关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
OpenModelZoo
/
feedback_jijiehao
Not watched
361
0
Code
Wiki
Issues 3 Datasets Cloudbrain
Issues
/
Detail
Labels Milestones
New Issue

#2 算子nn.Conv2d中group参数与dilation参数不兼容问题

Closed
created 3 years ago by zhangximing · 2 comments
zhangximing commented 3 years ago
name about labels
Bug Report Use this template for reporting a bug kind/bug

Environment

  • Hardware Environment(Ascend/GPU/CPU): 线下机器 /device ascend

  • Software Environment:
    -- MindSpore version (source or binary): MindSpore r1.2
    -- Python version (e.g., Python 3.7.5): Python 3.7.6
    -- OS platform and distribution (e.g., Linux Ubuntu 16.04): eulerosv2r8.aarch64
    -- GCC/Compiler version (if compiled from source): 7.5.0

Related testcase

测试代码如下


def test():
    import numpy as np
    from mindspore import Tensor
    from mindspore import dtype as mstype

    in_channels = 320
    shapes = 97
    data = Tensor(np.random.rand(10, in_channels, shapes, shapes), mstype.float32)

    kernel_size = 3
    op = nn.Conv2d(in_channels, in_channels, kernel_size, stride=1, dilation=2, group=in_channels, pad_mode='same')
    featuremap = op(data)

    print(featuremap)


if __name__ == '__main__':
    test()

Steps to reproduce the issue

  1. 实际使用代码请在gitee有发部分https://gitee.com/mindspore/mindspore/issues/I3I88M
  2. 测试出是Conv2d的问题，直接运行上述代码即可复现问题

Describe the current behavior

对于一下涉及的shape，运行情况不同

shape DilConv_3x3 DilConv_5x5
(10, 320,97,97) fail pass
(10, 640,49,49) fail fail
(10,1280,25,25) fail fail
(10,2560,13,13) pass pass

补充说明

发现是nn.Conv2d 算子可能有问题：
1、不设置group（group=1）、仅设置空洞率（dilation=2）的空洞卷积不报错；
2、设置group（group=input_channels）、不设置空洞率（dilation=1)的depthwise separable 卷积不报错。
可能是conv2d算子的group与dilation之间有什么问题导致出错

