10
R/man/prophet.Rd
View file

@ -8,10 +8,10 @@ prophet(df = NULL, growth = "linear", changepoints = NULL,
n.changepoints = 25, changepoint.range = 0.8,
yearly.seasonality = "auto", weekly.seasonality = "auto",
daily.seasonality = "auto", holidays = NULL,
seasonality.mode = "additive", seasonality.prior.scale = 10,
holidays.prior.scale = 10, changepoint.prior.scale = 0.05,
mcmc.samples = 0, interval.width = 0.8, uncertainty.samples = 1000,
fit = TRUE, ...)
append.holidays = NULL, seasonality.mode = "additive",
seasonality.prior.scale = 10, holidays.prior.scale = 10,
changepoint.prior.scale = 0.05, mcmc.samples = 0,
interval.width = 0.8, uncertainty.samples = 1000, fit = TRUE, ...)
}
\arguments{
\item{df}{(optional) Dataframe containing the history. Must have columns ds
@ -51,6 +51,8 @@ range of days around the date to be included as holidays. lower_window=-2
will include 2 days prior to the date as holidays. Also optionally can have
a column prior_scale specifying the prior scale for each holiday.}
\item{append.holidays}{country name or abbreviation (character).}
\item{seasonality.mode}{'additive' (default) or 'multiplicative'.}
\item{seasonality.prior.scale}{Parameter modulating the strength of the

4
R/tests/testthat/test_diagnostics.R
View file

@ -121,6 +121,7 @@ test_that("copy", {
weekly.seasonality = c(TRUE, FALSE),
daily.seasonality = c(TRUE, FALSE),
holidays = c('null', 'insert_dataframe'),
append.holidays = c(NULL, 'US'),
seasonality.mode = c('additive', 'multiplicative')
)
products <- expand.grid(inputs)
@ -139,6 +140,7 @@ test_that("copy", {
weekly.seasonality = products$weekly.seasonality[i],
daily.seasonality = products$daily.seasonality[i],
holidays = holidays,
append.holidays = products$append.holidays[i],
seasonality.prior.scale = 1.1,
holidays.prior.scale = 1.1,
changepoints.prior.scale = 0.1,
@ -153,7 +155,7 @@ test_that("copy", {
m1 <- prophet:::set_auto_seasonalities(m1)
m2 <- prophet:::prophet_copy(m1)
# Values should be copied correctly
args <- c('growth', 'changepoints', 'n.changepoints', 'holidays',
args <- c('growth', 'changepoints', 'n.changepoints', 'holidays', 'append.holidays',
'seasonality.prior.scale', 'holidays.prior.scale',
'changepoints.prior.scale', 'mcmc.samples', 'interval.width',
'uncertainty.samples', 'seasonality.mode', 'changepoint.range')

43
R/tests/testthat/test_prophet.R
View file

@ -349,6 +349,49 @@ test_that("fit_with_holidays", {
expect_error(predict(m), NA)
})
test_that("fit_with_append_holidays", {
skip_if_not(Sys.getenv('R_ARCH') != '/i386')
holidays <- data.frame(ds = c('2012-06-06', '2013-06-06'),
holiday = c('seans-bday', 'seans-bday'),
lower_window = c(0, 0),
upper_window = c(1, 1))
append.holidays = 'US'
# Test with holidays and append_holidays
m <- prophet(DATA,
holidays = holidays,
append.holidays = append.holidays,
uncertainty.samples = 0)
expect_error(predict(m), NA)
# There are training holidays missing in the test set
train2 <- DATA %>% head(155)
future2 <- DATA %>% tail(355)
model <- prophet(train2,
append.holidays = append.holidays,
uncertainty.samples = 0)
expect_error(predict(m, future2), NA)
# There are test holidays missing in the training set
train2 <- DATA %>% tail(355)
future2 <- DATA2
model <- prophet(train2,
append.holidays = append.holidays,
uncertainty.samples = 0)
expect_error(predict(m, future2), NA)
# Append_holidays with non-existing year
max.year <- generated_holidays %>%
dplyr::filter(country==append.holidays) %>%
dplyr::select(year) %>%
max()
train2 <- data.frame('ds'=c(paste(max.year+1, "-01-01", sep=''),
paste(max.year+1, "-01-02", sep='')),
'y'=1)
expect_warning(prophet(train2,
append.holidays = append.holidays))
# Append_holidays with non-existing country
append.holidays = 'Utopia'
expect_error(prophet(DATA,
append.holidays = append.holidays))
})
test_that("make_future_dataframe", {
skip_if_not(Sys.getenv('R_ARCH') != '/i386')
train.t <- DATA[1:234, ]

