How does one extract data so it emulates a typical Nutrition Label?

[yosun]

Playing around with some queries, but not versed enough in Nutrition to figure out how to structure it so it looks like a typical nutrition table (seems that ignoring certain nutrients in the listing is a step - lots of things labeled as 4:0). Maybe you can help?

SELECT nutrient.name, nutrition.amount FROM nutrition INNER JOIN food ON nutrition.food_id=food.id INNER JOIN nutrient ON nutrition.nutrient_id=nutrient.id WHERE food.id = ( SELECT food.id FROM food WHERE food.short_desc LIKE '%banana%' LIMIT 1 ) AND nutrition.amount>0 AND nutrient.id<607

https://www.fda.gov/food/guidanceregulation/guidancedocumentsregulatoryinformation/labelingnutrition/ucm063113.htm

[pierobot]

You don't have to worry about the nutrients that show as 4:0 and such.

The link you provided shows you what nutrients you need. Just find the nutrient ID for those and add them to your query.

For example, to construct the Fat section of a label

• 204 Total lipid (fat)
  • 605 Total trans-unsaturated
  • 606 Total saturated
  • 645 Total monounsaturated
  • 646 Total polyunsaturated

You also want to add the nutrient units to your query.

You will want to read SR 28 if you have not already. https://github.com/alyssaq/usda-sqlite/blob/master/data/sr28_doc.pdf

[BradKML]

I had similar ideas with https://github.com/schirinos/nutrient-db/issues/13 but they used a JSON generator that does not work on Python 3 currently

Since currently I would need a CSV-like data table for certain Data competitions.

And the other table that can be found is one with a naming scheme that is not in long form with https://github.com/buswedg/dataquest/blob/master/Data%20Analysis%20with%20Pandas/Dataframe%20Basics/data/food_info.csv Without using https://raw.githubusercontent.com/spbriggs/USDA-SR26/master/sr26/FOOD_DES.csv