Describe the expected behavior

以上代码能够跑通

Related log / screenshot

ssh://root@183.129.171.130:6483/root/archiconda3/envs/wks/bin/python3 -u /root/.pycharm_helpers/pydev/pydevd.py --multiproc --qt-support=auto --client 0.0.0.0 --port 36545 --file /home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py
Connected to pydev debugger (build 203.7148.72)
[WARNING][2021-04-10 21:57:30,528]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/dtype.py:145: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
[WARNING][2021-04-10 21:57:31,640]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:23: DeprecationWarning: Call to deprecated create function FileDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:31,640]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:42: DeprecationWarning: Call to deprecated create function FieldDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:31,641]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:63: DeprecationWarning: Call to deprecated create function Descriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:31,645]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/print_pb2.py:108: DeprecationWarning: Call to deprecated create function OneofDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:31,658]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:37: DeprecationWarning: Call to deprecated create function EnumValueDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:31,659]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:118: DeprecationWarning: Call to deprecated create function EnumDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:40,624]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/dtype.py:145: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
[WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:23: DeprecationWarning: Call to deprecated create function FileDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:42: DeprecationWarning: Call to deprecated create function FieldDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:63: DeprecationWarning: Call to deprecated create function Descriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:41,148]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/print_pb2.py:108: DeprecationWarning: Call to deprecated create function OneofDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:41,153]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:37: DeprecationWarning: Call to deprecated create function EnumValueDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[WARNING][2021-04-10 21:57:41,153]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:118: DeprecationWarning: Call to deprecated create function EnumDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool.
[ERROR] KERNEL(24249,python3):2021-04-10-21:57:48.131.671 [mindspore/ccsrc/backend/kernel_compiler/tbe/tbe_kernel_parallel_build.cc:99] TbeOpParallelBuild] task compile Failed, task id:8, cause:TBEException:ERROR:
[WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:30: DeprecationWarning: register is expired, please replace it with the func register_op_compute
[WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:58: DeprecationWarning: te.utils.para_check is expired, please replace it with tbe.common.utils.para_check
[WARNING][2021-04-10 21:57:46,052]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:97: DeprecationWarning: check_input_type in topi is expired, please do not use it
[WARNING][2021-04-10 21:57:46,131]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/util/util_frac_z.py:18: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3,and in 3.9 it will stop working
[WARNING][2021-04-10 21:57:46,344]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/lang/cce/te_compute/depthwise_conv2d_compute.py:65: DeprecationWarning: te.lang.cce.te_compute.depthwise_conv2d_compute is expired, please replace it with the func tbe.dsl.compute.depthwise_conv2d_compute
[WARNING][2021-04-10 21:57:46,345]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:868: DeprecationWarning: check_load3d_w_out_1_support in topi is expired, please do not use it
[WARNING][2021-04-10 21:57:46,359]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/utils/cce.py:37: DeprecationWarning: te.lang.cce.auto_schedule is deprecated, please replace it with the tbe.dsl.auto_schedule
Traceback (most recent call last):
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 129, in build_op
    res = op_func(*inputs_args, *outputs_args, *attrs_args, kernel_name=kernel_name)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/tbe/common/utils/para_check.py", line 529, in _in_wrapper
    return func(*args, **kwargs)
  File "/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/depthwise_conv2d.py", line 414, in depthwise_conv2d
    tvm.build_module.build(sch, tensor_list, "cce", name=kernel_name)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 900, in build
    evaluates=evaluates)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 610, in lower
    return cce_lower(sch, args, name, binds, simple_mode, evaluates)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 281, in cce_lower
    arg_list, cfg, evaluates, multi_process)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 80, in cce_static_lower
    stmt = ir_pass.EmitInsn(stmt)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/_ffi/_ctypes/function.py", line 209, in __call__
    raise get_last_ffi_error()
tvm._ffi.base.TVMError: Traceback (most recent call last):
  [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c]
  [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec]
  [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c]
  [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c]
  [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4]
  [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4]
  [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70]
  [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c]
  [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c]
  File "tilefy_pass.cc", line 2163
TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)])
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 174, in <module>
    result = compile_with_json(in_args)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 168, in compile_with_json
    ret = build_op(op_build, json_str, None)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 135, in build_op
    raise RuntimeError(e)
RuntimeError: Traceback (most recent call last):
  [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c]
  [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec]
  [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c]
  [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c]
  [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4]
  [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4]
  [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70]
  [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c]
  [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c]
  File "tilefy_pass.cc", line 2163
TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)])
input_args: {"SocInfo":{"autoTilingMode":"NO_TUNE","coreNum":"","coreType":"","l1Fusion":"false","l2Fusion":"false","l2Mode":"2","op_debug_level":"","op_impl_mode":"","op_impl_mode_list":[],"socVersion":"Ascend910B"},"impl_path":"","op_info":{"Type":"DepthwiseConv2dNative","attr_desc":[[1,1,1,1],[1,1,2,2],[2,2,2,2],"NCHW",0],"attrs":[{"name":"stride","valid":true,"value":[1,1,1,1]},{"name":"dilation","valid":true,"value":[1,1,2,2]},{"name":"pad_list","valid":true,"value":[2,2,2,2]},{"name":"format","valid":true,"value":"NCHW"},{"name":"offset_a","valid":true,"value":0}],"full_name":"Default/DepthwiseConv2dNative-op8","gen_model":"single","graph_id":1,"inputs":[[{"dtype":"float16","format":"NC1HWC0","name":"x_0","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}],[{"dtype":"float16","format":"C1HWNCoC0","name":"filter_0","ori_format":"NCHW","ori_shape":[1,320,3,3],"param_type":"required","range":[[1,1],[320,320],[3,3],[3,3]],"shape":[20,3,3,1,16,16],"valid":true}],[{"name":"bias_optional_2","valid":false}],[{"name":"offset_w_optional_2","valid":false}]],"is_dynamic_shape":false,"kernel_name":"DepthwiseConv2dNative_8172484865844151394_6","module_name":"impl.depthwise_conv2d","name":"depthwise_conv2d","outputs":[[{"dtype":"float16","format":"NC1HWC0","name":"y","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}]],"py_module_path":"/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe","socVersion":"Ascend910B"},"platform":"TBE"} trace: 