402
python/fbprophet/hdays.py
View file

@ -17,6 +17,7 @@ from datetime import date, timedelta
from convertdate.islamic import to_gregorian, from_gregorian
import warnings
# Official public holidays at a country level
# ------------ Holidays in Brazil---------------------
class Brazil(HolidayBase):
@ -89,7 +90,10 @@ class Indonesia(HolidayBase):
# Chinese New Year/ Spring Festival
name = "Chinese New Year"
self[LunarDate(year, 1, 1).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 1, 1).toSolarDate()
if ds.year == year:
self[ds] = name
# Day of Silence / Nyepi
# Note:
@ -126,10 +130,12 @@ class Indonesia(HolidayBase):
pass
# Ascension of the Prophet
islam_year = from_gregorian(year, 3, 17)[0]
y, m, d = to_gregorian(islam_year, 7, 27)
name = "Ascension of the Prophet"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 3, 17)[0]
y, m, d = to_gregorian(islam_year, 7, 27)
if y == year:
self[date(y, m, d)] = name
# Labor Day
name = "Labor Day"
@ -137,12 +143,17 @@ class Indonesia(HolidayBase):
# Ascension of Jesus Christ
name = "Ascension of Jesus"
hdate = easter(year) + rd(days=+39)
self[hdate] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) + rd(days=+39)
if ds.year == year:
self[ds] = name
# Buddha's Birthday
name = "Buddha's Birthday"
self[LunarDate(year, 4, 15).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 4, 15).toSolarDate()
if ds.year == year:
self[ds] = name
# Pancasila Day, since 2017
if year >= 2017:
@ -150,34 +161,43 @@ class Indonesia(HolidayBase):
self[date(year, 6, 1)] = name
# Eid al-Fitr
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
name = "Eid al-Fitr"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
if y1 == year:
self[date(y1, m2, d2)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
# Independence Day
name = "Independence Day"
self[date(year, 8, 17)] = name
# Feast of the Sacrifice
islam_year = from_gregorian(year, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
name = "Feast of the Sacrifice"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
if y == year:
self[date(y, m, d)] = name
# Islamic New Year
islam_year = from_gregorian(year, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
name = "Islamic New Year"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
if y == year:
self[date(y, m, d)] = name
# Birth of the Prophet
islam_year = from_gregorian(year, 11, 20)[0]
y, m, d = to_gregorian(islam_year + 1, 3, 12)
name = "Birth of the Prophet"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 11, 20)[0]
y, m, d = to_gregorian(islam_year + 1, 3, 12)
if y == year:
self[date(y, m, d)] = name
# Christmas
self[date(year, 12, 25)] = "Christmas"
@ -298,32 +318,41 @@ class India(HolidayBase):
# Day of Ashura
# 10th day of 1st Islamic month
islam_year = from_gregorian(year, 10, 1)[0]
y, m, d = to_gregorian(islam_year, 1, 10)
name = "Day of Ashura"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 10, 1)[0]
y, m, d = to_gregorian(islam_year, 1, 10)
if y == year:
self[date(y, m, d)] = name
# Mawlid, Birth of the Prophet
# 12th day of 3rd Islamic month
islam_year = from_gregorian(year, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
name = "Mawlid"
self[date(y, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
if y == year:
self[date(y, m, d)] = name
# Eid ul-Fitr
# 1st and 2nd day of 10th Islamic month
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
name = "Eid al-Fitr"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
# Eid al-Adha, i.e., Feast of the Sacrifice
islam_year = from_gregorian(year, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
name = "Feast of the Sacrifice"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
if y == year:
self[date(y, m, d)] = name
# --------------------------------
# Christian holidays
@ -345,23 +374,38 @@ class India(HolidayBase):
# Palm Sunday
name = "Palm Sunday"
self[easter(year) - rd(days=7)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=7)
if ds.year == year:
self[ds] = name
# Maundy Thursday
name = "Maundy Thursday"
self[easter(year) - rd(days=3)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=3)
if ds.year == year:
self[ds] = name
# Good Friday
name = "Good Friday"
self[easter(year) - rd(days=2)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=2)
if ds.year == year:
self[ds] = name
# Easter Sunday
name = "Easter Sunday"
self[easter(year)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset)
if ds.year == year:
self[ds] = name
# Feast of Pentecost
name = "Feast of Pentecost"
self[easter(year) + rd(days=49)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) + rd(days=49)
if ds.year == year:
self[ds] = name
# Fest of St. Theresa of Calcutta
name = "Fest of St. Theresa of Calcutta"
@ -416,14 +460,19 @@ class Malaysia(HolidayBase):
# Birthday of Prophet, Mawlid in India
# 12th day of 3rd Islamic month
islam_year = from_gregorian(year, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
name = "Birth of Prophet"
self[date(y, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
if y == year:
self[date(y, m, d)] = name
# Chinese New Year
name = "Chinese New Year"
self[LunarDate(year, 1, 1).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 1, 1).toSolarDate()
if ds.year == year:
self[ds] = name
# Tamil New Year
# Note: it's not necessarily 04/14
@ -434,7 +483,10 @@ class Malaysia(HolidayBase):
# Good Friday
name = "Good Friday"
self[easter(year) - rd(days=2)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=2)
if ds.year == year:
self[ds] = name
# Labor Day
name = "Labor Day"
@ -442,7 +494,10 @@ class Malaysia(HolidayBase):
# Buddha's Birthday
name = "Wesak Day"
self[LunarDate(year, 4, 15).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 4, 15).toSolarDate()
if ds.year == year:
self[ds] = name
# King's birthday
# https://www.thestar.com.my/news/nation/2017/04/26/
@ -471,28 +526,35 @@ class Malaysia(HolidayBase):
self[saturdays[-1]] = name
# Eid al-Fitr
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
name = "Eid al-Fitr"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
# Malaysia Day
name = "Malaysia Day"
self[date(year, 9, 16)] = name
# Feast of the Sacrifice
islam_year = from_gregorian(year, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
name = "Feast of the Sacrifice"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
if y == year:
self[date(y, m, d)] = name
# First Day of Muharram
islam_year = from_gregorian(year, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
name = "First Day of Muharram"
self[date(y, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
if y == year:
self[date(y, m, d)] = name
# Christmas
name = "Christmas Day"
@ -523,16 +585,35 @@ class Vietnam(HolidayBase):
# Vietnamese New Year
name = "Vietnamese New Year"
self[LunarDate(year - 1, 12, 30).toSolarDate()] = name
self[LunarDate(year, 1, 1).toSolarDate()] = name
self[LunarDate(year, 1, 2).toSolarDate()] = name
self[LunarDate(year, 1, 3).toSolarDate()] = name
self[LunarDate(year, 1, 4).toSolarDate()] = name
self[LunarDate(year, 1, 5).toSolarDate()] = name
for offset in range(-1, 2, 1):
try:
ds = LunarDate(year - 1 + offset, 12, 30).toSolarDate()
except ValueError:
ds = LunarDate(year - 1 + offset, 12, 29).toSolarDate()
if ds.year == year:
self[ds] = name
ds = LunarDate(year + offset, 1, 1).toSolarDate()
if ds.year == year:
self[ds] = name
ds = LunarDate(year + offset, 1, 2).toSolarDate()
if ds.year == year:
self[ds] = name
ds = LunarDate(year + offset, 1, 3).toSolarDate()
if ds.year == year:
self[ds] = name
ds = LunarDate(year + offset, 1, 4).toSolarDate()
if ds.year == year:
self[ds] = name
ds = LunarDate(year + offset, 1, 5).toSolarDate()
if ds.year == year:
self[ds] = name
# Hung Kings Commemorations
name = "Hung Kings Commemorations"
self[LunarDate(year, 3, 10).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 3, 10).toSolarDate()
if ds.year == year:
self[ds] = name
# Reunification Day
name = "Reunification Day"
@ -606,7 +687,10 @@ class Thailand(HolidayBase):
# Buddha's Birthday
name = "Buddha's Birthday"
self[LunarDate(year, 4, 15).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 4, 15).toSolarDate()
if ds.year == year:
self[ds] = name
# Coronation Day, removed in 2017
name = "Coronation Day"
@ -754,11 +838,17 @@ class Philippines(HolidayBase):
# Maundy Thursday
name = "Maundy Thursday"
self[easter(year) - rd(days=3)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=3)
if ds.year == year:
self[ds] = name
# Good Friday
name = "Good Friday"
self[easter(year) - rd(days=2)] = name
for offset in range(-1, 2, 1):
ds = easter(year + offset) - rd(days=2)
if ds.year == year:
self[ds] = name
# Day of Valor
name = "Day of Valor"
@ -774,15 +864,20 @@ class Philippines(HolidayBase):
# Eid al-Fitr
name = "Eid al-Fitr"
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
self[date(year, m1, d1) - timedelta(days=1)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y, m, d = to_gregorian(islam_year, 10, 1)
ds = date(y, m, d) - timedelta(days=1)
if ds.year == year:
self[ds] = name
# Eid al-Adha, i.e., Feast of the Sacrifice
name = "Feast of the Sacrifice"
islam_year = from_gregorian(year, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
if y == year:
self[date(y, m, d)] = name
# National Heroes' Day
name = "National Heroes' Day"
@ -849,15 +944,20 @@ class Turkey(HolidayBase):
# Eid al-Fitr
name = "Eid al-Fitr"
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
self[date(year, m1, d1) - timedelta(days=1)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y, m, d = to_gregorian(islam_year, 10, 1)
ds = date(y, m, d) - timedelta(days=1)
if ds.year == year:
self[ds] = name
# Eid al-Adha, i.e., Feast of the Sacrifice
name = "Feast of the Sacrifice"
islam_year = from_gregorian(year, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y, m, d = to_gregorian(islam_year, 12, 10)
if y == year:
self[date(y, m, d)] = name
class TU(Turkey):
@ -906,45 +1006,60 @@ class Pakistan(HolidayBase):
# Eid al-Adha, i.e., Feast of the Sacrifice
name = "Feast of the Sacrifice"
islam_year = from_gregorian(year, 8, 22)[0]
y1, m1, d1 = to_gregorian(islam_year, 12, 10)
y2, m2, d2 = to_gregorian(islam_year, 12, 11)
y3, m3, d3 = to_gregorian(islam_year, 12, 12)
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
self[date(year, m3, d3)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y1, m1, d1 = to_gregorian(islam_year, 12, 10)
y2, m2, d2 = to_gregorian(islam_year, 12, 11)
y3, m3, d3 = to_gregorian(islam_year, 12, 12)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
if y3 == year:
self[date(y3, m3, d3)] = name
# Eid al-Fitr
name = "Eid al-Fitr"
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
y3, m3, d3 = to_gregorian(islam_year, 10, 3)
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
self[date(year, m3, d3)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
y3, m3, d3 = to_gregorian(islam_year, 10, 3)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
if y3 == year:
self[date(y3, m3, d3)] = name
# Mawlid, Birth of the Prophet
# 12th day of 3rd Islamic month
islam_year = from_gregorian(year, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
name = "Mawlid"
self[date(y, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
if y == year:
self[date(y, m, d)] = name
# Day of Ashura
# 10th and 11th days of 1st Islamic month
islam_year = from_gregorian(year, 10, 1)[0]
y1, m1, d1 = to_gregorian(islam_year, 1, 10)
y2, m2, d2 = to_gregorian(islam_year, 1, 11)
name = "Day of Ashura"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 10, 1)[0]
y1, m1, d1 = to_gregorian(islam_year, 1, 10)
y2, m2, d2 = to_gregorian(islam_year, 1, 11)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
# Shab e Mairaj
islam_year = from_gregorian(year, 4, 13)[0]
y1, m1, d1 = to_gregorian(islam_year, 7, 27)
name = "Shab e Mairaj"
self[date(year, m1, d1)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 4, 13)[0]
y, m, d = to_gregorian(islam_year, 7, 27)
if y == year:
self[date(y, m, d)] = name
# Defence Day
name = "Defence Day"
@ -1052,44 +1167,60 @@ class Egypt(HolidayBase):
# Sham El Nessim
# The Monday following Orthodox Easter
name = "Sham El Nessim"
orthodox_easter = easter(year, method=2)
self[orthodox_easter + timedelta(days=1)] = name
for offset in range(-1, 2, 1):
orthodox_easter = easter(year + offset, method=2)
ds = orthodox_easter + timedelta(days=1)
if ds.year == year:
self[ds] = name
# Islamic New Year
islam_year = from_gregorian(year, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
name = "Islamic New Year"
self[date(year, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 9, 11)[0]
y, m, d = to_gregorian(islam_year + 1, 1, 1)
if y == year:
self[date(y, m, d)] = name
# Birthday of Prophet, Mawlid in India
# 12th day of 3rd Islamic month
islam_year = from_gregorian(year, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
name = "Birth of Prophet"
self[date(y, m, d)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 11, 20)[0]
y, m, d = to_gregorian(islam_year, 3, 12)
if y == year:
self[date(y, m, d)] = name
# Eid ul-Fitr
# 1st and 2nd day of 10th Islamic month
islam_year = from_gregorian(year, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
y3, m3, d3 = to_gregorian(islam_year, 10, 3)
name = "Eid al-Fitr"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
self[date(year, m3, d3)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 6, 15)[0]
y1, m1, d1 = to_gregorian(islam_year, 10, 1)
y2, m2, d2 = to_gregorian(islam_year, 10, 2)
y3, m3, d3 = to_gregorian(islam_year, 10, 3)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
if y3 == year:
self[date(y3, m3, d3)] = name
# Eid al-Adha, i.e., Feast of the Sacrifice
islam_year = from_gregorian(year, 8, 22)[0]
y1, m1, d1 = to_gregorian(islam_year, 12, 10)
y2, m2, d2 = to_gregorian(islam_year, 12, 11)
y3, m3, d3 = to_gregorian(islam_year, 12, 12)
y4, m4, d4 = to_gregorian(islam_year, 12, 13)
name = "Feast of the Sacrifice"
self[date(year, m1, d1)] = name
self[date(year, m2, d2)] = name
self[date(year, m3, d3)] = name
self[date(year, m4, d4)] = name
for offset in range(-1, 2, 1):
islam_year = from_gregorian(year + offset, 8, 22)[0]
y1, m1, d1 = to_gregorian(islam_year, 12, 10)
y2, m2, d2 = to_gregorian(islam_year, 12, 11)
y3, m3, d3 = to_gregorian(islam_year, 12, 12)
y4, m4, d4 = to_gregorian(islam_year, 12, 13)
if y1 == year:
self[date(y1, m1, d1)] = name
if y2 == year:
self[date(y2, m2, d2)] = name
if y3 == year:
self[date(y3, m3, d3)] = name
if y4 == year:
self[date(y4, m4, d4)] = name
class EG(Egypt):
@ -1116,7 +1247,10 @@ class China(HolidayBase):
# Chinese New Year/ Spring Festival
name = "Chinese New Year"
self[LunarDate(year, 1, 1).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 1, 1).toSolarDate()
if ds.year == year:
self[ds] = name
# Tomb-Sweeping Day
name = "Tomb-Sweeping Day"
@ -1129,11 +1263,17 @@ class China(HolidayBase):
# Dragon Boat Festival
name = "Dragon Boat Festival"
self[LunarDate(year, 5, 5).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 5, 5).toSolarDate()
if ds.year == year:
self[ds] = name
# Mid-Autumn Festival
name = "Mid-Autumn Festival"
self[LunarDate(year, 8, 15).toSolarDate()] = name
for offset in range(-1, 2, 1):
ds = LunarDate(year + offset, 8, 15).toSolarDate()
if ds.year == year:
self[ds] = name
# National Day
name = "National Day"

5
python/fbprophet/make_holidays.py
View file

@ -25,9 +25,8 @@ def get_holiday_names(country):
country: country name
Returns
------- a
Dataframe with 'ds' and 'holiday', which can directly feed
to 'holidays' params in Prophet
-------
A set of all possible holiday names of given country
"""
years = np.arange(1995, 2045)
try:

3
python/fbprophet/tests/test_diagnostics.py
View file

@ -141,6 +141,7 @@ class TestDiagnostics(TestCase):
# These values are created except for its default values
holiday = pd.DataFrame(
{'ds': pd.to_datetime(['2016-12-25']), 'holiday': ['x']})
append_holidays = 'US'
products = itertools.product(
['linear', 'logistic'], # growth
[None, pd.to_datetime(['2016-12-25'])], # changepoints
@ -150,6 +151,7 @@ class TestDiagnostics(TestCase):
[True, False], # weekly_seasonality
[True, False], # daily_seasonality
[None, holiday], # holidays
[None, append_holidays], # append_holidays
['additive', 'multiplicative'], # seasonality_mode
[1.1], # seasonality_prior_scale
[1.1], # holidays_prior_scale
@ -182,6 +184,7 @@ class TestDiagnostics(TestCase):
self.assertEqual(m1.holidays, m2.holidays)
else:
self.assertTrue((m1.holidays == m2.holidays).values.all())
self.assertEqual(m1.append_holidays, m2.append_holidays)
self.assertEqual(m1.seasonality_mode, m2.seasonality_mode)
self.assertEqual(m1.seasonality_prior_scale, m2.seasonality_prior_scale)
self.assertEqual(m1.changepoint_prior_scale, m2.changepoint_prior_scale)

62
python/scripts/generate_holidays_file.py Normal file
View file

@ -0,0 +1,62 @@
# Copyright (c) 2017-present, Facebook, Inc.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree. An additional grant
# of patent rights can be found in the PATENTS file in the same directory.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals
import pandas as pd
import numpy as np
import warnings
import holidays as hdays_part1
import fbprophet.hdays as hdays_part2
import inspect
def generate_holidays_file():
"""Generate csv file of all possible holiday names, ds,
and countries, year combination
"""
years = np.arange(1995, 2045, 1)
all_holidays = []
# class names in holiday packages which are not countries
class_to_exclude = set(['rd', 'datetime', 'date', 'HolidayBase', 'Calendar',
'LunarDate', 'timedelta', 'date'])
class_list2 = inspect.getmembers(hdays_part2, inspect.isclass)
country_set2 = set(list(zip(*class_list2))[0])
country_set2 -= class_to_exclude
for country in country_set2:
with warnings.catch_warnings():
warnings.simplefilter("ignore")
temp = getattr(hdays_part2, country)(years=years)
temp_df = pd.DataFrame(list(temp.items()),
columns=['ds', 'holiday'])
temp_df['country'] = country
all_holidays.append(temp_df)
class_list1 = inspect.getmembers(hdays_part1, inspect.isclass)
country_set1 = set(list(zip(*class_list1))[0])
country_set1 -= class_to_exclude
# Avoid overwrting holidays get from hdays_part2
country_set1 -= country_set2
for country in country_set1:
temp = getattr(hdays_part1, country)(years=years)
temp_df = pd.DataFrame(list(temp.items()),
columns=['ds', 'holiday'])
temp_df['country'] = country
all_holidays.append(temp_df)
generated_holidays = pd.concat(all_holidays, axis=0, ignore_index=True)
generated_holidays['year'] = generated_holidays.ds.apply(lambda x: x.year)
generated_holidays.to_csv("../R/data-raw/generated_holidays.csv")
if __name__ == "__main__":
# execute only if run as a script
generate_holidays_file()