[图片]
[图片]
2022-07-05 08:26:17.966986: W tensorflow/core/framework/common_shape_fns.cc:587][74618] The attr name paddings is deprecated. Please use explicit_paddings instead.
Traceback (most recent call last):
File "evaluate.py", line 278, in <module>
YoloTest().evaluate()
File "evaluate.py", line 44, in __init__
graph = load_graph(self.model_file)
File "evaluate.py", line 235, in load_graph
tf.import_graph_def(graph_def)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/importer.py", line 427, in import_graph_def
graph._c_graph, serialized, options) # pylint: disable=protected-access
tensorflow.python.framework.errors_impl.NotFoundError: Op type not registered 'Yolov3DetectionOutput' in binary running on localhost. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib, accessing (e.g.) `tf.contrib.resampler` should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed.
你好!请详细描述一下问题,方便定位。比如操作步骤,报错日志,问题截图信息等。
1、bangc算子填写:补齐bangc/PluginYolov3DetectionOutputOp/nms_detection.h
2、集成到cnplugin:cnplugin.h和.cc文件已经写好,
只需要将整个PluginYolov3DetectionOutputOp文件夹复制到env/Cambricon-CNPlugin-MLU270/pluginops,重新编译cnplugin,
3、集成到tensorflow:补齐tf-implementation/tf-1.14-detectionoutput/目录下的各个文件,按照readme.txt提示拷入对应目录,重新编译tensorflow。
4. 转换模型之后模型调用"Yolov3DetectionOutput"的算子进行,但是没有删除其他的节点(实验手册中有“注释掉其他后处理节点就行了”这句话)。
5. 运行'run_aicse.sh'进行测试,显示我的算子还没有集成到TensorFlow框架中。
请登录后评论