Traceback (most recent call last):
  File "/root/.pycharm_helpers/pydev/pydevd.py", line 1477, in _exec
    pydev_imports.execfile(file, globals, locals)  # execute the script
  File "/root/.pycharm_helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
python-BaseException
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 56, in <module>
    test()
  File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 50, in test
    featuremap = op(data)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__
    output = self.run_construct(cast_inputs, kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct
    output = self.construct(*cast_inputs, **kwargs)
  File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 34, in construct
    return self.op(x)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__
    output = self.run_construct(cast_inputs, kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct
    output = self.construct(*cast_inputs, **kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/layer/container.py", line 235, in construct
    input_data = cell(input_data)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__
    output = self.run_construct(cast_inputs, kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct
    output = self.construct(*cast_inputs, **kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/layer/conv.py", line 258, in construct
    output = self.conv2d(x, self.weight)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/ops/primitive.py", line 188, in __call__
    return _run_op(self, self.name, args)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/api.py", line 75, in wrapper
    results = fn(*arg, **kwargs)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/ops/primitive.py", line 556, in _run_op
    output = real_run_op(obj, op_name, args)
RuntimeError: mindspore/ccsrc/backend/kernel_compiler/tbe/tbe_kernel_parallel_build.cc:99 TbeOpParallelBuild] task compile Failed, task id:8, cause:TBEException:ERROR:
[WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:30: DeprecationWarning: register is expired, please replace it with the func register_op_compute
[WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:58: DeprecationWarning: te.utils.para_check is expired, please replace it with tbe.common.utils.para_check
[WARNING][2021-04-10 21:57:46,052]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:97: DeprecationWarning: check_input_type in topi is expired, please do not use it
[WARNING][2021-04-10 21:57:46,131]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/util/util_frac_z.py:18: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3,and in 3.9 it will stop working
[WARNING][2021-04-10 21:57:46,344]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/lang/cce/te_compute/depthwise_conv2d_compute.py:65: DeprecationWarning: te.lang.cce.te_compute.depthwise_conv2d_compute is expired, please replace it with the func tbe.dsl.compute.depthwise_conv2d_compute
[WARNING][2021-04-10 21:57:46,345]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:868: DeprecationWarning: check_load3d_w_out_1_support in topi is expired, please do not use it
[WARNING][2021-04-10 21:57:46,359]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/utils/cce.py:37: DeprecationWarning: te.lang.cce.auto_schedule is deprecated, please replace it with the tbe.dsl.auto_schedule
Traceback (most recent call last):
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 129, in build_op
    res = op_func(*inputs_args, *outputs_args, *attrs_args, kernel_name=kernel_name)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/tbe/common/utils/para_check.py", line 529, in _in_wrapper
    return func(*args, **kwargs)
  File "/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/depthwise_conv2d.py", line 414, in depthwise_conv2d
    tvm.build_module.build(sch, tensor_list, "cce", name=kernel_name)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 900, in build
    evaluates=evaluates)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 610, in lower
    return cce_lower(sch, args, name, binds, simple_mode, evaluates)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 281, in cce_lower
    arg_list, cfg, evaluates, multi_process)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 80, in cce_static_lower
    stmt = ir_pass.EmitInsn(stmt)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/_ffi/_ctypes/function.py", line 209, in __call__
    raise get_last_ffi_error()
tvm._ffi.base.TVMError: Traceback (most recent call last):
  [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c]
  [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec]
  [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c]
  [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c]
  [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4]
  [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4]
  [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70]
  [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c]
  [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c]
  File "tilefy_pass.cc", line 2163
TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)])
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 174, in <module>
    result = compile_with_json(in_args)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 168, in compile_with_json
    ret = build_op(op_build, json_str, None)
  File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 135, in build_op
    raise RuntimeError(e)
RuntimeError: Traceback (most recent call last):
  [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c]
  [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec]
  [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c]
  [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c]
  [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4]
  [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4]
  [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70]
  [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c]
  [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c]
  File "tilefy_pass.cc", line 2163
TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)])
input_args: {"SocInfo":{"autoTilingMode":"NO_TUNE","coreNum":"","coreType":"","l1Fusion":"false","l2Fusion":"false","l2Mode":"2","op_debug_level":"","op_impl_mode":"","op_impl_mode_list":[],"socVersion":"Ascend910B"},"impl_path":"","op_info":{"Type":"DepthwiseConv2dNative","attr_desc":[[1,1,1,1],[1,1,2,2],[2,2,2,2],"NCHW",0],"attrs":[{"name":"stride","valid":true,"value":[1,1,1,1]},{"name":"dilation","valid":true,"value":[1,1,2,2]},{"name":"pad_list","valid":true,"value":[2,2,2,2]},{"name":"format","valid":true,"value":"NCHW"},{"name":"offset_a","valid":true,"value":0}],"full_name":"Default/DepthwiseConv2dNative-op8","gen_model":"single","graph_id":1,"inputs":[[{"dtype":"float16","format":"NC1HWC0","name":"x_0","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}],[{"dtype":"float16","format":"C1HWNCoC0","name":"filter_0","ori_format":"NCHW","ori_shape":[1,320,3,3],"param_type":"required","range":[[1,1],[320,320],[3,3],[3,3]],"shape":[20,3,3,1,16,16],"valid":true}],[{"name":"bias_optional_2","valid":false}],[{"name":"offset_w_optional_2","valid":false}]],"is_dynamic_shape":false,"kernel_name":"DepthwiseConv2dNative_8172484865844151394_6","module_name":"impl.depthwise_conv2d","name":"depthwise_conv2d","outputs":[[{"dtype":"float16","format":"NC1HWC0","name":"y","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}]],"py_module_path":"/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe","socVersion":"Ascend910B"},"platform":"TBE"} trace: 
#

