@heripo/pdf-parser 0.1.31 → 0.1.33
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.ko.md +3 -3
- package/README.md +3 -3
- package/dist/index.cjs +1391 -377
- package/dist/index.cjs.map +1 -1
- package/dist/index.d.cts +21 -2
- package/dist/index.d.ts +21 -2
- package/dist/index.js +1391 -377
- package/dist/index.js.map +1 -1
- package/package.json +6 -6
package/README.ko.md
CHANGED
|
@@ -230,7 +230,7 @@ const tokenUsageReport = await pdfParser.parse(
|
|
|
230
230
|
reviewTasks: 4,
|
|
231
231
|
tables: 1,
|
|
232
232
|
},
|
|
233
|
-
|
|
233
|
+
modelConcurrency: 1,
|
|
234
234
|
workItemTimeoutMs: 900000,
|
|
235
235
|
maxRetries: {
|
|
236
236
|
textCorrection: 3,
|
|
@@ -299,7 +299,7 @@ await pdfParser.dispose();
|
|
|
299
299
|
|
|
300
300
|
### 로컬 모델 권장 실행 방식
|
|
301
301
|
|
|
302
|
-
보정 파이프라인은 로컬 VLM 기준으로 설계되었습니다. 큰 context 하나보다 작은 context를 자주 호출하고, deterministic validator와 retry, timeout, checkpoint/resume으로 안정성을 확보합니다. 처음에는 `concurrency.pages: 1`, `concurrency.tables: 1`, `
|
|
302
|
+
보정 파이프라인은 로컬 VLM 기준으로 설계되었습니다. 큰 context 하나보다 작은 context를 자주 호출하고, deterministic validator와 retry, timeout, checkpoint/resume으로 안정성을 확보합니다. 처음에는 `concurrency.pages: 1`, `concurrency.tables: 1`, `modelConcurrency: 1`, `temperature: 0`, 충분한 `workItemTimeoutMs`로 시작하고, 모델이 안정화된 뒤 동시성을 높이는 것을 권장합니다.
|
|
303
303
|
|
|
304
304
|
### 롤아웃 smoke test
|
|
305
305
|
|
|
@@ -679,7 +679,7 @@ interface PDFCorrectionOptions {
|
|
|
679
679
|
reviewAssistance?: number;
|
|
680
680
|
tableCorrection?: number;
|
|
681
681
|
};
|
|
682
|
-
|
|
682
|
+
modelConcurrency?: number; // 로컬 모델 요청 동시성 상한
|
|
683
683
|
workItemTimeoutMs?: number; // work item별 timeout
|
|
684
684
|
outputLanguage?: string; // 사람이 읽는 review reason 언어 (기본값: en-US)
|
|
685
685
|
pageGate?: {
|
package/README.md
CHANGED
|
@@ -230,7 +230,7 @@ const tokenUsageReport = await pdfParser.parse(
|
|
|
230
230
|
reviewTasks: 4,
|
|
231
231
|
tables: 1,
|
|
232
232
|
},
|
|
233
|
-
|
|
233
|
+
modelConcurrency: 1,
|
|
234
234
|
workItemTimeoutMs: 900000,
|
|
235
235
|
maxRetries: {
|
|
236
236
|
textCorrection: 3,
|
|
@@ -299,7 +299,7 @@ Text correction applies to every page with text or table content. The page gate
|
|
|
299
299
|
|
|
300
300
|
### Local Model Execution
|
|
301
301
|
|
|
302
|
-
The correction pipeline is optimized for local VLMs: small contexts, many calls, deterministic validation, retry loops, bounded concurrency, and resumable checkpoints. For local models, start with `concurrency.pages: 1`, `concurrency.tables: 1`, `
|
|
302
|
+
The correction pipeline is optimized for local VLMs: small contexts, many calls, deterministic validation, retry loops, bounded concurrency, and resumable checkpoints. For local models, start with `concurrency.pages: 1`, `concurrency.tables: 1`, `modelConcurrency: 1`, `temperature: 0`, and a generous `workItemTimeoutMs`. Increase concurrency only after the model is stable.
|
|
303
303
|
|
|
304
304
|
### Rollout Smoke Test
|
|
305
305
|
|
|
@@ -679,7 +679,7 @@ interface PDFCorrectionOptions {
|
|
|
679
679
|
reviewAssistance?: number;
|
|
680
680
|
tableCorrection?: number;
|
|
681
681
|
};
|
|
682
|
-
|
|
682
|
+
modelConcurrency?: number; // Bounded local model request concurrency
|
|
683
683
|
workItemTimeoutMs?: number; // Per-work-item timeout
|
|
684
684
|
outputLanguage?: string; // Human-readable review reason language (default: en-US)
|
|
685
685
|
pageGate?: {
|