[图片]
2023-03-27 17:27:17.345802: [cnmlError] Internal type: MLU::unary, Op name: UNSET dimension mismatch,output shape is wrong, the given shape is:(n=1, c=2,h=1,w=1, d=0, t=0, data in array:{1, 1, 1, 2, } , but the expected shape is:(n=2, c=1,h=1,w=1, d=0, t=0, data in array:{2, 1, 1, 1, } 2023-03-27 17:27:17.345911: [cnmlError] Internal type: MLU::unary, Op name: UNSET, Op update tensor shape failed[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/abs_internal.cpp][line: 30][cnml_abs_internal][thread:139733666461504][process:97723]: CNML error: CNML_STATUS_INVALIDPARAM[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 460][abs][thread:139733666461504][process:97723]: OpMethods::abs Op running on CPU!2023-03-27 17:27:17.358653: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET dimension mismatch,output shape is wrong, the given shape is:(n=1, c=2,h=1,w=1, d=0, t=0, data in array:{1, 1, 1, 2, } , but the expected shape is:(n=2, c=1,h=1,w=1, d=0, t=0, data in array:{2, 1, 1, 1, } 2023-03-27 17:27:17.358730: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op update tensor shape failed[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/eq_internal.cpp][line: 35][cnml_eq_internal][thread:139733666461504][process:97723]: CNML error: CNML_STATUS_INVALIDPARAM[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 1349][eq][thread:139733666461504][process:97723]: OpMethods::eq Op running on CPU![WARNING][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/mul_internal.cpp][line: 12][cnml_mul_internal][thread:139733666461504][process:97723]: self datatype: bool, other datatype: floatInconsistant datatype of self and other will cause the failuare on cnml op forward function.2023-03-27 17:27:17.369895: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op should follow these rules: input1_n == input2_n, now, input1_n=2, input2_n=1op_id:102023-03-27 17:27:17.369962: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op dimension mismatch[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/ne_internal.cpp][line: 35][cnml_ne_internal][thread:139733666461504][process:97723]: CNML error: CNML_STATUS_INVALIDPARAM[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 495][ne][thread:139733666461504][process:97723]: OpMethods::ne Op running on CPU![WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 638][bitwise_and_out][thread:139733666461504][process:97723]: OpMethods::bitwise_and_out Op running on CPU![ERROR][/torch/catch/torch_mlu/csrc/aten/core/tensor_impl.cpp][line: 873][cpu_data][thread:139733666461504][process:97723]:展开
您好,有以下两个信息请确认:1)“打印量化模型就会报错,不打印就会正常输出结果”,这是指在用量化后的模型,在mlu上做逐层推理吗?打印模型的操作是指?2)2楼的报错是在什么操作下触发的?从报错信息看是算子维度不匹配以及存在算子不支持的情况展开
https://github.com/bubbliiiing/deeplabv3-plus-pytorch 这个算法
2023-03-27 17:27:17.345802: [cnmlError] Internal type: MLU::unary, Op name: UNSET dimension mismatch,output shape is wrong, the given shape is:(n=1, c=2,h=1,w=1, d=0, t=0, data in array:{1, 1, 1, 2, } , but the expected shape is:(n=2, c=1,h=1,w=1, d=0, t=0, data in array:{2, 1, 1, 1, }
2023-03-27 17:27:17.345911: [cnmlError] Internal type: MLU::unary, Op name: UNSET, Op update tensor shape failed
[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/abs_internal.cpp][line: 30][cnml_abs_internal][thread:139733666461504][process:97723]:
CNML error: CNML_STATUS_INVALIDPARAM
[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 460][abs][thread:139733666461504][process:97723]: OpMethods::abs Op running on CPU!
2023-03-27 17:27:17.358653: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET dimension mismatch,output shape is wrong, the given shape is:(n=1, c=2,h=1,w=1, d=0, t=0, data in array:{1, 1, 1, 2, } , but the expected shape is:(n=2, c=1,h=1,w=1, d=0, t=0, data in array:{2, 1, 1, 1, }
2023-03-27 17:27:17.358730: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op update tensor shape failed
[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/eq_internal.cpp][line: 35][cnml_eq_internal][thread:139733666461504][process:97723]:
CNML error: CNML_STATUS_INVALIDPARAM
[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 1349][eq][thread:139733666461504][process:97723]: OpMethods::eq Op running on CPU!
[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/mul_internal.cpp][line: 12][cnml_mul_internal][thread:139733666461504][process:97723]:
self datatype: bool, other datatype: float
Inconsistant datatype of self and other will cause the failuare on cnml op forward function.
2023-03-27 17:27:17.369895: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op should follow these rules: input1_n == input2_n, now, input1_n=2, input2_n=1op_id:10
2023-03-27 17:27:17.369962: [cnmlError] Internal type: MLU::dyadic_all, Op name: UNSET, Op dimension mismatch
[ERROR][/torch/catch/torch_mlu/csrc/aten/operators/cnml/internal/ne_internal.cpp][line: 35][cnml_ne_internal][thread:139733666461504][process:97723]:
CNML error: CNML_STATUS_INVALIDPARAM
[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 495][ne][thread:139733666461504][process:97723]: OpMethods::ne Op running on CPU!
[WARNING][/torch/catch/torch_mlu/csrc/aten/operators/op_methods.cpp][line: 638][bitwise_and_out][thread:139733666461504][process:97723]: OpMethods::bitwise_and_out Op running on CPU!
[ERROR][/torch/catch/torch_mlu/csrc/aten/core/tensor_impl.cpp][line: 873][cpu_data][thread:139733666461504][process:97723]:
请登录后评论