2.4K Star 8.2K Fork 4.4K

GVPMindSpore / mindspore

 / 详情

[CT][MS][op] The Parameter segment_ids's ’verification of UnsortedSegmentMin is error

DONE
Bug-Report
创建于  
2020-05-21 19:29
name about labels
Bug Report Use this template for reporting a bug kind/bug

Environment

  • Hardware Environment(Ascend/GPU/CPU):

Uncomment only one /device <> line, hit enter to put that in a new line, and remove leading whitespaces from that line:

/device ascend

  • Software Environment:
    -- MindSpore version (source or binary):ME+GE
    -- Python version (e.g., Python 3.7.5):
    -- OS platform and distribution (e.g., Linux Ubuntu 16.04):
    -- GCC/Compiler version (if compiled from source):

Related testcase

def test_unsortedsegmentmin_exception_segments_less_of_range():
    input_shape = (4, 3)
    segment_ids = [2, 0, 1]
    num_segments = 3
    input = np.random.randn(*input_shape).astype(np.float32)
    segments = np.array(segment_ids).astype(np.int32)
    net = UnsortedSegmentMin(num_segments)
    fact = AnyNetFactory(net=net)
    with pytest.raises(ValueError):
      out1 = fact(Tensor(input), Tensor(segments))

test_unsortedsegmentmin.py:129:


../share/meta.py:225: in call
return self.net(*args)
../share/utils.py:61: in call
out = super().call(*args, **kwargs )
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/nn/cell.py:164: in call
out = self.compile_and_run(*inputs)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/nn/cell.py:327: in compile_and_run
return _executor(self, *inputs, phase=self.phase)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:427: in call
return self.run(obj, *args, phase=phase)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:455: in run
return self._exec_pip(obj, *args, phase=phase_real)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:67: in wrapper
results = fn(*arg, **kwargs)


self = <mindspore.common.api._Executor object at 0x7fcbb358ce50>, obj = WrapOp<>, phase = '0train.1590060194879452928'
args = ([[ 0.03051996 -0.14384444 1.536539 ]
[ 0.02804725 -0.20637493 0.9453783 ]
[ 0.52259713 -1.1823701 0.44734305]
[-1.3016951 -0.7353799 1.451657 ]], [2 0 1])
fn = <bound method UnsortedSegmentMin.construct of WrapOp<>>, converted = True

@_wrap_func
def _exec_pip(self, obj, *args, phase=''):
    """Execute the generated pipeline."""
    fn = obj.construct
    converted, arguments_dict, parse_method = _convert_function_arguments(fn, *args)
    if not converted:
        raise RuntimeError('Process method parameter is failure')
    args_list = tuple(arguments_dict.values())
    obj.__parse_method__ = parse_method
  return self._executor(args_list, phase)

E RuntimeError: mindspore/ccsrc/pipeline/pipeline_ge.cc:513 ExecDFGraph] Exec graph failed

/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:438: RuntimeError

Steps to reproduce the issue

  1. pytest -s test_unsortedsegmentmin.py::test_unsortedsegmentmin_exception_segments_less_of_range

Describe the current behavior

[ERROR] GE(96141,python):2020-05-21-19:23:16.691.090 [framework/domi/engine_manager/dnnengine_manager.cc:225]96141 GetDNNEngineName: ErrorNo: 1343242282(assign engine failed) GetDNNEngineName:Op type UnsortedSegmentMinD of ops kernel AIcoreEngine is unsupported, reason:Op Default/UnsortedSegmentMin-op0 not supported reason: The op desc is not supported in the operator implementation, consult the operator developer for details. Op store name is tbe-builtin.
The type of this op is not found in op store, check whether the op store has this type of op. Op store name is tbe-plugin.
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.115 [framework/domi/engine_manager/dnnengine_manager.cc:228]96141 GetDNNEngineName: ErrorNo: 1343242282(assign engine failed) Can't find any supported ops kernel and engine of Default/UnsortedSegmentMin-op0, type is UnsortedSegmentMinD
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.135 [framework/domi/graph/partition/engine_place.cc:54]96141 Run: ErrorNo: 1343229963(GE is not yet initialized or is finalized.) Can not find engine of op type UnsortedSegmentMinD
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.152 [framework/domi/graph/partition/graph_partition.cc:558]96141 Initialize: ErrorNo: -1(failed) Engine placer run failed.
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.168 [framework/domi/graph/partition/graph_partition.cc:908]96141 PartitionSubGraph: ErrorNo: 1343242240(Failed to initialize graph.) [GraphPartitioner]: initialize failed
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.183 [framework/domi/graph/partition/graph_partition.cc:872]96141 Partition: ErrorNo: -1(failed) Sub graph partition Failed
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.198 [framework/domi/graph/manager/graph_manager.cc:354]96141 PreRun: ErrorNo: -1(failed) Graph partition Failed
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.213 [framework/domi/graph/manager/graph_manager.cc:457]96141 StartForRunGraph: ErrorNo: -1(failed) PreRun Failed.
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.227 [framework/domi/graph/manager/graph_manager.cc:570]96141 RunGraph: ErrorNo: -1(failed) [RunGraph] StartForRunGraph failed!
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.243 [framework/domi/session/inner_session.cc:132]96141 RunGraph: ErrorNo: -1(failed) [InnerSession:0] run graph failed, graph_id=1.
[ERROR] GE(96141,python):2020-05-21-19:23:16.691.286 [framework/domi/client/ge_api.cc:385]96141 RunGraph: ErrorNo: -1(failed) Session RunGraph failed
[ERROR] ME(96141,python):2020-05-21-19:23:16.691.311 [mindspore/ccsrc/transform/graph_runner.cc:150] RunGraph] Call GE RunGraph Failed, ret is: 4294967295
[ERROR] ME(96141,python):2020-05-21-19:23:16.691.330 [mindspore/ccsrc/pipeline/pipeline_ge.cc:416] DoExecGraph] Exec graph failed
[ERROR] ME(96141,python):2020-05-21-19:23:16.691.354 [mindspore/ccsrc/pipeline/pipeline_ge.cc:513] ExecDFGraph] Exec graph failed

