# OCR 모델 make-model 명령어 오류

**URL:** <https://forums.furiosa.ai/t/ocr-make-model/311>\
**Category:** 한국어 (Korean)\
**Tags:** warboy\
**Created:** [July 25, 2025, 8:34am UTC](https://forums.furiosa.ai/t/ocr-make-model/311 "2025-07-25T08:34:42Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![keh](https://avatars.discourse-cdn.com/v4/letter/k/3bc359/32.png) [@keh](https://forums.furiosa.ai/u/keh)\
**Post date:** [July 25, 2025, 8:34am UTC](https://forums.furiosa.ai/t/ocr-make-model/311/1 "2025-07-25T08:34:42Z")

</div>

자체 개발한 OCR검출기 모델 onnx모델 변환 및 양자화를 진행하고 싶은데 warboy-vision make-model --config\_file “./ocr\_det.yaml” 명령어를 실행시키니 아래와 같은 에러가 발생합니다.

(warboy) clapi@sysadmin-B7105F48TV4HR-2T-N:~/warboy-vision-models/src/warboy/cfg/model\_config/object\_detection$ warboy-vision make-model --config\_file “./ocr\_det.yaml”  
Warning: The model is not a YOLO model. The need\_edit option is ignored.  
Load PyTorch Model from ./models/freeze\_detector.pt…  
Traceback (most recent call last):  
File “/home/clapi/miniconda3/envs/warboy/bin/warboy-vision”, line 8, in   
sys.exit(cli())  
File “/home/clapi/.local/lib/python3.10/site-packages/click/core.py”, line 1442, in **call**  
return self.main(\*args, \*\*kwargs)  
File “/home/clapi/.local/lib/python3.10/site-packages/click/core.py”, line 1363, in main  
rv = self.invoke(ctx)  
File “/home/clapi/.local/lib/python3.10/site-packages/click/core.py”, line 1830, in invoke  
return \_process\_result(sub\_ctx.command.invoke(sub\_ctx))  
File “/home/clapi/.local/lib/python3.10/site-packages/click/core.py”, line 1226, in invoke  
return ctx.invoke(self.callback, \*\*ctx.params)  
File “/home/clapi/.local/lib/python3.10/site-packages/click/core.py”, line 794, in invoke  
return callback(\*args, \*\*kwargs)  
File “/home/clapi/miniconda3/envs/warboy/lib/python3.10/site-packages/src/cli/make\_model/run\_make\_model.py”, line 35, in run\_make\_model  
onnx\_tools.export\_onnx(need\_edit=need\_edit)  
File “/home/clapi/miniconda3/envs/warboy/lib/python3.10/site-packages/src/warboy/tools/onnx\_tools.py”, line 39, in export\_onnx  
raise NotImplementedError(  
NotImplementedError: Export ONNX for object\_detection \>\> freeze\_detector is not implemented yet!

---

<div class="post-metadata">

**Author:** ![jongwook.kim](https://avatars.discourse-cdn.com/v4/letter/j/b5a626/32.png) [@jongwook.kim](https://forums.furiosa.ai/u/jongwook.kim)\
**Post date:** [July 26, 2025, 3:09am UTC](https://forums.furiosa.ai/t/ocr-make-model/311/2 "2025-07-26T03:09:55Z")

</div>

안녕하세요, 퓨리오사에이아이 김종욱입니다.

우선 warboy-vision-models 프로젝트의 경우 YOLO 시리즈 모델들을 지원하는 것을 기반으로 만들어져있습니다. 따라서 이외의 모델을 사용하시려고 하는 경우 에러가 발생할 수 있습니다.

기본적인 양자화에 대한 가이드는 아래를 참조하시는 것이 좋아보이며, 필요하시다면 작업에 도움을 드릴 수있도록 하겠습니다.

> **[모델 양자화 — FuriosaAI Warboy 및 Software 문서 0.10.2 문서](https://developer.furiosa.ai/docs/latest/ko/software/quantization.html)**