使用PC工作站（系统ubuntu22.04）训练的YOLO模型，训练使用的英伟达GTX4090，没有安装昇腾的板卡，想在PC上将模型转换为om模型放到Atlas 200I DK A2开发板上运行推理程序。但是转换时总是错误，无法运行atc工具，使用conda建立一个专属环境，搭建了CANN7.0.0RC1环境，仍然无法运行atc工具。使用官方网站上下载的yolov5s的onnx模型也无法成功。然后再Atlas 200I DK A2开发板上直接使用atc工具转换yolov5s案例模型，也不能成功。如何解决？因为没有昇腾的训练板卡，所以无法直接得到能在开发板上运行的模型，所以模型转换这个环节卡住了，请各位大佬给提供一个思路。查了各种资料，始终没能解决。Atlas 200I DK A2开发板上的版本是23.0.rc1，CANN7.0.0RC1，后期一直没有提供对应的高版本的系统，请问如果要自己升级，应该使用什么版本？
atc --model=yolov5s.onnx --framework=5 --output=yolov5s_bs2 --input_format=NCHW --soc_version=Ascend310B4 --input_fp16_nodes="images"
ATC start working now, please wait for a moment.
/home/ubuntu/Ascend/ascend-toolkit/latest/python/site-packages/tbe/dsl/unify_schedule/extract_image_patches_without_cbuf_schedule.py:317: SyntaxWarning: "is not" with a literal. Did you mean "!="?
if _ is not 1:
run stack spill compile error: In file included from <built-in>:1:
In file included from /home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_runtime_wrapper.h:39:
/home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_vector_types.h:19:2: warning: The vector size is manually reset. [-W#warnings]
#warning The vector size is manually reset.
^
In file included from <built-in>:1:
In file included from /home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_runtime_wrapper.h:68:
/home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_vector_intrinsics.h:20:10: fatal error: 'type_traits' file not found
#include <type_traits>
^~~~~~~~~~~~~
1 warning and 1 error generated.
这是在Atlas 200I DK A2开发板上运行atc工具提供的错误提示：
tc --model=yolo5s.onnx --framework=5 --output=yolo5s_1 --input_format=NCHW --soc_version=Ascend310B4 ATC start working now, please wait for a moment. ... ATC run failed, Please check the detail log, Try 'atc --help' for more information E10001: Value [yolo5s.onnx] for parameter [--model] is invalid. Reason: realpath error, errmsg:No such file or directory Solution: Try again with a valid argument. TraceBack (most recent call last): [Check][InputPath]model file yolo5s.onnx not found!![FUNC:CheckFlags][FILE:main_impl.cc][LINE:682] (base) root@davinci-mini:/home/HwHiAiUser/samples/notebooks/01-yolov8# atc --model=yolov5s.onnx --framework=5 --output=yolo5s_1 --input_format=NCHW --soc_version=Ascend310B4 ATC start working now, please wait for a moment. Exception in thread Thread-1: Traceback (most recent call last): File "/usr/local/miniconda3/lib/python3.9/threading.py", line 954, in _bootstrap_inner self.run() File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/utils/multiprocess_util.py", line 91, in run key, func, args, kwargs = self.task_q.get(timeout=TIMEOUT) File "<string>", line 2, in get File "/usr/local/miniconda3/lib/python3.9/multiprocessing/managers.py", line 809, in _callmethod kind, result = conn.recv() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 255, in recv buf = self._recv_bytes() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 419, in _recv_bytes buf = self._recv(4) File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 384, in _recv chunk = read(handle, remaining) ConnectionResetError: [Errno 104] Connection reset by peer Traceback (most recent call last): File "/usr/local/miniconda3/lib/python3.9/multiprocessing/util.py", line 300, in _run_finalizers finalizer() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/util.py", line 224, in __call__ res = self._callback(*self._args, **self._kwargs) File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/route.py", line 51, in wrapper return func(cls, *args, **kwargs) File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/route.py", line 207, in finalize cls.bank_process_pool.close() AttributeError: 'NoneType' object has no attribute 'close' ... ATC run failed, Please check the detail log, Try 'atc --help' for more information E40021: Failed to compile Op [Conv_235,[Conv_235,Conv_235]]. (oppath: [Compile /usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py failed with errormsg/stack: File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 635, in SocketClient s.connect(address) ConnectionRefusedError: [Errno 111] Connection refused ], optype: [Conv2D]) Solution: See the host log for details, and then check the Python stack where the error log is reported. TraceBack (most recent call last): Compile op[Conv_235,[Conv_235,Conv_235]] failed, oppath[/usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py], optype[Conv2D], taskID[275]. Please check op's compilation error message.[FUNC:ReportBuildErrMessage][FILE:fusion_manager.cc][LINE:763] Failed to compile Op [Conv_198,[Conv_198,Conv_198]]. (oppath: [Compile /usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py failed with errormsg/stack: File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 635, in SocketClient s.connect(address)
点赞
收藏
回复
分享
举报
已解决
atc --model=yolov5s.onnx --framework=5 --output=yolov5s_bs2 --input_format=NCHW --soc_version=Ascend310B4 --input_fp16_nodes="images"
ATC start working now, please wait for a moment.
