If you've landed on this page, it likely means you've visited our booth at Pycon 2017! 


Chicory is a food-tech startup that makes digital grocery shopping solutions. Our first product, the “Get Ingredients” button, allows consumers to add ingredients from recipes to online grocery carts at various retailers. But, because there is no standard for how to write out ingredients in recipes, mapping ingredient lines to products sold at the retailer becomes a challenge.

The Challenge: Beat Dolores.

We want to see if you can beat Dolores. You will be given a dataset of recipe lines and a standard set of ingredients. Your job is to write an algorithm that can accurately map the two datasets. The person whose submission has the highest accuracy will win a $150 Amazon gift card.  


  1. Download the challenge dataset here: Chicory Challenge

  2. Inside the archive you will find a list of raw recipe lines and a list of target ingredients.

  3. Create an algorithm that maps the recipe lines to the target ingredients.

  4. Post to the Chicory API at http://challenge.chicory.co/api/submission/ to get real-time feedback and a % accuracy of your data set.

  5. Tweak your algorithm

  6. Resubmit!

Each submission must include your full name, email address and the data that you want to validate. The data that you are posting should be a list of objects. Each object should contain the text input and the parsed ingredient. Use the API Token PYCON2017 when validating your submissions.

Below is an example of what validated data should look like:

  "fullname": "Emil Diaz",
  "email_address": "emil@chicory.co",
  "data": [
      "text": "1 pound cheddar cheese",
      "ingredient": "cheddar cheese"
      "text": "4 lbs beef roast (chuck, sirloin or rump)",
      "ingredient": "beef roasts"
      "text": "1 bag 26 oz. frozen hash browns",
      "ingredient": "hash browns"

To submit your data one can use curl as follows:

curl -X POST 'http://challenge.chicory.co/api/submission/' -H 'Authorization: Token PYCON2017' -H 'Content-Type: application/json' -d @your-json-file.json

1 Comment