javascript/makemeahanzi
1
0
Fork 0
mirror of https://github.com/skishore/makemeahanzi.git synced 2025-11-01 20:27:44 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update classifier to included trained neural net

Browse Source
This commit is contained in:
Shaunak Kishore
2015-09-13 00:35:41 -04:00
parent 515014d39a
commit 9f0a69554c
4 changed files with 22 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
lib/classifier.js Normal file
View File

@ -0,0 +1,17 @@
Meteor.startup(function() {
var weight = 0.8;
var dimensions = 8;
var input = new convnetjs.Vol(1, 1, dimensions);
var net = new convnetjs.Net();
net.fromJSON(TRAINED_NEURAL_NET);
function net_classifier(features) {
input.w = features;
var softmax = net.forward(input).w;
return softmax[1] - softmax[0];
}
this.combined_classifier = function(features) {
return hand_tuned_classifier(features) + weight*net_classifier(features);
}
});

1
lib/net.js Normal file
View File

File diff suppressed because one or more lines are too long

2
lib/stroke_extractor.js
View File

@ -503,7 +503,7 @@ this.get_glyph_render_data = function(glyph, manual_bridges, classifier) {
}
}
var log = [];
var bridges = get_bridges(endpoints, classifier || hand_tuned_classifier);
var bridges = get_bridges(endpoints, classifier || combined_classifier);
var strokes = extract_strokes(
paths, endpoints, manual_bridges || bridges, log);
var expected = UNIHAN_STROKE_COUNTS[glyph.name];

6
server/training.js
View File

@ -8,7 +8,7 @@ function evaluate(glyphs, classifier) {
return num_correct/glyphs.length;
}
Meteor.startup(function() {
function train_neural_net() {
var glyphs = Glyphs.find({'manual.verified': true}).fetch();
var sample = _.sample(glyphs, 400);
console.log('Hand-tuned accuracy:', evaluate(sample, hand_tuned_classifier));
@ -52,7 +52,7 @@ Meteor.startup(function() {
}
console.log('Iteration', iteration, 'mean loss:', loss/round_data.length);
}
console.log('Trained neural network.');
console.log('Trained neural network:', JSON.stringify(net.toJSON()));
function net_classifier(features) {
assert(input.w.length === features.length);
@ -73,4 +73,4 @@ Meteor.startup(function() {
console.log('Weight', weights[i], 'combined accuracy:',
evaluate(sample, combined_classifier(weights[i])));
}
});
}