Package 'm5'

Classify time series of the particular items

Description

Each time series in the dataset can be assigned one of the following classes:

Usage

m5_demand_type(data)

Arguments

data	The result of the m5_prepare function; tiny_m5 can be passed as well.

Details

• Smooth (ADI < 1.32 and CV² < 0.49).

• Intermittent (ADI >= 1.32 and CV² < 0.49)

• Erratic (ADI < 1.32 and CV² >= 0.49)

• Lumpy (ADI >= 1.32 and CV² >= 0.49)

Value

A data.table containing item ids (item_id and store_id), ADI and CV2 scores (adi and cv2 respectively) as well as the final class chosen based on the aforementioned scores (demand_type).

References

Syntetos A. A. and Boylan J. E., 2005, The accuracy of intermittent demand estimates. International Journal of Forecasting 21: 303–314 Forecast Error Measures: Intermittent Demand

Examples

head(m5_demand_type(tiny_m5))

---

Download and unzip the raw data to the specified directory

Description

Download and unzip the raw data to the specified directory

Usage

m5_download(path, unzip = TRUE)

Arguments

path	A directory name to save the zip file
unzip	Automatically uznip the file when the downloading is finished. Default: TRUE. The exdir argument in the unzip function is the directory name the file was downloaded into.

Value

Returns nothing; the result of the function call is the m5.zip file downloading and extraction.

Note

If you struggle with timeout problems, please increase the timeout value using ⁠options(timeout=<new_timeout_value>)⁠

References

• M5 Forecasting - Accuracy

• M5 Forecasting - Uncertainty

Examples

## Not run: 
m5_download('data')

## End(Not run)

---

Load raw CSV files using data.table::fread() function

Description

Load raw CSV files using data.table::fread() function

Usage

m5_get_raw_evaluation(path)

m5_get_raw_validation(path)

Arguments

path	The directory with the unzipped M5 data files

Value

The function returns a list of five data.tables:

• sales_train (evaluation/validation)

• sales_test (evaluation/validation)

• sell_prices

• calendar

• weights (evaluation/validation)

References

m5-forecasts repo by Nixtla

Examples

## Not run: 
library(m5)
library(zeallot)

m5_download('data')
c(sales_train,
  sales_test,
  sell_prices,
  calendar,
  ) %<-% m5_get_raw_evaluation('data')

## End(Not run)

---

Prepare the ready-to-use M5 data in one data.frame

Description

It's a memory-efficient function, which uses data.table under the hood. However, it still not recommended to use this function on PCs with < 16GB RAM. In such case, consider to use a custom solution based on ⁠[arrow](https://arrow.apache.org/docs/r/)⁠ or ⁠[disk.frame](https://diskframe.com/index.html)⁠

Usage

m5_prepare(sales_train, sales_test, calendar, sell_prices)

Arguments

sales_train	A data.frame with M5 train data
sales_test	A data.frame with M5 test data
calendar	A data.frame with M5 calendar
sell_prices	A data.frame with M5 sell_prices

Value

A data.table composed from input objects, which contains the following columns:

• item_id

• dept_id

• cat_id

• store_id

• state_id

• d - day ordinal number

• value - number of sold items

• wm_yr_wk - week identifier

• weekday - weekday name (character)

• wday - weekday as an integer

• month

• year

• event_name_1 - special event name, like holidays etc.

• event_type_1 - special event type

• event_name_2 - as above

• event_type_2 - as above

• snap - promotion flag

• sell_price

Examples

library(m5)
library(zeallot)
## Not run: 

m5_download('data')
c(sales_train,
  sales_test,
  sell_prices,
  calendar,
  weights) %<-% m5_get_raw_evaluation('data')

m5_data  <-
   m5_prepare(sales_train, sales_test, calendar, sell_prices)

## End(Not run)

---

A subset from M5 Walmart Challenge Dataset in one data frame

Description

A piece of data cut from the training dataset used in the M5 challenges on Kaggle. M5 is a challenge from a series organized by Spyros Makridakis.

Usage

tiny_m5

Format

item_id

The id of the product

dept_id

The id of the department the product belongs to

cat_id

The id of the category the product belongs to

store_id

The id of the store where the product is sold

state_id

The State where the store is located

value

The number of sold units

date

The date in a “y-m-d” format

wm_yr_wk

The id of the week the date belongs to

weekday

The type of the day (Saturday, Sunday, …, Friday)

wday

The id of the weekday, starting from Saturday

month

The month of the date

year

The year of the date

event_name_1

If the date includes an event, the name of this event

event_type_1

If the date includes an event, the type of this event

event_name_2

If the date includes a second event, the name of this event

event_type_2

If the date includes a second event, the type of this event

snap

A binary variable (0 or 1) indicating whether the stores of CA, TX or WI allow SNAP1 purchases on the examined date. 1 indicates that SNAP purchases are allowed

sell_price

The price of the product for the given week/store. The price is provided per week (average across seven days). If not available, this means that the product was not sold during the examined week. Note that although prices are constant at weekly basis, they may change through time (both training and test set)

See Also

M5 Forecasting - Accuracy

M5 Forecasting - Uncertainty

The M5 competition: Background, organization, and implementation

Other Walmart datasets in timetk

Examples

library(m5)
# Head of tiny_m5
head(tiny_m5)

---