StackGenVis: Alignment of Data, Algorithms, and Models for Stacking Ensemble Learning Using Performance Metrics
https://doi.org/10.1109/TVCG.2020.3030352
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
|
4 years ago | |
---|---|---|
.. | ||
.eslintrc.json | 4 years ago | |
.travis.yml | 4 years ago | |
index.js | 4 years ago | |
package.json | 4 years ago | |
readme.md | 4 years ago | |
test.js | 4 years ago |
readme.md
detect-kerning

Calculate kerning pairs for a font.
const kerning = require('detect-kerning')
let pairs = kerning('Roboto')
/*
{
'A”': -10,
'W.': -5,
'P,': -3,
...
}
*/
// get px kerning for 16px font-size
let px = 16 * pairs['AV'] / 1000
pairs = kerning(family|familyList, pairs|range|options?)
Detect kerning pairs for the font family or stack of families and return their kerning in 1000 units/em. Optionally pass specific kerning pairs to check, or a unicode range, by default all printable ASCII character pairs are detected from the [32, 126]
range. Alternatively, an options object can define:
options.pairs
- specific pairs to check;options.range
- unicode range to detect pairs from;options.fontSize
- base font size to use for check. Can affect performance, by default 16.options.threshold
- font size (em) ratio to detect kerning, by default 0.05. Values below that number can bloat kerning table size.
Related
- css-font for parsing font-family from css font string.
License
© 2018 Dmitry Yv. MIT License