Special notes for this issue

这是gitee的issues链接：
https://gitee.com/mindspore/mindspore/issues/I3I88M

| name | about | labels | | ---------- | ------------------------------------- | -------- | | Bug Report | Use this template for reporting a bug | kind/bug | ## Environment - **Hardware Environment(`Ascend`/`GPU`/`CPU`)**: 线下机器 /device ascend - **Software Environment**: -- MindSpore version (source or binary): MindSpore r1.2 -- Python version (e.g., Python 3.7.5): Python 3.7.6 -- OS platform and distribution (e.g., Linux Ubuntu 16.04): eulerosv2r8.aarch64 -- GCC/Compiler version (if compiled from source): 7.5.0 ## Related testcase 测试代码如下 ``` def test(): import numpy as np from mindspore import Tensor from mindspore import dtype as mstype in_channels = 320 shapes = 97 data = Tensor(np.random.rand(10, in_channels, shapes, shapes), mstype.float32) kernel_size = 3 op = nn.Conv2d(in_channels, in_channels, kernel_size, stride=1, dilation=2, group=in_channels, pad_mode='same') featuremap = op(data) print(featuremap) if __name__ == '__main__': test() ``` ## Steps to reproduce the issue 1. 实际使用代码请在gitee有发部分https://gitee.com/mindspore/mindspore/issues/I3I88M 2. 测试出是Conv2d的问题，直接运行上述代码即可复现问题 3. ## Describe the current behavior 对于一下涉及的shape，运行情况不同 | shape | DilConv_3x3 | DilConv_5x5 | | ------------------- | ----------- | ----------- | | (10, 320,97,97) | fail | pass | | (10, 640,49,49) | fail | fail | | (10,1280,25,25) | fail | fail | | (10,2560,13,13) | pass | pass | ### 补充说明 发现是nn.Conv2d 算子可能有问题： 1、不设置group（group=1）、仅设置空洞率（dilation=2）的空洞卷积不报错； 2、设置group（group=input_channels）、不设置空洞率（dilation=1)的depthwise separable 卷积不报错。 可能是conv2d算子的group与dilation之间有什么问题导致出错 ## Describe the expected behavior 以上代码能够跑通 ## Related log / screenshot ``` ssh://root@183.129.171.130:6483/root/archiconda3/envs/wks/bin/python3 -u /root/.pycharm_helpers/pydev/pydevd.py --multiproc --qt-support=auto --client 0.0.0.0 --port 36545 --file /home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py Connected to pydev debugger (build 203.7148.72) [WARNING][2021-04-10 21:57:30,528]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/dtype.py:145: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here. Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations [WARNING][2021-04-10 21:57:31,640]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:23: DeprecationWarning: Call to deprecated create function FileDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:31,640]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:42: DeprecationWarning: Call to deprecated create function FieldDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:31,641]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:63: DeprecationWarning: Call to deprecated create function Descriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:31,645]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/print_pb2.py:108: DeprecationWarning: Call to deprecated create function OneofDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:31,658]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:37: DeprecationWarning: Call to deprecated create function EnumValueDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:31,659]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:118: DeprecationWarning: Call to deprecated create function EnumDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:40,624]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/dtype.py:145: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here. Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations [WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:23: DeprecationWarning: Call to deprecated create function FileDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:42: DeprecationWarning: Call to deprecated create function FieldDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:41,147]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/checkpoint_pb2.py:63: DeprecationWarning: Call to deprecated create function Descriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:41,148]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/print_pb2.py:108: DeprecationWarning: Call to deprecated create function OneofDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:41,153]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:37: DeprecationWarning: Call to deprecated create function EnumValueDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [WARNING][2021-04-10 21:57:41,153]/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/train/mind_ir_pb2.py:118: DeprecationWarning: Call to deprecated create function EnumDescriptor(). Note: Create unlinked descriptors is going to go away. Please use get/find descriptors from generated code or query the descriptor_pool. [ERROR] KERNEL(24249,python3):2021-04-10-21:57:48.131.671 [mindspore/ccsrc/backend/kernel_compiler/tbe/tbe_kernel_parallel_build.cc:99] TbeOpParallelBuild] task compile Failed, task id:8, cause:TBEException:ERROR: [WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:30: DeprecationWarning: register is expired, please replace it with the func register_op_compute [WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:58: DeprecationWarning: te.utils.para_check is expired, please replace it with tbe.common.utils.para_check [WARNING][2021-04-10 21:57:46,052]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:97: DeprecationWarning: check_input_type in topi is expired, please do not use it [WARNING][2021-04-10 21:57:46,131]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/util/util_frac_z.py:18: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3,and in 3.9 it will stop working [WARNING][2021-04-10 21:57:46,344]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/lang/cce/te_compute/depthwise_conv2d_compute.py:65: DeprecationWarning: te.lang.cce.te_compute.depthwise_conv2d_compute is expired, please replace it with the func tbe.dsl.compute.depthwise_conv2d_compute [WARNING][2021-04-10 21:57:46,345]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:868: DeprecationWarning: check_load3d_w_out_1_support in topi is expired, please do not use it [WARNING][2021-04-10 21:57:46,359]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/utils/cce.py:37: DeprecationWarning: te.lang.cce.auto_schedule is deprecated, please replace it with the tbe.dsl.auto_schedule Traceback (most recent call last): File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 129, in build_op res = op_func(*inputs_args, *outputs_args, *attrs_args, kernel_name=kernel_name) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/tbe/common/utils/para_check.py", line 529, in _in_wrapper return func(*args, **kwargs) File "/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/depthwise_conv2d.py", line 414, in depthwise_conv2d tvm.build_module.build(sch, tensor_list, "cce", name=kernel_name) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 900, in build evaluates=evaluates) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 610, in lower return cce_lower(sch, args, name, binds, simple_mode, evaluates) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 281, in cce_lower arg_list, cfg, evaluates, multi_process) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 80, in cce_static_lower stmt = ir_pass.EmitInsn(stmt) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/_ffi/_ctypes/function.py", line 209, in __call__ raise get_last_ffi_error() tvm._ffi.base.TVMError: Traceback (most recent call last): [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c] [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec] [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c] [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c] [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4] [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4] [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70] [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c] [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c] File "tilefy_pass.cc", line 2163 TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)]) During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 174, in <module> result = compile_with_json(in_args) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 168, in compile_with_json ret = build_op(op_build, json_str, None) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 135, in build_op raise RuntimeError(e) RuntimeError: Traceback (most recent call last): [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c] [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec] [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c] [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c] [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4] [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4] [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70] [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c] [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c] File "tilefy_pass.cc", line 2163 TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)]) input_args: {"SocInfo":{"autoTilingMode":"NO_TUNE","coreNum":"","coreType":"","l1Fusion":"false","l2Fusion":"false","l2Mode":"2","op_debug_level":"","op_impl_mode":"","op_impl_mode_list":[],"socVersion":"Ascend910B"},"impl_path":"","op_info":{"Type":"DepthwiseConv2dNative","attr_desc":[[1,1,1,1],[1,1,2,2],[2,2,2,2],"NCHW",0],"attrs":[{"name":"stride","valid":true,"value":[1,1,1,1]},{"name":"dilation","valid":true,"value":[1,1,2,2]},{"name":"pad_list","valid":true,"value":[2,2,2,2]},{"name":"format","valid":true,"value":"NCHW"},{"name":"offset_a","valid":true,"value":0}],"full_name":"Default/DepthwiseConv2dNative-op8","gen_model":"single","graph_id":1,"inputs":[[{"dtype":"float16","format":"NC1HWC0","name":"x_0","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}],[{"dtype":"float16","format":"C1HWNCoC0","name":"filter_0","ori_format":"NCHW","ori_shape":[1,320,3,3],"param_type":"required","range":[[1,1],[320,320],[3,3],[3,3]],"shape":[20,3,3,1,16,16],"valid":true}],[{"name":"bias_optional_2","valid":false}],[{"name":"offset_w_optional_2","valid":false}]],"is_dynamic_shape":false,"kernel_name":"DepthwiseConv2dNative_8172484865844151394_6","module_name":"impl.depthwise_conv2d","name":"depthwise_conv2d","outputs":[[{"dtype":"float16","format":"NC1HWC0","name":"y","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}]],"py_module_path":"/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe","socVersion":"Ascend910B"},"platform":"TBE"} trace: Traceback (most recent call last): File "/root/.pycharm_helpers/pydev/pydevd.py", line 1477, in _exec pydev_imports.execfile(file, globals, locals) # execute the script File "/root/.pycharm_helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile python-BaseException exec(compile(contents+"\n", file, 'exec'), glob, loc) File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 56, in <module> test() File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 50, in test featuremap = op(data) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__ output = self.run_construct(cast_inputs, kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct output = self.construct(*cast_inputs, **kwargs) File "/home/HRBEU-MedAI/zhangximing/autodeeplab-mindspore-zhangximing/src/models/test.py", line 34, in construct return self.op(x) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__ output = self.run_construct(cast_inputs, kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct output = self.construct(*cast_inputs, **kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/layer/container.py", line 235, in construct input_data = cell(input_data) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 365, in __call__ output = self.run_construct(cast_inputs, kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/cell.py", line 323, in run_construct output = self.construct(*cast_inputs, **kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/nn/layer/conv.py", line 258, in construct output = self.conv2d(x, self.weight) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/ops/primitive.py", line 188, in __call__ return _run_op(self, self.name, args) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/common/api.py", line 75, in wrapper results = fn(*arg, **kwargs) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/ops/primitive.py", line 556, in _run_op output = real_run_op(obj, op_name, args) RuntimeError: mindspore/ccsrc/backend/kernel_compiler/tbe/tbe_kernel_parallel_build.cc:99 TbeOpParallelBuild] task compile Failed, task id:8, cause:TBEException:ERROR: [WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:30: DeprecationWarning: register is expired, please replace it with the func register_op_compute [WARNING][2021-04-10 21:57:45,962]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/abs.py:58: DeprecationWarning: te.utils.para_check is expired, please replace it with tbe.common.utils.para_check [WARNING][2021-04-10 21:57:46,052]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:97: DeprecationWarning: check_input_type in topi is expired, please do not use it [WARNING][2021-04-10 21:57:46,131]/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/util/util_frac_z.py:18: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3,and in 3.9 it will stop working [WARNING][2021-04-10 21:57:46,344]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/lang/cce/te_compute/depthwise_conv2d_compute.py:65: DeprecationWarning: te.lang.cce.te_compute.depthwise_conv2d_compute is expired, please replace it with the func tbe.dsl.compute.depthwise_conv2d_compute [WARNING][2021-04-10 21:57:46,345]/root/archiconda3/envs/wks/lib/python3.7/site-packages/topi/cce/util.py:868: DeprecationWarning: check_load3d_w_out_1_support in topi is expired, please do not use it [WARNING][2021-04-10 21:57:46,359]/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/utils/cce.py:37: DeprecationWarning: te.lang.cce.auto_schedule is deprecated, please replace it with the tbe.dsl.auto_schedule Traceback (most recent call last): File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 129, in build_op res = op_func(*inputs_args, *outputs_args, *attrs_args, kernel_name=kernel_name) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/tbe/common/utils/para_check.py", line 529, in _in_wrapper return func(*args, **kwargs) File "/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe/impl/depthwise_conv2d.py", line 414, in depthwise_conv2d tvm.build_module.build(sch, tensor_list, "cce", name=kernel_name) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 900, in build evaluates=evaluates) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/build_module.py", line 610, in lower return cce_lower(sch, args, name, binds, simple_mode, evaluates) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 281, in cce_lower arg_list, cfg, evaluates, multi_process) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/cce_build_module.py", line 