onnx转om报错
收藏回复举报
onnx转om报错
t('forum.solved') 已解决
发表于2025-01-07 18:58:31
0 查看

【行业】:政企

【服务器型号】:Atlas 800T A2

【版本信息】:

--CANN版本:CANN 7.0.0

--Python版本:Python 3.9.18

--操作系统版本:Ubuntu 20.04.6 LTS

1、onnx模型推理正常

2、atc转om模型报错

转模型命令:

atc --framework=5 --model=video_model.onnx --input_format=NCHW --input_shape="input:1,3,160,160" --output=weights --log=debug --soc_version=Ascend910B3 

报错详情

ATC start working now, please wait for a moment. 

... 

ATC run failed, Please check the detail log, Try 'atc --help' for more information 

E10042: GenerateOfflineModel execute failed. 

        TraceBack (most recent call last): 

        op[/Concat], the input shape dims should be equal except merge axis,shapes:[[1, ], [1, 1, ], ]axis:0[FUNC:ConcatInferShapeCommon][FILE:split_combination_ops.cc][LINE:898] 

        Call InferShapeAndType for node:/Concat(ConcatD) failed[FUNC:Infer][FILE:infershape_pass.cc][LINE:119] 

        process pass InferShapePass on node:/Concat failed, ret:4294967295[FUNC:RunPassesOnNode][FILE:base_pass.cc][LINE:571] 

        build graph failed, graph id:0, ret:1343242270[FUNC:BuildModelWithGraphId][FILE:ge_generator.cc][LINE:1615] 

本帖最后由 匿名用户2025/01/07 19:01:17 编辑

我要发帖子