vacancy-similarity-onnx

iOS Vacancy Similarity Model (ONNX)

Usage with @xenova/transformers

import { pipeline } from '@xenova/transformers';

// Load the model
const extractor = await pipeline(
  'feature-extraction',
  '/Users/timurceberda/Developer/telegram-ios-academy-foundation-pro/apps/miniapp/models/ios-vacancy-similarity-onnx',
  { local_files_only: true }
);

// Get embeddings
const vacancy1 = await extractor('iOS разработчик Swift UIKit');
const vacancy2 = await extractor('Разработчик мобильных приложений iOS');

// Calculate cosine similarity
function cosineSimilarity(a, b) {
  let dot = 0, normA = 0, normB = 0;
  for (let i = 0; i < a.length; i++) {
    dot += a[i] * b[i];
    normA += a[i] * a[i];
    normB += b[i] * b[i];
  }
  return dot / (Math.sqrt(normA) * Math.sqrt(normB));
}

const similarity = cosineSimilarity(vacancy1.data, vacancy2.data);
console.log('Similarity:', similarity);

Model Info

This model was fine-tuned on iOS developer vacancies from HH.ru for improved similarity matching in the iOS vacancy aggregator app.

Base model: sentence-transformers/all-MiniLM-L6-v2 Embedding dimension: 384 Training pairs: 2000+