mirror of
https://github.com/explosion/spaCy.git
synced 2025-01-12 10:16:27 +03:00
Remove charts from model direcory and add speed benchmarks
With speed benchmarks, charts ended up taking up too much space – and they were mostly data porn and not particularly useful anyways. Instead, we might add a "Compare" page that fetches all models and lets the user compare two or more models in terms of accuracy, speed etc.
This commit is contained in:
ines
parent
47fd254ba7
commit
ae2ad5becc
|
|
@ -14,7 +14,7 @@
|
|||
|
||||
- MODEL_META = public.models._data.MODEL_META
|
||||
- MODEL_LICENSES = public.models._data.MODEL_LICENSES
|
||||
- MODEL_ACCURACY = public.models._data.MODEL_ACCURACY
|
||||
- MODEL_BENCHMARKS = public.models._data.MODEL_BENCHMARKS
|
||||
- EXAMPLE_SENTENCES = public.models._data.EXAMPLE_SENTENCES
|
||||
|
||||
- IS_PAGE = (SECTION != "index") && !landing
|
||||
|
|
|
|||
|
|
@ -26,7 +26,7 @@ for id in CURRENT_MODELS
|
|||
| about this model, see the overview of the
|
||||
| #[+a(gh("spacy-models") + "/releases") latest model releases].
|
||||
|
||||
+table(data-tpl=id data-tpl-key="table")
|
||||
+table.o-block-small(data-tpl=id data-tpl-key="table")
|
||||
+row
|
||||
+cell #[+label Language]
|
||||
+cell #[+tag=comps.lang] #{LANGUAGES[comps.lang]}
|
||||
|
|
@ -56,22 +56,20 @@ for id in CURRENT_MODELS
|
|||
select.o-field__select.u-text-small(data-tpl=id data-tpl-key="compat")
|
||||
.o-empty(data-tpl=id data-tpl-key="compat-versions")
|
||||
|
||||
section(data-tpl=id data-tpl-key="accuracy-wrapper" style="display: none")
|
||||
+grid.o-no-block
|
||||
+grid-col("third")
|
||||
+h(4) Accuracy
|
||||
+table.o-block-small
|
||||
for label, field in MODEL_ACCURACY
|
||||
+row(style="display: none")
|
||||
+cell.u-nowrap
|
||||
+label=label
|
||||
if MODEL_META[field]
|
||||
| #[+help(MODEL_META[field]).u-color-subtle]
|
||||
+cell.u-text-right(data-tpl=id data-tpl-key=field)
|
||||
| n/a
|
||||
|
||||
+grid-col("two-thirds")
|
||||
+h(4) Comparison
|
||||
+chart(id).u-padding-small
|
||||
section(data-tpl=id data-tpl-key="benchmarks" style="display: none")
|
||||
+grid.o-block-small
|
||||
for keys, label in MODEL_BENCHMARKS
|
||||
.u-flex-full.u-padding-small(data-tpl=id data-tpl-key=label.toLowerCase() style="display: none")
|
||||
+table.o-block-small
|
||||
+row("head")
|
||||
+head-cell(colspan="2")=(MODEL_META["benchmark_" + label] || label)
|
||||
for label, field in keys
|
||||
+row(style="display: none")
|
||||
+cell.u-nowrap
|
||||
+label=label
|
||||
if MODEL_META[field]
|
||||
| #[+help(MODEL_META[field]).u-color-subtle]
|
||||
+cell.u-text-right(data-tpl=id data-tpl-key=field)
|
||||
| n/a
|
||||
|
||||
p.u-text-small.u-color-dark(data-tpl=id data-tpl-key="notes")
|
||||
|
|
|
|||
|
|
@ -6,9 +6,6 @@ if quickstart
|
|||
if IS_PAGE
|
||||
script(src="/assets/js/in-view.min.js")
|
||||
|
||||
if HAS_MODELS
|
||||
script(src="/assets/js/chart.min.js")
|
||||
|
||||
if environment == "deploy"
|
||||
script(async src="https://www.google-analytics.com/analytics.js")
|
||||
|
||||
|
|
@ -35,7 +32,7 @@ script
|
|||
| };
|
||||
|
||||
if HAS_MODELS
|
||||
| new ModelLoader('!{MODELS_REPO}', !{JSON.stringify(CURRENT_MODELS)}, !{JSON.stringify(MODEL_LICENSES)}, !{JSON.stringify(MODEL_ACCURACY)});
|
||||
| new ModelLoader('!{MODELS_REPO}', !{JSON.stringify(CURRENT_MODELS)}, !{JSON.stringify(MODEL_LICENSES)}, !{JSON.stringify(MODEL_BENCHMARKS)});
|
||||
|
||||
if environment == "deploy"
|
||||
| window.ga=window.ga||function(){
|
||||
|
|
|
|||
|
|
@ -108,22 +108,12 @@ class ModelLoader {
|
|||
* @param {Object} licenses - License IDs mapped to URLs.
|
||||
* @param {Object} accKeys - Available accuracy keys mapped to display labels.
|
||||
*/
|
||||
constructor(repo, models = [], licenses = {}, accKeys = {}) {
|
||||
constructor(repo, models = [], licenses = {}, benchmarkKeys = {}) {
|
||||
this.url = `https://raw.githubusercontent.com/${repo}/master`;
|
||||
this.repo = `https://github.com/${repo}`;
|
||||
this.modelIds = models;
|
||||
this.licenses = licenses;
|
||||
this.accKeys = accKeys;
|
||||
this.chartColor = '#09a3d5';
|
||||
this.chartOptions = {
|
||||
type: 'bar',
|
||||
options: { responsive: true, scales: {
|
||||
yAxes: [{ label: 'Accuracy', ticks: { suggestedMin: 70 }}],
|
||||
xAxes: [{ barPercentage: 0.425 }]
|
||||
}}
|
||||
}
|
||||
Chart.defaults.global.legend.position = 'bottom';
|
||||
Chart.defaults.global.defaultFontFamily = "-apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Helvetica, Arial, sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol'";
|
||||
this.benchKeys = benchmarkKeys;
|
||||
this.init();
|
||||
}
|
||||
|
||||
|
|
@ -171,7 +161,7 @@ class ModelLoader {
|
|||
/**
|
||||
* Update model details in tables. Currently quite hacky :(
|
||||
*/
|
||||
render({ lang, name, version, sources, pipeline, vectors, url, author, license, accuracy, size, description, notes }) {
|
||||
render({ lang, name, version, sources, pipeline, vectors, url, author, license, accuracy, speed, size, description, notes }) {
|
||||
const modelId = `${lang}_${name}`;
|
||||
const model = `${modelId}-${version}`;
|
||||
const template = new Templater(modelId);
|
||||
|
|
@ -194,11 +184,31 @@ class ModelLoader {
|
|||
if (license) template.fill('license', this.licenses[license] ? getLink(license, this.licenses[license]) : license, true);
|
||||
|
||||
template.get('download').setAttribute('href', `${this.repo}/releases/tag/${model}`);
|
||||
if (accuracy) this.renderAccuracy(template, accuracy, modelId);
|
||||
|
||||
this.renderBenchmarks(template, accuracy, speed);
|
||||
this.renderCompat(template, modelId);
|
||||
template.get('table').removeAttribute('data-loading');
|
||||
}
|
||||
|
||||
renderBenchmarks(template, accuracy = {}, speed = {}) {
|
||||
if (!accuracy && !speed) return;
|
||||
template.get('benchmarks').style.display = 'block';
|
||||
this.renderTable(template, 'parser', accuracy, val => val.toFixed(2));
|
||||
this.renderTable(template, 'ner', accuracy, val => val.toFixed(2));
|
||||
this.renderTable(template, 'speed', speed, Math.round);
|
||||
}
|
||||
|
||||
renderTable(template, id, benchmarks, convertVal = val => val) {
|
||||
if (!this.benchKeys[id] || !Object.keys(this.benchKeys[id]).some(key => benchmarks[key])) return;
|
||||
const keys = Object.keys(this.benchKeys[id]).map(k => benchmarks[k] ? k : false).filter(k => k);
|
||||
template.get(id).style.display = 'block';
|
||||
for (let key of keys) {
|
||||
template
|
||||
.fill(key, this.convertNumber(convertVal(benchmarks[key])))
|
||||
.parentElement.style.display = 'table-row';
|
||||
}
|
||||
}
|
||||
|
||||
renderCompat(template, modelId) {
|
||||
template.get('compat-wrapper').style.display = 'table-row';
|
||||
const options = Object.keys(this.compat).map(v => `<option value="${v}">v${v}</option>`).join('');
|
||||
|
|
@ -211,24 +221,6 @@ class ModelLoader {
|
|||
});
|
||||
}
|
||||
|
||||
renderAccuracy(template, accuracy, modelId, compare=false) {
|
||||
template.get('accuracy-wrapper').style.display = 'block';
|
||||
const metaKeys = Object.keys(this.accKeys).map(k => accuracy[k] ? k : false).filter(k => k);
|
||||
for (let key of metaKeys) {
|
||||
template.fill(key, accuracy[key].toFixed(2)).parentElement.style.display = 'table-row';
|
||||
}
|
||||
|
||||
this.chartOptions.options.legend = { display: compare }
|
||||
new Chart(`chart_${modelId}`, Object.assign({}, this.chartOptions, { data: {
|
||||
datasets: [{
|
||||
label: modelId,
|
||||
data: metaKeys.map(key => accuracy[key].toFixed(2)),
|
||||
backgroundColor: this.chartColor
|
||||
}],
|
||||
labels: metaKeys.map(key => this.accKeys[key])
|
||||
}}))
|
||||
}
|
||||
|
||||
getLatestVersion(model, compat = {}) {
|
||||
for (let spacy_v of Object.keys(compat)) {
|
||||
const models = compat[spacy_v];
|
||||
|
|
|
|||
|
|
@ -50,25 +50,27 @@
|
|||
"ents_f": "Entities (F-score)",
|
||||
"ents_p": "Entities (precision)",
|
||||
"ents_r": "Entities (recall)",
|
||||
"cpu": "words per second on CPU",
|
||||
"gpu": "words per second on GPU",
|
||||
"pipeline": "Processing pipeline components in order",
|
||||
"sources": "Sources of training data"
|
||||
"sources": "Sources of training data",
|
||||
"benchmark_parser": "Parser accuracy",
|
||||
"benchmark_ner": "NER accuracy",
|
||||
"benchmark_speed": "Speed"
|
||||
},
|
||||
|
||||
"MODEL_LICENSES": {
|
||||
"CC BY-SA": "https://creativecommons.org/licenses/by-sa/3.0/",
|
||||
"CC BY-SA": "https://creativecommons.org/licenses/by-sa/3.0/",
|
||||
"CC BY-SA 3.0": "https://creativecommons.org/licenses/by-sa/3.0/",
|
||||
"CC BY-NC": "https://creativecommons.org/licenses/by-nc/3.0/",
|
||||
"CC BY-NC": "https://creativecommons.org/licenses/by-nc/3.0/",
|
||||
"CC BY-NC 3.0": "https://creativecommons.org/licenses/by-nc/3.0/",
|
||||
"GPL": "http://www.gnu.de/documents/gpl.en.html"
|
||||
"GPL": "http://www.gnu.de/documents/gpl.en.html"
|
||||
},
|
||||
|
||||
"MODEL_ACCURACY": {
|
||||
"uas": "UAS",
|
||||
"las": "LAS",
|
||||
"tags_acc": "POS",
|
||||
"ents_f": "NER F",
|
||||
"ents_p": "NER P",
|
||||
"ents_r": "NER R"
|
||||
"MODEL_BENCHMARKS": {
|
||||
"parser": { "uas": "UAS", "las": "LAS", "tags_acc": "POS" },
|
||||
"ner": { "ents_f": "NER F", "ents_p": "NER P", "ents_r": "NER R" },
|
||||
"speed": { "nwords": "Words", "cpu": "w/s CPU", "gpu": "w/s GPU" }
|
||||
},
|
||||
|
||||
"LANGUAGES": {
|
||||
|
|
|
|||
|
|
@ -609,8 +609,8 @@ include _includes/_mixins
|
|||
+code(false, "json").o-no-block "CC BY-SA 3.0": "http://..."
|
||||
|
||||
+row
|
||||
+cell #[code MODEL_ACCURACY]
|
||||
+cell Display labels for accuracy keys.
|
||||
+cell #[code MODEL_BENCHMARKS]
|
||||
+cell Display labels for accuracy and speed.
|
||||
+cell
|
||||
+code(false, "json").o-no-block "ents_f": "NER F"
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user