Skip to content

the Converted Text file can't be visualized by et_visualizer #76

@lulala-s

Description

@lulala-s

Describe the Bug

mistake report is the following thing

Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.8/dist-packages/chakra/et_visualizer/et_visualizer.py", line 81, in <module>
    main()
  File "/usr/local/lib/python3.8/dist-packages/chakra/et_visualizer/et_visualizer.py", line 40, in main
    while decode_message(et, node):
  File "/usr/local/lib/python3.8/dist-packages/chakra/third_party/utils/protolib.py", line 138, in decodeMessage
    message.ParseFromString(buf)
google.protobuf.message.DecodeError: Error parsing message

the command which I used to visualizer is this

python3 -m chakra.et_visualizer.et_visualizer\
    --input_filename ../../outputs/convert_result/Resnet50_DataParallel/Resnet50_DataParallel.0.et\
    --output_filename ../../outputs/pic_result/Resnet50.pdf

Steps to Reproduce

I use the following command to convert the txt to et file

python3 -m chakra.et_converter.et_converter\
    --input_type Text\
    --input_filename ../../inputs/workload/ASTRA-sim-1.0/MLP_ModelParallel.txt\
    --output_filename ../../outputs/convert_result/MLP_ModelParallel\
    --num_npus 64\
    --num_dims 1\
    --num_passes 1

and the result is like this

����
�schema���1.0.2-chakra.0.0.4���

input_file����MODEL											
6											
layer_64_1_mlp0	-1	32291	ALLGATHER	37632	32291	ALLREDUCE	37632	12864	NONE	0	3229
layer_64_1_mlp1	-1	7488	ALLGATHER	65536	7488	ALLREDUCE	65536	3648	NONE	0	749
layer_64_1_mlp2	-1	7488	ALLGATHER	65536	7488	ALLREDUCE	65536	3456	NONE	0	749
layer_64_1_mlp3	-1	14144	ALLGATHER	147456	14144	ALLREDUCE	147456	10368	NONE	0	1414
layer_64_1_mlp4	-1	7488	ALLGATHER	65536	7488	ALLREDUCE	65536	3648	NONE	0	749
layer_64_2_mlp5	-1	9984	ALLGATHER	65536	9984	ALLREDUCE	65536	3456	NONE	0	998%��COMP_NODE_layer_64_1_mlp0_FWD��8���Q���(COMM_COLL_NODE_layer_64_1_mlp0_ALLGATHER��*�

Expected Behavior

Maybe is the convert file has some problems?

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions