AttributeError: 'bool' object has no attribute 'sum' when use_gpu: False

[ltbd78]

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem

• 系统环境/System Environment：Python 3.12

• 版本号/Version

  • Paddle：2.6.1
  • PaddleOCR：2.7.3
  • PaddleNLP: 2.6.1 (also tried 2.8.0, 2.5.x, 2.7.x)

• 问题相关组件/Related components：
  Note only two changes differ:

  • in configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml, use_gpu: False
  • in configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml, use_gpu: False

• 运行指令/Command Code:

python3 ./tools/infer_kie_token_ser_re.py \
  -c configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml \
  -o Architecture.Backbone.checkpoints=./pretrained_model/re_vi_layoutxlm_xfund_pretrained/best_accuracy \
  Global.infer_img=./test_data/passport_card_1.jpg \
  -c_ser configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml \
  -o_ser Architecture.Backbone.checkpoints=./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy

• 完整报错/Complete Error Message：

I0516 16:45:36.134816 55311040 kernel_dispatch.h:102] Get BackendSet from tensor
Traceback (most recent call last):
  File "/Users/linsuhan/Documents/GitHub/PaddleOCR/./tools/infer_kie_token_ser_re.py", line 216, in <module>
    result = ser_re_engine(data)
             ^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/PaddleOCR/./tools/infer_kie_token_ser_re.py", line 151, in __call__
    preds = self.model(re_input)
            ^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddle/nn/layer/layers.py", line 1429, in __call__
    return self.forward(*inputs, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/PaddleOCR/ppocr/modeling/architectures/base_model.py", line 85, in forward
    x = self.backbone(x)
        ^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddle/nn/layer/layers.py", line 1429, in __call__
    return self.forward(*inputs, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/PaddleOCR/ppocr/modeling/backbones/vqa_layoutlm.py", line 248, in forward
    x = self.model(
        ^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddle/nn/layer/layers.py", line 1429, in __call__
    return self.forward(*inputs, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1329, in forward
    loss, pred_relations = self.extractor(sequence_output, entities, relations)
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddle/nn/layer/layers.py", line 1429, in __call__
    return self.forward(*inputs, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1223, in forward
    relations, entities = self.build_relation(relations, entities)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/linsuhan/Documents/GitHub/paddle/.venv/lib/python3.12/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1189, in build_relation
    if negative_mask.sum() > 0:
       ^^^^^^^^^^^^^^^^^
AttributeError: 'bool' object has no attribute 'sum'

[zhangyubo0722]

可以评估一下验证集精度，看看是不是精度比较低，感觉像是没推理出来内容

[564142183]

指标评估命令

python3 tools/eval.py -c configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml -o Architecture.Backbone.checkpoints=./output/ser_vi_layoutxlm_xfund_zh/best_accuracy

你好，这个问题有解决吗？