feat(ocr): add PaddleOCR GPU Docker image and FastAPI OCR server with entrypoint; implement OCR endpoints and consensus extraction testing

test/test.node.ts

@@ -7,7 +7,7 @@ import * as os from 'os';
 const OLLAMA_URL = 'http://localhost:11434';
 const MODEL = 'openbmb/minicpm-v4.5:q8_0';
 
-const BANK_STATEMENT_PROMPT = `You are a bank statement parser. Extract EVERY transaction from the table.
+const EXTRACT_PROMPT = `You are a bank statement parser. Extract EVERY transaction from the table.
 
 Read the Amount column carefully:
 - "- 21,47 €" means DEBIT, output as: -21.47
@@ -16,7 +16,7 @@ Read the Amount column carefully:
 
 For each row output: {"date":"YYYY-MM-DD","counterparty":"NAME","amount":-21.47}
 
-Do not skip any rows. Return complete JSON array:`;
+Do not skip any rows. Return ONLY the JSON array, no explanation.`;
 
 interface ITransaction {
   date: string;
@@ -53,12 +53,12 @@ function convertPdfToImages(pdfPath: string): string[] {
 }
 
 /**
- * Extract transactions from images using Ollama with streaming
+ * Single extraction pass
  */
-async function extractTransactionsStreaming(images: string[]): Promise<ITransaction[]> {
+async function extractOnce(images: string[], passNum: number): Promise<ITransaction[]> {
   const payload = {
     model: MODEL,
-    prompt: BANK_STATEMENT_PROMPT,
+    prompt: EXTRACT_PROMPT,
     images,
     stream: true,
     options: {
@@ -86,7 +86,8 @@ async function extractTransactionsStreaming(images: string[]): Promise<ITransact
   let fullText = '';
   let lineBuffer = '';
 
-  // Stream and print output (buffer until newline for cleaner display)
+  console.log(`[Pass ${passNum}] Extracting...`);
+
   while (true) {
     const { done, value } = await reader.read();
     if (done) break;
@@ -116,13 +117,11 @@ async function extractTransactionsStreaming(images: string[]): Promise<ITransact
     }
   }
 
-  // Print any remaining buffer
   if (lineBuffer) {
     console.log(lineBuffer);
   }
   console.log('');
 
-  // Parse JSON from response
   const startIdx = fullText.indexOf('[');
   const endIdx = fullText.lastIndexOf(']') + 1;
 
@@ -133,6 +132,60 @@ async function extractTransactionsStreaming(images: string[]): Promise<ITransact
   return JSON.parse(fullText.substring(startIdx, endIdx));
 }
 
+/**
+ * Create a hash of transactions for comparison
+ */
+function hashTransactions(transactions: ITransaction[]): string {
+  return transactions
+    .map((t) => `${t.date}|${t.amount.toFixed(2)}`)
+    .sort()
+    .join(';');
+}
+
+/**
+ * Extract with majority voting - run until 2 passes match
+ */
+async function extractWithConsensus(images: string[], maxPasses: number = 5): Promise<ITransaction[]> {
+  const results: Array<{ transactions: ITransaction[]; hash: string }> = [];
+  const hashCounts: Map<string, number> = new Map();
+
+  for (let pass = 1; pass <= maxPasses; pass++) {
+    const transactions = await extractOnce(images, pass);
+    const hash = hashTransactions(transactions);
+
+    results.push({ transactions, hash });
+    hashCounts.set(hash, (hashCounts.get(hash) || 0) + 1);
+
+    console.log(`[Pass ${pass}] Got ${transactions.length} transactions (hash: ${hash.substring(0, 20)}...)`);
+
+    // Check if we have consensus (2+ matching)
+    const count = hashCounts.get(hash)!;
+    if (count >= 2) {
+      console.log(`[Consensus] Reached after ${pass} passes (${count} matching results)`);
+      return transactions;
+    }
+
+    // After 2 passes, if no match yet, continue
+    if (pass >= 2) {
+      console.log(`[Pass ${pass}] No consensus yet, trying again...`);
+    }
+  }
+
+  // No consensus reached - return the most common result
+  let bestHash = '';
+  let bestCount = 0;
+  for (const [hash, count] of hashCounts) {
+    if (count > bestCount) {
+      bestCount = count;
+      bestHash = hash;
+    }
+  }
+
+  const best = results.find((r) => r.hash === bestHash)!;
+  console.log(`[No consensus] Using most common result (${bestCount}/${maxPasses} passes)`);
+  return best.transactions;
+}
+
 /**
  * Compare extracted transactions against expected
  */
@@ -227,16 +280,15 @@ for (const testCase of testCases) {
     // Convert PDF to images
     console.log('Converting PDF to images...');
     const images = convertPdfToImages(testCase.pdfPath);
-    console.log(`Converted: ${images.length} pages`);
+    console.log(`Converted: ${images.length} pages\n`);
 
-    // Extract transactions with streaming output
-    console.log('Extracting transactions (streaming)...\n');
-    const extracted = await extractTransactionsStreaming(images);
-    console.log(`Extracted: ${extracted.length} transactions`);
+    // Extract with consensus voting
+    const extracted = await extractWithConsensus(images);
+    console.log(`\nFinal: ${extracted.length} transactions`);
 
     // Compare results
     const result = compareTransactions(extracted, expected);
-    console.log(`Matches: ${result.matches}/${result.total}`);
+    console.log(`Accuracy: ${result.matches}/${result.total}`);
 
     if (result.errors.length > 0) {
       console.log('Errors:');