/home/ubuntu/Ascend/ascend-toolkit/latest/python/site-packages/tbe/dsl/unify_schedule/extract_image_patches_without_cbuf_schedule.py:317: SyntaxWarning: "is not" with a literal. Did you mean "!="?
if _ is not 1:
run stack spill compile error: In file included from <built-in>:1:
In file included from /home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_runtime_wrapper.h:39:
/home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_vector_types.h:19:2: warning: The vector size is manually reset. [-W#warnings]
#warning The vector size is manually reset.
^
In file included from <built-in>:1:
In file included from /home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_runtime_wrapper.h:68:
/home/ubuntu/Ascend/ascend-toolkit/7.0.RC1/x86_64-linux/ccec_compiler/lib/clang/15.0.5/include/__clang_cce_vector_intrinsics.h:20:10: fatal error: 'type_traits' file not found
#include <type_traits>
^~~~~~~~~~~~~
1 warning and 1 error generated.
这是在Atlas 200I DK A2开发板上运行atc工具提供的错误提示：
tc --model=yolo5s.onnx --framework=5 --output=yolo5s_1 --input_format=NCHW --soc_version=Ascend310B4 ATC start working now, please wait for a moment. ... ATC run failed, Please check the detail log, Try 'atc --help' for more information E10001: Value [yolo5s.onnx] for parameter [--model] is invalid. Reason: realpath error, errmsg:No such file or directory Solution: Try again with a valid argument. TraceBack (most recent call last): [Check][InputPath]model file yolo5s.onnx not found!![FUNC:CheckFlags][FILE:main_impl.cc][LINE:682] (base) root@davinci-mini:/home/HwHiAiUser/samples/notebooks/01-yolov8# atc --model=yolov5s.onnx --framework=5 --output=yolo5s_1 --input_format=NCHW --soc_version=Ascend310B4 ATC start working now, please wait for a moment. Exception in thread Thread-1: Traceback (most recent call last): File "/usr/local/miniconda3/lib/python3.9/threading.py", line 954, in _bootstrap_inner self.run() File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/utils/multiprocess_util.py", line 91, in run key, func, args, kwargs = self.task_q.get(timeout=TIMEOUT) File "<string>", line 2, in get File "/usr/local/miniconda3/lib/python3.9/multiprocessing/managers.py", line 809, in _callmethod kind, result = conn.recv() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 255, in recv buf = self._recv_bytes() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 419, in _recv_bytes buf = self._recv(4) File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 384, in _recv chunk = read(handle, remaining) ConnectionResetError: [Errno 104] Connection reset by peer Traceback (most recent call last): File "/usr/local/miniconda3/lib/python3.9/multiprocessing/util.py", line 300, in _run_finalizers finalizer() File "/usr/local/miniconda3/lib/python3.9/multiprocessing/util.py", line 224, in __call__ res = self._callback(*self._args, **self._kwargs) File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/route.py", line 51, in wrapper return func(cls, *args, **kwargs) File "/usr/local/Ascend/ascend-toolkit/latest/python/site-packages/tbe/common/repository_manager/route.py", line 207, in finalize cls.bank_process_pool.close() AttributeError: 'NoneType' object has no attribute 'close' ... ATC run failed, Please check the detail log, Try 'atc --help' for more information E40021: Failed to compile Op [Conv_235,[Conv_235,Conv_235]]. (oppath: [Compile /usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py failed with errormsg/stack: File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 635, in SocketClient s.connect(address) ConnectionRefusedError: [Errno 111] Connection refused ], optype: [Conv2D]) Solution: See the host log for details, and then check the Python stack where the error log is reported. TraceBack (most recent call last): Compile op[Conv_235,[Conv_235,Conv_235]] failed, oppath[/usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py], optype[Conv2D], taskID[275]. Please check op's compilation error message.[FUNC:ReportBuildErrMessage][FILE:fusion_manager.cc][LINE:763] Failed to compile Op [Conv_198,[Conv_198,Conv_198]]. (oppath: [Compile /usr/local/Ascend/ascend-toolkit/7.0.RC1/opp/built-in/op_impl/ai_core/tbe/impl/conv2d.py failed with errormsg/stack: File "/usr/local/miniconda3/lib/python3.9/multiprocessing/connection.py", line 635, in SocketClient s.connect(address)