Registering

In order to use Kaggle as a board, you need to authenticate first by creating a token file from kaggle.com/me/account.

board_register_kaggle(token = "path/to/kaggle.json")

Notice that board_register_kaggle() is just an alias with named parameters to board_register(); the previous code is equivalent to:

board_register("kaggle", token = "path/to/kaggle.json")

Once the board is registered, you can pin and search using pin(), pin_get() and pin_find().

Pinning

Like in other boards, you can create pins for iris and mtcars by setting board to Kaggle’s board name,

pin(iris, description = "The iris data set", board = "kaggle")
pin(mtcars, description = "The motor trend cars data set", board = "kaggle")

After a pin is created, the pin also becomes available in the Kaggle’s dataset website; by default, they are created as private datasets.

You can also retrieve pins back from this repo using the now familiar pin_get() function. However, since Kaggle names require at least six characters, pins appends -pin to names that are shorter than Kaggle’s required size.

pin_get("iris-pin", board = "kaggle")
# A tibble: 150 x 5
   Sepal.Length Sepal.Width Petal.Length Petal.Width Species
          <dbl>       <dbl>        <dbl>       <dbl> <fct>  
 1          5.1         3.5          1.4         0.2 setosa 
 2          4.9         3            1.4         0.2 setosa 
 3          4.7         3.2          1.3         0.2 setosa 
 4          4.6         3.1          1.5         0.2 setosa 
 5          5           3.6          1.4         0.2 setosa 
 6          5.4         3.9          1.7         0.4 setosa 
 7          4.6         3.4          1.4         0.3 setosa 
 8          5           3.4          1.5         0.2 setosa 
 9          4.4         2.9          1.4         0.2 setosa 
10          4.9         3.1          1.5         0.1 setosa 
# … with 140 more rows

Discovering

You can then search pins in Kaggle using pin_find(), which by default search all boards but you can also explicitly request to use this particular Kaggle board:

pin_find("motor", board = "kaggle")
# A tibble: 11 x 4
   name                                         description                          type  board
   <chr>                                        <chr>                                <chr> <chr>
 1 javierluraschi/mtcars                        The motor trend cars data set        files kagg…
 2 wkirgsn/electric-motor-temperature           Electric Motor Temperature           files kagg…
 3 new-york-city/nypd-motor-vehicle-collisions  NYPD Motor Vehicle Collisions        files kagg…
 4 rodrigodomingos/brazilian-insurance-motor-m… Brazilian Motor Insurance Market     files kagg…
 5 dan195/ebaymotorcycles                       Ebay Motorcycle Prices               files kagg…
 6 new-york-state/nys-motor-vehicle-crashes-an… NYS Motor Vehicle Crashes and Insur… files kagg…
 7 floser/french-motor-claims-datasets-fremtpl… French Motor Claims Datasets freMTP… files kagg…
 8 floser/swedish-motor-insurance               Swedish Motor Insurance              files kagg…
 9 kkairu/motorbike-ride-cancellation-data      MotorBike Ride Cancellation Data     files kagg…
10 shreedevi/motor-insurance                    Motor Insurance                      files kagg…
11 cdc/cdc-motor-vehicle-occupant-death-rate    CDC Motor Vehicle Occupant Death Ra… files kagg…

Kaggle boards contain extended fields that go beyond the fields pins requires, to retrieve all the additional fields use extended = TRUE:

pin_find("motor", board = "kaggle", extended = TRUE)
# A tibble: 8 x 27
      id subtitle tags  creatorName creatorUrl totalBytes url   lastUpdated downloadCount
   <int> <chr>    <lis> <chr>       <chr>           <dbl> <chr> <chr>               <int>
1 236410 140 hrs… <df[… Kirgsn      wkirgsn      49334849 http… 2019-06-19…          6785
2  31447 From Ne… <df[… Kaggle Team kaggleteam   71098646 http… 2019-11-12…           638
3    605 An intr… <df[… Rodrigo Do… rodrigodo…     374779 http… 2017-01-11…           615
4  34387 From Ne… <df[… Kaggle Team kaggleteam          0 http… 2019-07-03…           479
5  12630 ""       <df[… floser      floser          20476 http… 2018-02-10…           427
6   2182 Ebay Mo… <df[… Dan         dan195         262517 http… 2017-08-23…           701
7 118991 Predict… <df[… floser      floser        6955329 http… 2019-02-11…           212
8 158751 ""       <df[… donpiano    donpiano       226652 http… 2019-04-07…           111
# … with 18 more variables: isPrivate <lgl>, isReviewed <lgl>, isFeatured <lgl>,
#   licenseName <chr>, description <chr>, ownerName <chr>, ownerRef <chr>, kernelCount <int>,
#   topicCount <int>, viewCount <int>, voteCount <int>, currentVersionNumber <int>,
#   files <list>, versions <list>, usabilityRating <dbl>, name <chr>, board <chr>, type <chr>

