ingredients-parse-crf
v1.0.0
Published
NodeJS port of NYTimes CRF Ingredient Phrase Tagger
Downloads
3
Readme
CRF Ingredient Phrase Tagger
This is a fork of SimonBoorer/node-ingredient-phrase-tagger but without the unnecessary bloat, code formatting tools, and vulnerable packages. This is intended to be published as a nodejs module that can be imported.
NodeJS port of NYTimes CRF Ingredient Phrase Tagger.
This repo contains scripts to extract the Quantity, Unit, Name, and Comments from unstructured ingredient phrases. Given the following input:
1 pound carrots, young ones if possible
Kosher salt, to taste
2 tablespoons sherry vinegar
2 tablespoons honey
2 tablespoons extra-virgin olive oil
1 medium-size shallot, peeled and finely diced
1/2 teaspoon fresh thyme leaves, finely chopped
Black pepper, to taste
This tool produces something like:
[
{
"qty": "1",
"unit": "pound",
"name": "carrots",
"comment": ", young ones if possible",
"display": "<span class='qty'>1<\/span><span class='unit'>pound<\/span><span class='name'>carrots<\/span><span class='comment'>, young ones if possible<\/span>",
"input": "1 pound carrots, young ones if possible"
},
{
"comment": "Kosher to taste",
"name": "salt",
"other": ",",
"display": "<span class='comment'>Kosher<\/span><span class='name'>salt<\/span><span class='other'>,<\/span><span class='comment'>to taste<\/span>",
"input": "Kosher salt, to taste"
},
{
"qty": "2",
"unit": "tablespoon",
"name": "sherry vinegar",
"display": "<span class='qty'>2<\/span><span class='unit'>tablespoons<\/span><span class='name'>sherry vinegar<\/span>",
"input": "2 tablespoons sherry vinegar"
},
{
"qty": "2",
"unit": "tablespoon",
"name": "honey",
"display": "<span class='qty'>2<\/span><span class='unit'>tablespoons<\/span><span class='name'>honey<\/span>",
"input": "2 tablespoons honey"
},
{
"qty": "2",
"unit": "tablespoon",
"comment": "extra-virgin",
"name": "olive oil",
"display": "<span class='qty'>2<\/span><span class='unit'>tablespoons<\/span><span class='comment'>extra-virgin<\/span><span class='name'>olive oil<\/span>",
"input": "2 tablespoons extra-virgin olive oil"
},
{
"qty": "1",
"comment": "medium-size, peeled and finely diced",
"name": "shallot",
"display": "<span class='qty'>1<\/span><span class='comment'>medium-size<\/span><span class='name'>shallot<\/span><span class='comment'>, peeled and finely diced<\/span>",
"input": "1 medium-size shallot, peeled and finely diced"
},
{
"qty": "1\/2",
"unit": "teaspoon",
"comment": "fresh, finely chopped",
"name": "thyme leaves",
"display": "<span class='qty'>1\/2<\/span><span class='unit'>teaspoon<\/span><span class='comment'>fresh<\/span><span class='name'>thyme leaves<\/span><span class='comment'>, finely chopped<\/span>",
"input": "1\/2 teaspoon fresh thyme leaves, finely chopped"
},
{
"name": "Black pepper",
"other": ",",
"comment": "to taste",
"display": "<span class='name'>Black pepper<\/span><span class='other'>,<\/span><span class='comment'>to taste<\/span>",
"input": "Black pepper, to taste"
}
]
Usage
Training
import path from "path";
import { Trainer } from "ingredient-phrase-tagger";
const trainer = new Trainer();
const data_path = path.join(__dirname, "nyt-ingredients-snapshot-2015.csv");
const model_filename = path.join(__dirname, "model.crfsuite");
trainer.append(data_path, 20000, 0).then(() => {
trainer.train(model_filename);
});
// output: ./model.crfsuite
Testing
import path from "path";
import { Tagger } from "ingredient-phrase-tagger";
const tagger = new Tagger();
const model_filename = path.join(__dirname, "model.crfsuite");
const is_opened = tagger.open(model_filename);
console.log("File model is opened:", is_opened);
const input = [
"1 pound carrots, young ones if possible",
"Kosher salt, to taste",
"2 tablespoons sherry vinegar",
"2 tablespoons honey",
"2 tablespoons extra-virgin olive oil",
"1 medium-size shallot, peeled and finely diced",
"1/2 teaspoon fresh thyme leaves, finely chopped",
"Black pepper, to taste"
];
const data = tagger.tag(input);