使用微调后的paddleocr_vl模型进行推理时报错
#17193
-
|
使用保存的模型检查点推理报错AttributeError: 'PaddleOCRVLConfig' object has no attribute 'quantization_config' |
Beta Was this translation helpful? Give feedback.
Answered by
LT-07
Nov 27, 2025
Replies: 1 comment
-
|
报错原因为使用微调后的模型推理时，paddlx==3.3.3会默认从config.json中加载一个quantization_config，而微调后的模型没有生成。paddlex==3.3.3不兼容，升级paddlx到3.3.9后得以解决。 |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
LT-07
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
报错原因为使用微调后的模型推理时，paddlx==3.3.3会默认从config.json中加载一个quantization_config，而微调后的模型没有生成。paddlex==3.3.3不兼容，升级paddlx到3.3.9后得以解决。