 def test_unsortedsegmentmin_exception_segments_less_of_range():
        input_shape = (4, 3)
        segment_ids = [2, 0, 1]
        num_segments = 3
        input = np.random.randn(*input_shape).astype(np.float32)
        segments = np.array(segment_ids).astype(np.int32)
        net = UnsortedSegmentMin(num_segments)
        fact = AnyNetFactory(net=net)
        with pytest.raises(ValueError):
>           out1 = fact(Tensor(input), Tensor(segments))

test_unsortedsegmentmin.py:129:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
../share/meta.py:225: in __call__
    return self.net(*args)
../share/utils.py:61: in __call__
    out = super().__call__(*args, **kwargs )
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/nn/cell.py:164: in __call__
    out = self.compile_and_run(*inputs)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/nn/cell.py:327: in compile_and_run
    return _executor(self, *inputs, phase=self.phase)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:427: in __call__
    return self.run(obj, *args, phase=phase)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:455: in run
    return self._exec_pip(obj, *args, phase=phase_real)
/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:67: in wrapper
    results = fn(*arg, **kwargs)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

self = <mindspore.common.api._Executor object at 0x7fcbb358ce50>, obj = WrapOp<>, phase = '0train.1590060194879452928'
args = ([[ 0.03051996 -0.14384444  1.536539  ]
 [ 0.02804725 -0.20637493  0.9453783 ]
 [ 0.52259713 -1.1823701   0.44734305]
 [-1.3016951  -0.7353799   1.451657  ]], [2 0 1])
fn = <bound method UnsortedSegmentMin.construct of WrapOp<>>, converted = True

    @_wrap_func
    def _exec_pip(self, obj, *args, phase=''):
        """Execute the generated pipeline."""
        fn = obj.construct
        converted, arguments_dict, parse_method = _convert_function_arguments(fn, *args)
        if not converted:
            raise RuntimeError('Process method parameter is failure')
        args_list = tuple(arguments_dict.values())
        obj.__parse_method__ = parse_method
>       return self._executor(args_list, phase)
E       RuntimeError: mindspore/ccsrc/pipeline/pipeline_ge.cc:513 ExecDFGraph] Exec graph failed

/root/miniconda3/envs/zb_ge/lib/python3.7/site-packages/mindspore/common/api.py:438: RuntimeError

Describe the expected behavior

raise exception

Related log / screenshot

Special notes for this issue

评论 (3)

HidyLi 创建了Bug-Report
HidyLi 计划截止日期设置为2020-12-31
HidyLi 关联项目设置为MindSpore Bug Tracking System
HidyLi 关联仓库设置为MindSpore/mindspore
HidyLi 负责人设置为liuxiao93
HidyLi 里程碑设置为B-ME
HidyLi 计划开始日期设置为2020-05-21
HidyLi 优先级设置为次要
HidyLi 添加了
 
kind/bug
标签
HidyLi 添加了device/ascend(已删除)标签
展开全部操作日志

test_unsortedsegmentmin_exception_segments_less_of_range
test_unsortedsegmentmin_exception_segments_great_of_range

输入图片说明
已加入校验,请重新测试

liuxiao93 添加协作者liuxiao93
liuxiao93 负责人liuxiao93 修改为HidyLi
liuxiao93 任务状态TODO 修改为VALIDATION
liuxiao93 里程碑B-NET 修改为B-ComponentTest
HidyLi 任务状态VALIDATION 修改为DONE
HidyLi 移除了device/ascend(已删除)标签

登录 后才可以发表评论

状态
负责人
项目
里程碑
Pull Requests
关联的 Pull Requests 被合并后可能会关闭此 issue
分支
开始日期   -   截止日期
-
置顶选项
优先级
预计工期 (小时)
参与者(3)
6557666 hidyli 1648689999 6561470 liangchenghui 1584762793
Python
1
https://gitee.com/mindspore/mindspore.git
git@gitee.com:mindspore/mindspore.git
mindspore
mindspore
mindspore

搜索帮助