(base) HwHiAiUser@orangepiaipro-20t:~/mindyolo$ python deploy/export.py --config ./configs/yolov7/yolov7.yaml --weight ./yolov7_300e_mAP508-734ac919.ckpt --file_format MINDIR --device_target Ascend
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
return self._float_to_str(self.smallest_subnormal)
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
return self._float_to_str(self.smallest_subnormal)
2025-08-21 16:29:45,072 [WARNING] Parse Model, args: nearest, keep str type
2025-08-21 16:29:45,318 [WARNING] Parse Model, args: nearest, keep str type
2025-08-21 16:29:46,960 [INFO] number of network params, total: 37.672465M, trainable: 37.622682M
[WARNING] ME(31617:255085889290272,MainProcess):2025-08-21-16:30:02.337.260 [mindspore/train/serialization.py:1809] For 'load_param_into_net', remove parameter prefix name: ema., continue to load.
2025-08-21 16:30:02,369 [INFO] Load checkpoint from [./yolov7_300e_mAP508-734ac919.ckpt] success.
2025-08-21 16:30:27,050 [INFO] Export completed.
(base) HwHiAiUser@orangepiaipro-20t:~/mindyolo$ python deploy/mslite_predict.py --mindir_path yolov7.mindir --config ./configs/yolov7/yolov7.yaml --image_path ./88qNBT3ZPX.jpg /home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
return self._float_to_str(self.smallest_subnormal)
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
return self._float_to_str(self.smallest_subnormal)
2025-08-21 16:30:52,739 [INFO] mslite model init...
[WARNING] GE_ADPT(32627,e7ffc3192020,python):2025-08-21-16:30:55.224.453 [mindspore/ccsrc/transform/graph_ir/transform_util.cc:185] GetGeTensorDesc] undefined data type :14
[WARNING] GE_ADPT(32627,e7ffc3192020,python):2025-08-21-16:30:55.224.506 [mindspore/ccsrc/transform/graph_ir/op_adapter.cc:646] UpdateMultiOutputDesc] Create op: Default/Return-op0 output descriptor failed!
.[ERROR] ME(32627,e7ffc3192020,python):2025-08-21-16:32:39.961.289 [mindspore/lite/src/extendrt/kernel/ascend/model/model_process.cc:819] Resize] Not support dynamic input
[ERROR] ME(32627,e7ffc3192020,python):2025-08-21-16:32:39.961.351 [mindspore/lite/src/extendrt/kernel/ascend/src/custom_ascend_kernel.cc:254] OnNewInputShapes] Failed to Resize
[ERROR] ME(32627,e7ffc3192020,python):2025-08-21-16:32:39.961.369 [mindspore/lite/src/extendrt/kernel/ascend/src/custom_ascend_kernel.cc:200] Resize] Failed to resize inputs
[ERROR] ME(32627,e7ffc3192020,python):2025-08-21-16:32:39.961.385 [mindspore/lite/src/extendrt/session/single_op_session.cc:571] OnNewInputShapes] Failed to resize custom ascend kernel
Traceback (most recent call last):
File "/home/HwHiAiUser/mindyolo/deploy/mslite_predict.py", line 190, in <module>
infer(args)
File "/home/HwHiAiUser/mindyolo/deploy/mslite_predict.py", line 170, in infer
result_dict = detect(
File "/home/HwHiAiUser/mindyolo/deploy/mslite_predict.py", line 101, in detect
model.resize(inputs,[list(img.shape)])
File "/home/HwHiAiUser/.local/lib/python3.9/site-packages/mindspore_lite/model.py", line 493, in resize
super(Model, self).resize(inputs, dims)
File "/home/HwHiAiUser/.local/lib/python3.9/site-packages/mindspore_lite/base_model.py", line 153, in resize
raise RuntimeError(f"resize failed! Error is {ret.ToString()}")
RuntimeError: resize failed! Error is Common error code.
(base) HwHiAiUser@orangepiaipro-20t:~/mindyolo$ python deploy/mslite_predict.py --mindir_path yolov7_300e_mAP508-734ac919-6d65d27c.mindir --config ./configs/yolov7/yolov7.yaml --image_path ./88qNBT3ZPX.jpg
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
return self._float_to_str(self.smallest_subnormal)
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
return self._float_to_str(self.smallest_subnormal)
2025-08-21 16:23:09,135 [INFO] mslite model init...
[WARNING] GE_ADPT(28828,e7ffe5e1d020,python):2025-08-21-16:23:11.548.220 [mindspore/ccsrc/transform/graph_ir/transform_util.cc:185] GetGeTensorDesc] undefined data type :14
[WARNING] GE_ADPT(28828,e7ffe5e1d020,python):2025-08-21-16:23:11.548.275 [mindspore/ccsrc/transform/graph_ir/op_adapter.cc:646] UpdateMultiOutputDesc] Create op: Default/Return-op0 output descriptor failed!
.2025-08-21 16:24:18,871 [INFO] perform nms...
2025-08-21 16:24:18,873 [INFO] Predict result is: {'category_id': [17, 18], 'bbox': [[0.781, 228.844, 346.25, 452.812], [309.688, 49.469, 490.312, 627.812]], 'score': [0.94829, 0.82553]}
2025-08-21 16:24:18,874 [INFO] Speed: 69736.3/1.4/69737.7 ms inference/NMS/total per 640x640 image at batch-size 1;
2025-08-21 16:24:18,874 [INFO] Detect a image success.
2025-08-21 16:24:18,939 [INFO] predict completed.
使用香橙派AIpro20T，mindyolo==0.5.0，mindspore==2.4.10，mindspore-lite==2.4.10
使用export.py导出的mindir无法在mslite_predict.py上运行
但是官方提供的mindir可以正常运行