Custom Model Inference

[kamalkannan79]

from paddleocr import PaddleOCR
custom_ocr=PaddleOCR(use_angle_cls=True,rec_model_dir="/content/drive/MyDrive/PaddleOCR/Inference_rec",det_model_dir="/content/drive/MyDrive/PaddleOCR/Inference_dir",rec_char_dict_path="/content/drive/MyDrive/PaddleOCR/ppocr/utils/en_dict.txt",use_gpu=True)
ocr = PaddleOCR(use_angle_cls=True, lang="en")
import cv2
import matplotlib.pyplot as plt
img_path = "/content/drive/MyDrive/PaddleOCR/Data_Anno/crop_20230629_151906.jpg.jpg"
img = cv2.imread(img_path)
plt.imshow(img)
result = custom_ocr.ocr(img)

while inferenceing i received valueerror
ValueError: (InvalidArgument) The size of Op(Conv) inputs should not be 0.
[Hint: Expected in_dims[i] != 0, but received in_dims[i]:0 == 0:0.] (at /paddle/paddle/phi/infermeta/binary.cc:465)
[operator < conv2d > error]

how to resolve this???????

[Sarveshsh8]

sure

[Liufeiran123]

same error!

[RickyGunawan09]

how u resolve this error ? i get the same error when using quick start for layout analysis

[elonzh]

PPStructure 预训练模型结构和 PaddleDetection 提供的配置文件模型结构不一样，需要使用 deploy/python/infer.py 进行推理，目前似乎没有提供代码方便地调用推理。