-
-
Save harthur/8bda4439b76e7ca19587 to your computer and use it in GitHub Desktop.
Neural network indentation detection
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
var brain = require("brain"); | |
// initalize network from previously trained state | |
var trained = {}; // Too big to display! see https://gist.github.com/harthur/3d5c24a04e509284940d | |
var network = new brain.NeuralNetwork().fromJSON(trained); | |
function detectIndent(lines) { | |
// get widths used in file and what % of file | |
var percentages = getWidthPercentages(lines); | |
var results = network.run(percentages); | |
// from results, pick indent with the highest signal | |
var max = 0,indent = null; | |
for (var width in results) { | |
var prob = results[width]; | |
if (prob > max) { | |
indent = width; | |
max = prob; | |
} | |
} | |
return indent; | |
} | |
/** | |
* Get indent widths used, and the % of lines using each. | |
* Return ex: {2: 0.33, 4: 0.51, 6: 0.13, 8: 0.03, 17: 0.01} | |
*/ | |
function getWidthPercentages(lines) { | |
var counts = {}; // # spaces -> # lines with that width | |
var total = 0; // # of indented lines | |
lines.forEach(function (text) { | |
var width = leadingSpaces(text); | |
if (width > 0) { | |
counts[width] = (counts[width] || 0) + 1; | |
total++; | |
} | |
}); | |
// turn counts into % of lines | |
for (var width in counts) { | |
counts[width] = counts[width] / total; | |
} | |
return counts; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment