qingqing01
commented
5 years ago @sunxiaolong Please look at this question.
Open xiegegege opened 5 years ago
qingqing01
commented
5 years ago @sunxiaolong Please look at this question.
xiegegege
commented
5 years ago 最新的报错：
Invoke operator roi_align_grad error.
Python Callstacks:
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/framework.py", line 1773, in append_op
attrs=kwargs.get("attrs", None))
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
return self.main_program.current_block().append_op(*args, **kwargs)
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/layers/nn.py", line 7453, in roi_align
"sampling_ratio": sampling_ratio
File "/ssd1/xiege/model_7.23/models/PaddleCV/PaddleDetection/ppdet/modeling/roi_extractors/roi_extractor.py", line 91, in call
sampling_ratio=self.sampling_ratio)
File "/ssd1/xiege/model_7.23/models/PaddleCV/PaddleDetection/ppdet/modeling/architectures/cascade_rcnn.py", line 127, in build
roi_feat = self.roi_extractor(body_feats, proposals, spatial_scale)
File "/ssd1/xiege/model_7.23/models/PaddleCV/PaddleDetection/ppdet/modeling/architectures/cascade_rcnn.py", line 168, in train
return self.build(feed_vars, 'train')
File "tools/train.py", line 99, in main
train_fetches = model.train(feedvars)
File "tools/train.py", line 211, in
suyali
commented
5 years ago 遇到了同样的问题，怎么解决的呀？@xiegegege
xiegegege
commented
5 years ago 这个他们已经在修了，应该快merge了~
使用paddle 1.5分支，在CPU下运行PaddleDetection中的cascade模型，有如下报错： Traceback (most recent call last): File "tools/train.py", line 204, in
main()
File "tools/train.py", line 159, in main
outs = exe.run(train_compile_program, fetch_list=train_values)
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/executor.py", line 665, in run
return_numpy=return_numpy)
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/executor.py", line 527, in _run_parallel
exe.run(fetch_var_names, fetch_var_name)
paddle.fluid.core_avx.EnforceNotMet: Invoke operator distribute_fpn_proposals error.
Python Callstacks:
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/framework.py", line 1764, in append_op
attrs=kwargs.get("attrs", None))
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
return self.main_program.current_block().append_op(*args, **kwargs)
File "/opt/_internal/cpython-2.7.11-ucs2/lib/python2.7/site-packages/paddle/fluid/layers/detection.py", line 2837, in distribute_fpn_proposals
'refer_scale': refer_scale
File "/ssd1/xiege/model_7.1/models/PaddleCV/PaddleDetection/ppdet/modeling/roi_extractors/roi_extractor.py", line 76, in call
rois, k_min, k_max, self.canconical_level, self.canonical_size)
File "/ssd1/xiege/model_7.1/models/PaddleCV/PaddleDetection/ppdet/modeling/architectures/cascade_rcnn.py", line 119, in build
roi_feat = self.roi_extractor(body_feats, proposals, spatial_scale)
File "/ssd1/xiege/model_7.1/models/PaddleCV/PaddleDetection/ppdet/modeling/architectures/cascade_rcnn.py", line 160, in train
return self.build(feed_vars, 'train')
File "tools/train.py", line 95, in main
train_fetches = model.train(feed_vars)
File "tools/train.py", line 204, in
main()
C++ Callstacks:
Enforce failed. Expected begin_idx < end_idx, but received begin_idx:0 >= end_idx:0.
The start row index must be lesser than the end row index. at [/ssd1/xiege/paddle_wheel/Paddle_2.7/Paddle/paddle/fluid/framework/tensor.cc:78]