Error in training Slanet for table recognition

[vani-mcm]

I am new to paddlepaddle. Pls help me to solve this error. System environment : ubuntu 20.04 python version : 3.8.18 Installed libraries using requirements.txt file in paddle repo The error message:

1. ValueError: could not broadcast input array from shape (8,) into shape (4,)

2. return self._fmt.find(self.asctime_search) >= 0 RecursionError: maximum recursion depth exceeded while calling a Python object

3. SystemError: (Fatal) Blocking queue is killed because the data reader raises an exception. [Hint: Expected killed != true, but received killed:1 == true:1.] (at /paddle/paddle/fluid/operators/reader/blocking_queue.h:175) If any information is needed, I will provide the necessary information.

[sandy-git-only]

I got the same problem: ValueError: could not broadcast input array from shape (8,) into shape (4,)

I have changed the values below: ver : release/2.6 I used the pretrained model : ch_ppstructure_mobile_v2.0_SLANet

• box_format: &box_format xyxyxyxy
• loc_reg_num: &loc_reg_num 8 However it didnt work.

[weiwuxian1998]

i find the original box labeling format is [][][][], and the first and the third box is the top left and bottom right cords, so you can set box[0::2] in label_ops.py line 731, and can solve it