在对模型的权重进行量化之后,将其保存为TF服务,尽管能够比以前更慢,但我能够得到预测。
但是,在Quantizing权重以及模型之后,然后将其保存以用于服务和使用它,我从客户端收到以下错误 在tf服务docker容器(tensorflow / serving:latest-gpu)中的V100上运行此命令-客户端和模型
_Rendezvous:<_的RPC终止,终止于:status = StatusCode.UNIMPLEMENTED details =“在[1,9,4]和 尚不支持[221,1,4]。 [[{{节点 anchors_3 / add_2 / eightbit}}]] [[{{node filter_detections / map / while / non_max_suppression_47 / NonMaxSuppressionV3}}]]“ debug_error_string = “ {” created“:” @ 1552349235.607723578“,” description“:”从收到错误 peer”,“文件”:“ src / core / lib / surface / call.cc”,“ file_line”:1017,“ grpc_message”:“广播 尚不支持[1,9,4]和[221,1,4]之间的值。\ n \ t [[{{node anchors_3 / add_2 / eightbit}}]] \ n \ t [[{{node filter_detections / map / while / non_max_suppression_47 / NonMaxSuppressionV3}}]“”,“ grpc_status”:12}“
在TF服务端,出现以下错误
2019-03-12 00:07:13.149087:我 外部/ org_tensorflow / tensorflow /核心/内核/quantized_add_op.cc:546] ndims = 3 2019-03-12 00:07:13.149233:我 外部/ org_tensorflow / tensorflow /核心/内核/quantized_add_op.cc:547] bcast.x_reshape()= [1,9,4] 2019-03-12 00:07:13.149309:我 外部/ org_tensorflow / tensorflow /核心/内核/quantized_add_op.cc:549] bcast.y_reshape()= [221,1,4] 2019-03-12 00:07:13.149348:我 外部/ org_tensorflow / tensorflow /核心/内核/quantized_add_op.cc:551] bcast.x_bcast()= [221,1,1] 2019-03-12 00:07:13.149385:我 外部/ org_tensorflow / tensorflow /核心/内核/quantized_add_op.cc:553] bcast.y_bcast()= [1,9,1]
模型是图像检测模型
MetaGraphDef with tag-set: 'serve' contains the following SignatureDefs:
signature_def['serving_default']:
The given SavedModel SignatureDef contains the following input(s):
inputs['input_image'] tensor_info:
dtype: DT_FLOAT
shape: (-1, -1, -1, 3)
name: input_1_2:0
The given SavedModel SignatureDef contains the following output(s):
outputs['filtered_detections/map/TensorArrayStack/TensorArrayGatherV3:0'] tensor_info:
dtype: DT_FLOAT
shape: (-1, 300, 4)
name: filtered_detections/map/TensorArrayStack/TensorArrayGatherV3:0
outputs['filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3:0'] tensor_info:
dtype: DT_FLOAT
shape: (-1, 300)
name: filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3:0
outputs['filtered_detections/map/TensorArrayStack_2/TensorArrayGatherV3:0'] tensor_info:
dtype: DT_INT32
shape: (-1, 300)
name: filtered_detections/map/TensorArrayStack_2/TensorArrayGatherV3:0
Method name is: tensorflow/serving/predict