Or pin_info() to retrieve all the information associated with a particular pin,

pin_info("electric-motor-temperature", board = "kaggle")
# Source: kaggle<wkirgsn/electric-motor-temperature> [files]
# Description: Electric Motor Temperature
# Extended:
#   - id: 236410
#   - subtitle: 140 hrs recordings from a permanent magnet synchronous motor (PMSM)
#   - tags: (ref) time series, electrical components, electrical engineering, automobiles, regression...
#   - creatorName: Kirgsn
#   - creatorUrl: wkirgsn
#   - totalBytes: 49334849
#   - url: https://www.kaggle.com/wkirgsn/electric-motor-temperature
#   - lastUpdated: 2019-06-19T11:12:41.147Z
#   - downloadCount: 6785
#   - isPrivate: FALSE
#   - isReviewed: TRUE
#   - isFeatured: FALSE
#   - licenseName: Data files © Original Authors
#   - ownerName: Kirgsn
#   - ownerRef: wkirgsn
#   - kernelCount: 24
#   - topicCount: 1
#   - viewCount: 40705
#   - voteCount: 190
#   - currentVersionNumber: 2
#   - usabilityRating: 1
#   - extension: zip

Notice that the given ‘motor’ search keyword is searched in the name and description fields for the pins available in this repo.

Competitions

Datasets are also searched within competitions and are referenced by the c/ prefix. For instance, when searching for crowdflower notice the c/crowdflower-weather-twitter entry:

pin_find("crowdflower", board = "kaggle")
# A tibble: 10 x 4
   name                                                    description                                     type  board 
   <chr>                                                   <chr>                                           <chr> <chr> 
 1 awsaf49/ecommerce-search-result-relevane-by-crowdflower eCommerce Search Result Relevane by Crowdflower files kaggle
 2 c/crowdflower-search-relevance                          Crowdflower Search Results Relevance            files kaggle
 3 c/crowdflower-weather-twitter                           Partly Sunny with a Chance of Hashtags          files kaggle
 4 crowdflower/first-gop-debate-twitter-sentiment          First GOP Debate Twitter Sentiment              files kaggle
 5 crowdflower/handwritten-names                           Handwritten Names                               files kaggle
 6 crowdflower/narrativity-in-scientific-publishing        Narrativity in Scientific Publishing            files kaggle
 7 crowdflower/political-social-media-posts                Political Social Media Posts                    files kaggle
 8 crowdflower/twitter-airline-sentiment                   Twitter US Airline Sentiment                    files kaggle
 9 crowdflower/twitter-user-gender-classification          Twitter User Gender Classification              files kaggle
10 humancomp/worker-activity-crowdflower                   Workers Browser Activity in CrowdFlower Tasks   files kaggle

You can then download the data associated to the c/crowdflower-weather-twitter competition:

pin_get("c/crowdflower-weather-twitter", board = "kaggle")
[1] "/Users/javierluraschi/Library/Caches/pins/kaggle/crowdflower-weather-twitter/sampleSubmission.csv"
[2] "/Users/javierluraschi/Library/Caches/pins/kaggle/crowdflower-weather-twitter/test.csv"            
[3] "/Users/javierluraschi/Library/Caches/pins/kaggle/crowdflower-weather-twitter/train.csv"           
[4] "/Users/javierluraschi/Library/Caches/pins/kaggle/crowdflower-weather-twitter/variableNames.txt"   

Sharing

When working in teams, you might want to share your pins with others. For Kaggle boards, this is accomplished by adding users or making the dataset public through Kaggle’s website.

Once you share with specific users, they can follow the same steps to register a Kaggle board which allows them to download pins shared with them or upload their own pins.