tools/export_model.py error

youngstu commented 3 years ago

python tools/export_model.py --config ... runnning error.

   File "/tmp/tmpn07wvto_.py", line 27, in forward
    false_fn_1, (), (), (out,))
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/dygraph_to_static/convert_operators.py", line 210, in convert_ifelse
    return _run_py_ifelse(pred, true_fn, false_fn, true_args, false_args)
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/dygraph_to_static/convert_operators.py", line 235, in _run_py_ifelse
    return true_fn(*true_args) if pred else false_fn(*false_args)
    File "/mnt/disk_2t/shaoxiong/handpp_work/research/PaddleDetection-Yolov5/ppdet/modeling/architectures/meta_arch.py", line 28, in forward (* user code *)
    out = self.get_pred()
    File "/mnt/disk_2t/shaoxiong/handpp_work/research/PaddleDetection-Yolov5/ppdet/modeling/architectures/yolo.py", line 201, in get_pred (* user code *)
    return self._forward()
    File "/mnt/disk_2t/shaoxiong/handpp_work/research/PaddleDetection-Yolov5/ppdet/modeling/architectures/yolo.py", line 160, in _forward (* user code *)
    body_feats = self.backbone(self.inputs)
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/layers.py", line 898, in __call__
    outputs = self.forward(*inputs, **kwargs)
    File "/tmp/tmp7igr6af4.py", line 29, in forward
    for_loop_condition_0, for_loop_body_0, [__for_loop_var_index_1, x, i])
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/dygraph_to_static/convert_operators.py", line 44, in convert_while_loop
    loop_vars = _run_py_while(cond, body, loop_vars)
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/dygraph_to_static/convert_operators.py", line 58, in _run_py_while
    loop_vars = body(*loop_vars)
    File "/mnt/disk_2t/shaoxiong/handpp_work/research/PaddleDetection-Yolov5/ppdet/modeling/backbones/cspdarknet.py", line 48, in forward (* user code *)
    x = m(x)  # run
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/dygraph/layers.py", line 898, in __call__
    outputs = self.forward(*inputs, **kwargs)
    File "/mnt/disk_2t/shaoxiong/handpp_work/research/PaddleDetection-Yolov5/ppdet/modeling/architectures/common.py", line 227, in forward (* user code *)
    y = paddle.concat([x[:, :, ::2, ::2], x[:, :, 1::2, ::2], x[:, :, ::2, 1::2], x[:, :, 1::2, 1::2]], 1)
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/framework.py", line 1832, in __getitem__
    return _getitem_impl_(self, item)
    File "/home/ar/shaoxiong/envs/anaconda3/envs/mobilehand/lib/python3.7/site-packages/paddle/fluid/framework.py", line 831, in _getitem_impl_
    assert (step == -1)
    AssertionError

Sharpiless commented 3 years ago

I'll check it later

youngstu commented 3 years ago

Could it save_inference_model? I want to save model.pdmodel and model.pdiparams in order to transform to paddleLite.

Sharpiless commented 3 years ago

是由于inference花式切片不支持导致的问题，新版本已经转换到 stried_slice 操作来解决这个问题 @youngstu

Sharpiless / PaddleDetection-Yolov5

tools/export_model.py error #2