80, in cce_static_lower stmt = ir_pass.EmitInsn(stmt) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/te/tvm/_ffi/_ctypes/function.py", line 209, in __call__ raise get_last_ffi_error() tvm._ffi.base.TVMError: Traceback (most recent call last): [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c] [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec] [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c] [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c] [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4] [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4] [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70] [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c] [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c] File "tilefy_pass.cc", line 2163 TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)]) During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 174, in <module> result = compile_with_json(in_args) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 168, in compile_with_json ret = build_op(op_build, json_str, None) File "/root/archiconda3/envs/wks/lib/python3.7/site-packages/mindspore/_extends/parallel_compile/tbe_compiler/compiler.py", line 135, in build_op raise RuntimeError(e) RuntimeError: Traceback (most recent call last): [bt] (8) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14b117c) [0xffff76dbc17c] [bt] (7) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate_(tvm::ir::ProducerConsumer const*, tvm::Stmt const&)+0x6c) [0xffff769230ec] [bt] (6) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::IRMutator::Mutate(tvm::Stmt)+0x6c) [0xffff765f658c] [bt] (5) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::NodeFunctor<tvm::Stmt (tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*)>::operator()(tvm::runtime::ObjectRef const&, tvm::Stmt const&, tvm::ir::IRMutator*) const+0x5c) [0xffff765f628c] [bt] (4) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(+0x14af8c4) [0xffff76dba8c4] [bt] (3) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::Mutate_(tvm::ir::AttrStmt const*, tvm::Stmt const&)+0x63c) [0xffff76f4efc4] [bt] (2) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(std::_Function_handler<tvm::Stmt (tvm::Stmt), tvm::ir::tileir::Tilefy::str2function_::{lambda(tvm::Stmt)#4}>::_M_invoke(std::_Any_data const&, tvm::Stmt&&)+0x40) [0xffff76c72a70] [bt] (1) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(tvm::ir::tileir::Tilefy::MutateSetFmatrix(tvm::Stmt const&)+0x66c) [0xffff76f35a1c] [bt] (0) /usr/local/Ascend/fwkacllib/lib64/libtvm.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4c) [0xffff7658ce6c] File "tilefy_pass.cc", line 2163 TVMError: [EB0000] Check failed: set_fmatrix_shape_[1] % ow == 0 (24 vs. 0) : W tiling unsupportedCurrent IR Stmt:im2col_row_major.local.L1[((((((i1.c*171072) + (i2.c*144)) + (i3.c*144)) + (i4.c*48)) + (i5.c*16)) + i6.c)] = select((((((i4.c*2) + floordiv(i2.c, 97)) < 2) || (((i5.c*2) + floormod(i2.c, 97)) < 2)) || (98 < ((i5.c*2) + floormod(i2.c, 97)))), 0h, fmap.local.L1[((((((i1.c*24832) + (i4.c*3104)) + (i5.c*32)) + (i2.c*16)) + i6.c) - 3136)]) input_args: {"SocInfo":{"autoTilingMode":"NO_TUNE","coreNum":"","coreType":"","l1Fusion":"false","l2Fusion":"false","l2Mode":"2","op_debug_level":"","op_impl_mode":"","op_impl_mode_list":[],"socVersion":"Ascend910B"},"impl_path":"","op_info":{"Type":"DepthwiseConv2dNative","attr_desc":[[1,1,1,1],[1,1,2,2],[2,2,2,2],"NCHW",0],"attrs":[{"name":"stride","valid":true,"value":[1,1,1,1]},{"name":"dilation","valid":true,"value":[1,1,2,2]},{"name":"pad_list","valid":true,"value":[2,2,2,2]},{"name":"format","valid":true,"value":"NCHW"},{"name":"offset_a","valid":true,"value":0}],"full_name":"Default/DepthwiseConv2dNative-op8","gen_model":"single","graph_id":1,"inputs":[[{"dtype":"float16","format":"NC1HWC0","name":"x_0","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}],[{"dtype":"float16","format":"C1HWNCoC0","name":"filter_0","ori_format":"NCHW","ori_shape":[1,320,3,3],"param_type":"required","range":[[1,1],[320,320],[3,3],[3,3]],"shape":[20,3,3,1,16,16],"valid":true}],[{"name":"bias_optional_2","valid":false}],[{"name":"offset_w_optional_2","valid":false}]],"is_dynamic_shape":false,"kernel_name":"DepthwiseConv2dNative_8172484865844151394_6","module_name":"impl.depthwise_conv2d","name":"depthwise_conv2d","outputs":[[{"dtype":"float16","format":"NC1HWC0","name":"y","ori_format":"NCHW","ori_shape":[10,320,97,97],"param_type":"required","range":[[10,10],[320,320],[97,97],[97,97]],"shape":[10,20,97,97,16],"valid":true}]],"py_module_path":"/usr/local/Ascend/opp/op_impl/built-in/ai_core/tbe","socVersion":"Ascend910B"},"platform":"TBE"} trace: # ``` ## Special notes for this issue 这是gitee的issues链接： https://gitee.com/mindspore/mindspore/issues/I3I88M
zhangximing added the
mindspore-assistant
label 3 years ago
zhangximing added the
技术
label 3 years ago
zhangximing removed the
技术
label 3 years ago
zhangximing commented 3 years ago
Poster

