How to use a dictionary for a large data frame in R?

Question

I read the answers about creating dictionary in r.

equivalent of a python dict in R

Is there a dictionary functionality in R

And I have a question: how could I use this in a large dataset? Data structure is like this:

dput of a subsample is:

structure(list(...1 = c("category 1", NA, NA, NA, "total", "category 2", 
NA, NA, NA, "total"), Items = c("product 1", "product 2", "product 3", 
"product 4", NA, "product 1", "product 2", "product 3", "product 4", 
NA), price = c(1, 2, 3, 4, 10, 3, 4, 5, 6, 18)), row.names = c(NA, 
-10L), class = c("tbl_df", "tbl", "data.frame"))

And I want the result be like:

categoryx: {prodcut1:1, product2:2, product3:3....}

What could I do if the there are 1000 categories and the number of products for each category is different? The answers in above two links, values of each key should be added manually, I don't how to use it for a large dataset.

Or is there other method (except create dictionaries) that could let me extract information of each category easily?

Could someone give ideas about this question? Thanks.

Is it possible to have a result like a dictionary(or list) of dictionaries in python?

such as dict={category1: {prodcut1:1, product2:2, product3:3....}, category2: {prodcut1:3, product2:4, product3:5....} }

So I could know categories's index and use the index to extract information from dict, and maybe it is like such a dataframe:

            item      price

categoryx    product1   2
             product2   3

so I could do operations for specific category?

Answer 1

The first column name started with ..., so it was renamed to 'grp', then use fill from tidyr to replace the NA elements with the previous non-NA element, filter out the rows where the 'Items' are NA, unite the columns 'Items', 'price' to a single column by concatenating with sep as ":", grouped by 'grp', we summarise the 'ItemsPrice' by creating a single string with str_c

library(dplyr)
library(tidyr)
library(stringr)
df1 %>% 
   rename(grp = `...1`) %>% 
   fill(grp) %>%
   filter(!is.na(Items)) %>% 
   unite(ItemsPrice, Items, price, sep=":") %>%
   group_by(grp) %>%
   summarise(ItemsPrice = str_c(ItemsPrice, collapse = ", "))

-output

# A tibble: 2 x 2
#  grp        ItemsPrice                                        
#  <chr>      <chr>                                             
#1 category 1 product 1:1, product 2:2, product 3:3, product 4:4
#2 category 2 product 1:3, product 2:4, product 3:5, product 4:6

Answer 2

A list of hashmap dictionaries:

dat <-
  structure(
    list(
      ...1 = c("category 1", NA, NA, NA, "total", "category 2",
               NA, NA, NA, "total"),
      Items = c(
        "product 1",
        "product 2",
        "product 3",
        "product 4",
        NA,
        "product 1",
        "product 2",
        "product 3",
        "product 4",
        NA
      ),
      price = c(1, 2, 3, 4, 10, 3, 4, 5, 6, 18)
    ),
    row.names = c(NA,-10L),
    class = c("tbl_df", "tbl", "data.frame")
  )

library(hashmap)

dat_clean <- tidyr::fill(dat[!is.na(dat[["Items"]]), ], 1)

list_of_dicts <- lapply(split(dat_clean, dat_clean[[1]]), function(d){
  hashmap(d[["Items"]], d[["price"]])  
})

list_of_dicts
# $`category 1`
# ## (character) => (numeric)  
# ## [product 1] => [+1.000000]
# ## [product 3] => [+3.000000]
# ## [product 4] => [+4.000000]
# ## [product 2] => [+2.000000]
# 
# $`category 2`
# ## (character) => (numeric)  
# ## [product 1] => [+3.000000]
# ## [product 3] => [+5.000000]
# ## [product 4] => [+6.000000]
# ## [product 2] => [+4.000000]


# get totals:
lapply(list_of_dicts, function(dict){
  sum(dict$values())
})
# $`category 1`
# [1] 10
# 
# $`category 2`
# [1] 18

Answer 3

You can use zoo::na.locf to fill the category values

names(df)[1] <- 'category'
df$category <- zoo::na.locf(df$category)
df <- subset(df, category != 'total')
df

# A tibble: 8 x 3
#  category   Items     price
#  <chr>      <chr>     <dbl>
#1 category 1 product 1     1
#2 category 1 product 2     2
#3 category 1 product 3     3
#4 category 1 product 4     4
#5 category 2 product 1     3
#6 category 2 product 2     4
#7 category 2 product 3     5
#8 category 2 product 4     6

I would keep the data as above in long format since all the libraries and base R allow grouped operations. So you can calculate anything for each category. I don't see any benefit of complicating the structure beyond this.

If you want the data to be in separate dataframes we can use split.

list_df <- split(df[-1], df$category)

Now you can have access to each individual category in separate dataframe. For example, to get data for category 1 you can do :

list_df$`category 1`

# A tibble: 4 x 2
#  Items     price
#  <chr>     <dbl>
#1 product 1     1
#2 product 2     2
#3 product 3     3
#4 product 4     4