您好,我在使用CANN7.0RC1对模型(经过量化后的Llama3.1onnx模型)进行转换时,报错信息如下:
ATC run failed, Please check the detail log, Try 'atc --help' for more information
E79999: Inner Error!
E79999 onnx_plugin op_name[/model/Trilu], now only support one input, not support k input[FUNC:ParseParamsTrilu][FILE:trilu_plugin.cc][LINE:38]
TraceBack (most recent call last):
Model parse to graph failed, graph name:llama3.om.[FUNC:ModelParseToGraph][FILE:onnx_parser.cc][LINE:925]
ATC model parse ret fail.[FUNC:ParseGraph][FILE:omg.cc][LINE:780]
其中,模型转换指令为:
atc --framework=5 --model="llama3.onnx" --output="llama3.om" --input_format=ND --input_shape="input_ids:1,1;attention_mask:1,1025;position_ids:1,1;past_key_values:32,2,1,32,1024,128" --log=debug --soc_version=Ascend310B1 --precision_mode=must_keep_origin_dtype
您好,我在使用CANN7.0RC1对模型(经过量化后的Llama3.1onnx模型)进行转换时,报错信息如下:
ATC run failed, Please check the detail log, Try 'atc --help' for more information
E79999: Inner Error!
E79999 onnx_plugin op_name[/model/Trilu], now only support one input, not support k input[FUNC:ParseParamsTrilu][FILE:trilu_plugin.cc][LINE:38]
TraceBack (most recent call last):
Model parse to graph failed, graph name:llama3.om.[FUNC:ModelParseToGraph][FILE:onnx_parser.cc][LINE:925]
ATC model parse ret fail.[FUNC:ParseGraph][FILE:omg.cc][LINE:780]
其中,模型转换指令为:
atc --framework=5 --model="llama3.onnx" --output="llama3.om" --input_format=ND --input_shape="input_ids:1,1;attention_mask:1,1025;position_ids:1,1;past_key_values:32,2,1,32,1024,128" --log=debug --soc_version=Ascend310B1 --precision_mode=must_keep_origin_dtype