问题尚未解决，bug尚未修复。目前使用替代方案来取代原来的nn.Conv2d，但是时间复杂度有点高。

class DilSepCov(nn.Cell):
    def __init__(self,
                 channels,
                 kernel_size,
                 stride,
                 pad_mode='pad',
                 padding=0,
                 dilation=1,
                 has_bias=False,
                 weight_init='HeNormal'
                 ):
        super(DilSepCov, self).__init__()
        self.channels = channels
        self.ops = nn.CellList([])
        for i in range(channels):
            self.ops.append(nn.Conv2d(1, 1, kernel_size=kernel_size, stride=stride, pad_mode=pad_mode,
                                      padding=padding, dilation=dilation,  has_bias=has_bias, weight_init=weight_init))
        self.concat = ops.Concat(axis=1)

    def construct(self, x):
        output = ()
        for i in range(1, self.channels):
            output += (self.ops[i](x[:, i:i+1:1, :, :]), )
        output = self.concat(output)
        return output
问题尚未解决，bug尚未修复。目前使用替代方案来取代原来的nn.Conv2d，但是时间复杂度有点高。 ``` class DilSepCov(nn.Cell): def __init__(self, channels, kernel_size, stride, pad_mode='pad', padding=0, dilation=1, has_bias=False, weight_init='HeNormal' ): super(DilSepCov, self).__init__() self.channels = channels self.ops = nn.CellList([]) for i in range(channels): self.ops.append(nn.Conv2d(1, 1, kernel_size=kernel_size, stride=stride, pad_mode=pad_mode, padding=padding, dilation=dilation, has_bias=has_bias, weight_init=weight_init)) self.concat = ops.Concat(axis=1) def construct(self, x): output = () for i in range(1, self.channels): output += (self.ops[i](x[:, i:i+1:1, :, :]), ) output = self.concat(output) return output ```
zhangximing commented 2 years ago
Poster

mindspore r1.5版本已修复此bug，issue关闭

mindspore r1.5版本已修复此bug，issue关闭
zhangximing closed this issue 2 years ago
Sign in to join this conversation.
No Branch/Tag Specified
Clear branch or tag
Branches Tags
Labels
Apply labels
Clear labels
mindspore-assistant modelArts平台问题
使用平台过程中，发生错误 其他 平台功能优化
平台使用流程不符合逻辑，或平台缺少某些通用功能 技术 社区平台问题
使用平台过程中，发生错误
No Label
mindspore-assistant
modelArts平台问题
其他
平台功能优化
技术
社区平台问题
Milestone
Set milestone
Clear milestone
No items
No Milestone
Assignees
Assign users
Clear assignees
No Assignees
1 Participants
Notifications
Due Date

No due date set.

Dependencies

This issue currently doesn't have any dependencies.

Write Preview
Loading…
Cancel
Save
There is no content yet.
Delete Branch '%!s(MISSING)'

Deleting a branch is permanent. It CANNOT be undone. Continue?

No
Yes
Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit