312 lines
10 KiB
TypeScript
312 lines
10 KiB
TypeScript
/**
|
|
* Invoice extraction using Qwen3-VL 8B Vision (Direct)
|
|
*
|
|
* Single-step pipeline: PDF → Images → Qwen3-VL → JSON
|
|
* Uses /no_think to disable reasoning mode for fast, direct responses.
|
|
*
|
|
* Qwen3-VL outperforms PaddleOCR-VL on certain invoice formats.
|
|
*/
|
|
import { tap, expect } from '@git.zone/tstest/tapbundle';
|
|
import * as fs from 'fs';
|
|
import * as path from 'path';
|
|
import { execSync } from 'child_process';
|
|
import * as os from 'os';
|
|
import { ensureMiniCpm } from './helpers/docker.js';
|
|
|
|
const OLLAMA_URL = 'http://localhost:11434';
|
|
const VISION_MODEL = 'qwen3-vl:8b';
|
|
|
|
interface IInvoice {
|
|
invoice_number: string;
|
|
invoice_date: string;
|
|
vendor_name: string;
|
|
currency: string;
|
|
net_amount: number;
|
|
vat_amount: number;
|
|
total_amount: number;
|
|
}
|
|
|
|
/**
|
|
* Convert PDF to PNG images using ImageMagick
|
|
*/
|
|
function convertPdfToImages(pdfPath: string): string[] {
|
|
const tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'pdf-convert-'));
|
|
const outputPattern = path.join(tempDir, 'page-%d.png');
|
|
|
|
try {
|
|
// 150 DPI is sufficient for invoice extraction, reduces context size
|
|
execSync(
|
|
`convert -density 150 -quality 90 "${pdfPath}" -background white -alpha remove "${outputPattern}"`,
|
|
{ stdio: 'pipe' }
|
|
);
|
|
|
|
const files = fs.readdirSync(tempDir).filter((f) => f.endsWith('.png')).sort();
|
|
const images: string[] = [];
|
|
|
|
for (const file of files) {
|
|
const imagePath = path.join(tempDir, file);
|
|
const imageData = fs.readFileSync(imagePath);
|
|
images.push(imageData.toString('base64'));
|
|
}
|
|
|
|
return images;
|
|
} finally {
|
|
fs.rmSync(tempDir, { recursive: true, force: true });
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Extract invoice data directly from images using Qwen3-VL Vision
|
|
* Uses /no_think to disable reasoning mode for fast, direct JSON output
|
|
*/
|
|
async function extractInvoiceFromImages(images: string[]): Promise<IInvoice> {
|
|
console.log(` [Vision] Processing ${images.length} page(s) with Qwen3-VL`);
|
|
|
|
// /no_think disables Qwen3's reasoning mode - crucial for getting direct output
|
|
const prompt = `/no_think
|
|
Look at this invoice and extract these fields. Reply with ONLY JSON, no explanation.
|
|
|
|
- invoice_number
|
|
- invoice_date (format: YYYY-MM-DD)
|
|
- vendor_name
|
|
- currency (EUR, USD, or GBP)
|
|
- net_amount
|
|
- vat_amount
|
|
- total_amount
|
|
|
|
JSON: {"invoice_number":"...","invoice_date":"YYYY-MM-DD","vendor_name":"...","currency":"EUR","net_amount":0,"vat_amount":0,"total_amount":0}`;
|
|
|
|
const response = await fetch(`${OLLAMA_URL}/api/chat`, {
|
|
method: 'POST',
|
|
headers: { 'Content-Type': 'application/json' },
|
|
body: JSON.stringify({
|
|
model: VISION_MODEL,
|
|
messages: [{
|
|
role: 'user',
|
|
content: prompt,
|
|
images: images, // Pass all pages
|
|
}],
|
|
stream: false,
|
|
options: {
|
|
num_predict: 512,
|
|
temperature: 0.0,
|
|
},
|
|
}),
|
|
});
|
|
|
|
if (!response.ok) {
|
|
const err = await response.text();
|
|
throw new Error(`Ollama API error: ${response.status} - ${err}`);
|
|
}
|
|
|
|
const data = await response.json();
|
|
let content = data.message?.content || '';
|
|
|
|
console.log(` [Vision] Response (${content.length} chars): ${content.substring(0, 200)}...`);
|
|
|
|
// Parse JSON from response
|
|
if (content.startsWith('```json')) content = content.slice(7);
|
|
else if (content.startsWith('```')) content = content.slice(3);
|
|
if (content.endsWith('```')) content = content.slice(0, -3);
|
|
content = content.trim();
|
|
|
|
const startIdx = content.indexOf('{');
|
|
const endIdx = content.lastIndexOf('}') + 1;
|
|
|
|
if (startIdx < 0 || endIdx <= startIdx) {
|
|
throw new Error(`No JSON found: ${content.substring(0, 300)}`);
|
|
}
|
|
|
|
const parsed = JSON.parse(content.substring(startIdx, endIdx));
|
|
|
|
return {
|
|
invoice_number: parsed.invoice_number || null,
|
|
invoice_date: parsed.invoice_date || null,
|
|
vendor_name: parsed.vendor_name || null,
|
|
currency: parsed.currency || 'EUR',
|
|
net_amount: parseFloat(parsed.net_amount) || 0,
|
|
vat_amount: parseFloat(parsed.vat_amount) || 0,
|
|
total_amount: parseFloat(parsed.total_amount) || 0,
|
|
};
|
|
}
|
|
|
|
/**
|
|
* Normalize date to YYYY-MM-DD
|
|
*/
|
|
function normalizeDate(dateStr: string | null): string {
|
|
if (!dateStr) return '';
|
|
if (/^\d{4}-\d{2}-\d{2}$/.test(dateStr)) return dateStr;
|
|
|
|
const monthMap: Record<string, string> = {
|
|
JAN: '01', FEB: '02', MAR: '03', APR: '04', MAY: '05', JUN: '06',
|
|
JUL: '07', AUG: '08', SEP: '09', OCT: '10', NOV: '11', DEC: '12',
|
|
};
|
|
|
|
let match = dateStr.match(/^(\d{1,2})-([A-Z]{3})-(\d{4})$/i);
|
|
if (match) {
|
|
return `${match[3]}-${monthMap[match[2].toUpperCase()] || '01'}-${match[1].padStart(2, '0')}`;
|
|
}
|
|
|
|
match = dateStr.match(/^(\d{1,2})[\/.](\d{1,2})[\/.](\d{4})$/);
|
|
if (match) {
|
|
return `${match[3]}-${match[2].padStart(2, '0')}-${match[1].padStart(2, '0')}`;
|
|
}
|
|
|
|
return dateStr;
|
|
}
|
|
|
|
/**
|
|
* Compare extracted vs expected
|
|
*/
|
|
function compareInvoice(extracted: IInvoice, expected: IInvoice): { match: boolean; errors: string[] } {
|
|
const errors: string[] = [];
|
|
|
|
const extNum = extracted.invoice_number?.replace(/\s/g, '').toLowerCase() || '';
|
|
const expNum = expected.invoice_number?.replace(/\s/g, '').toLowerCase() || '';
|
|
if (extNum !== expNum) {
|
|
errors.push(`invoice_number: expected "${expected.invoice_number}", got "${extracted.invoice_number}"`);
|
|
}
|
|
|
|
if (normalizeDate(extracted.invoice_date) !== normalizeDate(expected.invoice_date)) {
|
|
errors.push(`invoice_date: expected "${expected.invoice_date}", got "${extracted.invoice_date}"`);
|
|
}
|
|
|
|
if (Math.abs(extracted.total_amount - expected.total_amount) > 0.02) {
|
|
errors.push(`total_amount: expected ${expected.total_amount}, got ${extracted.total_amount}`);
|
|
}
|
|
|
|
if (extracted.currency?.toUpperCase() !== expected.currency?.toUpperCase()) {
|
|
errors.push(`currency: expected "${expected.currency}", got "${extracted.currency}"`);
|
|
}
|
|
|
|
return { match: errors.length === 0, errors };
|
|
}
|
|
|
|
/**
|
|
* Find test cases
|
|
*/
|
|
function findTestCases(): Array<{ name: string; pdfPath: string; jsonPath: string }> {
|
|
const testDir = path.join(process.cwd(), '.nogit/invoices');
|
|
if (!fs.existsSync(testDir)) return [];
|
|
|
|
const files = fs.readdirSync(testDir);
|
|
const testCases: Array<{ name: string; pdfPath: string; jsonPath: string }> = [];
|
|
|
|
for (const pdf of files.filter((f) => f.endsWith('.pdf'))) {
|
|
const baseName = pdf.replace('.pdf', '');
|
|
const jsonFile = `${baseName}.json`;
|
|
if (files.includes(jsonFile)) {
|
|
testCases.push({
|
|
name: baseName,
|
|
pdfPath: path.join(testDir, pdf),
|
|
jsonPath: path.join(testDir, jsonFile),
|
|
});
|
|
}
|
|
}
|
|
|
|
return testCases.sort((a, b) => a.name.localeCompare(b.name));
|
|
}
|
|
|
|
/**
|
|
* Ensure Qwen3-VL 8B model is available
|
|
*/
|
|
async function ensureQwen3Vl(): Promise<boolean> {
|
|
try {
|
|
const response = await fetch(`${OLLAMA_URL}/api/tags`);
|
|
if (response.ok) {
|
|
const data = await response.json();
|
|
const models = data.models || [];
|
|
if (models.some((m: { name: string }) => m.name === VISION_MODEL)) {
|
|
console.log(`[Ollama] Model already available: ${VISION_MODEL}`);
|
|
return true;
|
|
}
|
|
}
|
|
} catch {
|
|
console.log('[Ollama] Cannot check models');
|
|
return false;
|
|
}
|
|
|
|
console.log(`[Ollama] Pulling model: ${VISION_MODEL}...`);
|
|
const pullResponse = await fetch(`${OLLAMA_URL}/api/pull`, {
|
|
method: 'POST',
|
|
headers: { 'Content-Type': 'application/json' },
|
|
body: JSON.stringify({ name: VISION_MODEL, stream: false }),
|
|
});
|
|
|
|
return pullResponse.ok;
|
|
}
|
|
|
|
// Tests
|
|
|
|
tap.test('setup: ensure Qwen3-VL is running', async () => {
|
|
console.log('\n[Setup] Checking Qwen3-VL 8B...\n');
|
|
|
|
// Ensure Ollama service is running
|
|
const ollamaOk = await ensureMiniCpm();
|
|
expect(ollamaOk).toBeTrue();
|
|
|
|
// Ensure Qwen3-VL 8B model
|
|
const visionOk = await ensureQwen3Vl();
|
|
expect(visionOk).toBeTrue();
|
|
|
|
console.log('\n[Setup] Ready!\n');
|
|
});
|
|
|
|
const testCases = findTestCases();
|
|
console.log(`\nFound ${testCases.length} invoice test cases (Qwen3-VL Vision)\n`);
|
|
|
|
let passedCount = 0;
|
|
let failedCount = 0;
|
|
const times: number[] = [];
|
|
|
|
for (const testCase of testCases) {
|
|
tap.test(`should extract invoice: ${testCase.name}`, async () => {
|
|
const expected: IInvoice = JSON.parse(fs.readFileSync(testCase.jsonPath, 'utf-8'));
|
|
console.log(`\n=== ${testCase.name} ===`);
|
|
console.log(`Expected: ${expected.invoice_number} | ${expected.invoice_date} | ${expected.total_amount} ${expected.currency}`);
|
|
|
|
const start = Date.now();
|
|
const images = convertPdfToImages(testCase.pdfPath);
|
|
console.log(` Pages: ${images.length}`);
|
|
|
|
const extracted = await extractInvoiceFromImages(images);
|
|
console.log(` Extracted: ${extracted.invoice_number} | ${extracted.invoice_date} | ${extracted.total_amount} ${extracted.currency}`);
|
|
const elapsed = Date.now() - start;
|
|
times.push(elapsed);
|
|
|
|
const result = compareInvoice(extracted, expected);
|
|
|
|
if (result.match) {
|
|
passedCount++;
|
|
console.log(` Result: MATCH (${(elapsed / 1000).toFixed(1)}s)`);
|
|
} else {
|
|
failedCount++;
|
|
console.log(` Result: MISMATCH (${(elapsed / 1000).toFixed(1)}s)`);
|
|
result.errors.forEach((e) => console.log(` - ${e}`));
|
|
}
|
|
|
|
expect(result.match).toBeTrue();
|
|
});
|
|
}
|
|
|
|
tap.test('summary', async () => {
|
|
const total = testCases.length;
|
|
const accuracy = total > 0 ? (passedCount / total) * 100 : 0;
|
|
const totalTime = times.reduce((a, b) => a + b, 0) / 1000;
|
|
const avgTime = times.length > 0 ? totalTime / times.length : 0;
|
|
|
|
console.log(`\n======================================================`);
|
|
console.log(` Invoice Extraction Summary (Qwen3-VL Vision)`);
|
|
console.log(`======================================================`);
|
|
console.log(` Method: Qwen3-VL 8B Direct Vision (/no_think)`);
|
|
console.log(` Passed: ${passedCount}/${total}`);
|
|
console.log(` Failed: ${failedCount}/${total}`);
|
|
console.log(` Accuracy: ${accuracy.toFixed(1)}%`);
|
|
console.log(`------------------------------------------------------`);
|
|
console.log(` Total time: ${totalTime.toFixed(1)}s`);
|
|
console.log(` Avg per inv: ${avgTime.toFixed(1)}s`);
|
|
console.log(`======================================================\n`);
|
|
});
|
|
|
|
export default tap.start();
|