Oneflow-Inc / libai

LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training
https://libai.readthedocs.io
Apache License 2.0
389 stars 55 forks source link

mock model training demo【dont merge】 #497

Open xiezipeng-ML opened 1 year ago

xiezipeng-ML commented 1 year ago
# 2卡数据并行
[04/19 03:32:49 lb.models.utils.graph_base]: Start compling the train graph which may take some time. Please wait for a moment ...
F20230419 03:32:54.952033 3055036 shape.cpp:30] Check failed: index < tp()->NumAxes() (0 vs. 0)  Shape: () visit index: 0 > num_axes: 0
*** Check failure stack trace: ***
    @     0x7f90eff21c9a  google::LogMessage::Fail()
    @     0x7f90eff24bd1  google::LogMessage::SendToLog()
    @     0x7f90eff217c9  google::LogMessage::Flush()
    @     0x7f90eff254b9  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f90e3d1feac  oneflow::ConstShapeMixIn<>::At()
    @     0x7f90e868117e  oneflow::SubTskGphBuilderUtil::HasEmptySliceIfSplit()
    @     0x7f90e867ea77  oneflow::SliceBoxingSubTskGphBuilder::Build()
    @     0x7f90e8659811  oneflow::ChainSubTskGphBuilder::Build()
    @     0x7f90e866ee3a  oneflow::FlatSubTskGphBuilder::Build()
    @     0x7f90e866ddad  oneflow::DispatchHierarchicalSubTskGphBuilder::Build()
    @     0x7f90e870f232  oneflow::TaskGraph::BldSubTskGphByBoxing()
    @     0x7f90e87112e3  oneflow::TaskGraph::TaskGraph()
    @     0x7f90e87a6563  oneflow::Compiler::Compile()
    @     0x7f90e7ce1131  oneflow::NNGraph::CompilePlanForRuntime()
    @     0x7f91c13179a3  (unknown)
    @     0x7f91c118087a  (unknown)
    @           0x507f97  cfunction_call
    @           0x4f0f1c  _PyObject_MakeTpCall
    @           0x505ed0  method_vectorcall
    @           0x4ecebf  _PyEval_EvalFrameDefault
    @           0x4f8923  function_code_fastcall
    @           0x505bb1  method_vectorcall
    @           0x4ecebf  _PyEval_EvalFrameDefault
    @           0x4e729a  _PyEval_EvalCode
    @           0x505cbd  method_vectorcall
    @           0x5063f4  PyObject_Call
    @           0x4ec77b  _PyEval_EvalFrameDefault
    @           0x4e729a  _PyEval_EvalCode
    @           0x4f8645  _PyFunction_Vectorcall
    @           0x4f07f5  _PyObject_FastCallDictTstate
    @           0x503616  _PyObject_Call_Prepend
    @           0x5cc3f3  slot_tp_call
F20230419 03:32:55.400154 3055037 rpc_client.cpp:49] Check failed: stub->CallMethod<ctrl_method>(&client_ctx, request_, &response_).error_code() == grpc::StatusCode::OK (14 vs. 0) 
*** Check failure stack trace: ***
    @     0x7f585adf0c9a  google::LogMessage::Fail()
    @     0x7f585adf3bd1  google::LogMessage::SendToLog()
    @     0x7f585adf07c9  google::LogMessage::Flush()
    @     0x7f585adf44b9  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f584ec1d381  oneflow::RpcClient::PullKV()
    @     0x7f584ec1d514  oneflow::RpcClient::PullKV()
    @     0x7f5852bb066e  oneflow::NNGraph::CompilePlanForRuntime()
    @     0x7f592c1e69a3  (unknown)
    @     0x7f592c04f87a  (unknown)
    @           0x507f97  cfunction_call
    @           0x4f0f1c  _PyObject_MakeTpCall
    @           0x505ed0  method_vectorcall
    @           0x4ecebf  _PyEval_EvalFrameDefault
    @           0x4f8923  function_code_fastcall
    @           0x505bb1  method_vectorcall
    @           0x4ecebf  _PyEval_EvalFrameDefault
    @           0x4e729a  _PyEval_EvalCode
    @           0x505cbd  method_vectorcall
    @           0x5063f4  PyObject_Call
    @           0x4ec77b  _PyEval_EvalFrameDefault
    @           0x4e729a  _PyEval_EvalCode
    @           0x4f8645  _PyFunction_Vectorcall
    @           0x4f07f5  _PyObject_FastCallDictTstate
    @           0x503616  _PyObject_Call_Prepend
    @           0x5cc3f3  slot_tp_call
    @           0x506498  PyObject_Call
    @           0x4ec77b  _PyEval_EvalFrameDefault
    @           0x4e729a  _PyEval_EvalCode
    @           0x4f8645  _PyFunction_Vectorcall
    @           0x4e882b  _PyEval_EvalFrameDefault
    @           0x4f8923  function_code_fastcall
    @           0x4e882b  _PyEval_EvalFrameDefault
Killing subprocess 3055036
Killing subprocess 3055037
Traceback (most recent call last):
  File "/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/lib/python3.9/runpy.py", line 197, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/lib/python3.9/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/lib/python3.9/site-packages/oneflow/distributed/launch.py", line 240, in <module>
    main()
  File "/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/lib/python3.9/site-packages/oneflow/distributed/launch.py", line 228, in main
    sigkill_handler(signal.SIGTERM, None)
  File "/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/lib/python3.9/site-packages/oneflow/distributed/launch.py", line 196, in sigkill_handler
    raise subprocess.CalledProcessError(
subprocess.CalledProcessError: Command '['/home/xiezipeng/anaconda3/envs/oneflow-dev-gcc9/bin/python3', '-u', 'tools/train_net.py', '--config-file', 'projects/mock_gpt_train/configs/training.py']' died with <Signals.SIGABRT: 6>.