Open Datasets used in DataHack Premier League

opensource
dpl

#1

Please use this thread for contributing any datasets you found useful for building models for DataHack Premier League.

Some food for thought - match data for various venues, weather data for venue etc.

Looking forward to seeing the datasets.


#2

Just a question. How to procure the various data? Is there any source from where we can download/ procure the same? Or manually prepare data in a spreadsheet after getting the scorecards from websites like Cricinfo/ Cricbuzz?


#3

The data is available in the site as Train data, Test data and sample data. You can download from there.


#4

Can we prepare the dataset based on the information available over the internet and use the same to make the prediction?


#5

You can use outside dataset for modelling but it must be included in the submission.


#6

Is there a slack channel for DPL also?


#7

Hi,
Is the data shared with us or we have to take it for internet.


#8

Yes just search for #datahackpremierleague on the AV Slack channel.


#9

@getvishalsingh - you can get the base data from https://datahack.analyticsvidhya.com/contest/datahack-premier-league/

You are free to use more data. But, if you do, make sure to include it in your code files, so that we can verify your codes.

Regards,
Kunal


#11

The following datasets will help:

  1. Playing XI of historical IPL games
  2. Weather data - historical and future forecast
  3. Player classification - Batsman, Bowler, All-rounder, Wicket Keeper, Captain, Vice-captain

Does anyone know any site from where we can download these data?


#12

I scraped this data for a friend, I think you may find this useful.

It’s the IPL Auction Data 2018 it includes fields like

sold_to(team name), player, role(batsman/bowler/wk etc), final price, sold_or_not, base price

Hope you find this useful,

Sanad :slight_smile:cricbuzz_ipl.csv (20 KB)


#13

@arunsankar,

I think this site can help you - http://www.cricbuzz.com/

Regards,
Sanad :slight_smile:


#14

Thanks for sharing @mohdsanadzakirizvi It would be awesome if we can get this data for all season. I think the auction price is a very important feature as it is correlated directly to player’s current performance and experience in international cricket.