Giters

microsoft / nn-Meter

A DNN inference latency prediction toolkit for accurately modeling and predicting the latency on diverse edge devices.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

google.protobuf.message.DecodeError: Error parsing message with type 'tensorflow.GraphDef'

howardgriffin opened this issue · comments

commented

when I use 'nn-meter predict --predictor cortexA76cpu_tflite21 --predictor-version 1.0 --tensorflow mobilenetv3small_0.onnx ' or 'nn-meter predict --predictor cortexA76cpu_tflite21 --tensorflow mobilenetv3small_0.json' in my command line, this error occured. Any suggestions?

Traceback (most recent call last):
File "/opt/conda/bin/nn-meter", line 8, in
sys.exit(nn_meter_cli())
File "/opt/conda/lib/python3.7/site-packages/nn_meter/nn_meter_cli.py", line 182, in nn_meter_cli
args.func(args)
File "/opt/conda/lib/python3.7/site-packages/nn_meter/nn_meter_cli.py", line 54, in apply_latency_predictor_cli
latency = predictor.predict(model, model_type) # in unit of ms
File "/opt/conda/lib/python3.7/site-packages/nn_meter/predictor/nn_meter_predictor.py", line 102, in predict
graph = model_file_to_graph(model, model_type, input_shape, apply_nni=apply_nni)
File "/opt/conda/lib/python3.7/site-packages/nn_meter/ir_converter/utils.py", line 41, in model_file_to_graph
converter = FrozenPbConverter(filename)
File "/opt/conda/lib/python3.7/site-packages/nn_meter/ir_converter/frozenpb_converter/frozenpb_converter.py", line 15, in init
parser = FrozenPbParser(file_name)
File "/opt/conda/lib/python3.7/site-packages/nn_meter/ir_converter/frozenpb_converter/frozenpb_parser.py", line 19, in init
graph.ParseFromString(f.read())
google.protobuf.message.DecodeError: Error parsing message with type 'tensorflow.GraphDef'

commented

Hi, thanks for using nn-Meter! In command line, the file format need to be specified before the file name. You can try ... --onnx mobilenetv3small_0.onnx instead of ... --tensorflow mobilenetv3small_0.onnx. The whole command line should be nn-meter predict --predictor cortexA76cpu_tflite21 --predictor-version 1.0 --onnx mobilenetv3small_0.onnx. Similarly, the prediction of json file should be nn-meter predict --predictor cortexA76cpu_tflite21 --nn-meter-ir mobilenetv3small_0.json.

We have noticed that the example command in README is wrong. We're sorry for that. The README command will be fixed soon.

commented

It does.
And I have another question: the json model and the pb model have the same latency, but the onnx model seems to have different latency from the json and pb model of the same hardware. How can this be explained？

commented

Thanks for raising this question! It's a bug in onnx model parser and we have fixed the bug in PR #60. Now the onnx model latency is 11.842ms. The slight gap between onnx model and json & pb model is caused by the different implementation of onnx model and pb model.