一、问题现象(附报错日志上下文):
cdws@dev-52673460-bb97-4046-871c-ae6387bdb329-jc6cr:~$ python
Python 3.7.5 (default, Oct 14 2022, 06:42:07)
[GCC 7.5.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> import torch_npu
[W OperatorEntry.cpp:121] Warning: Overriding a previously registered kernel for the same operator and the same dispatch key
operator: aten::_has_compatible_shallow_copy_type(Tensor self, Tensor from) -> (bool)
registered at /tmp/pytorch_v1.8.1/build/aten/src/ATen/RegisterSchema.cpp:20
dispatch key: Math
previous kernel: registered at /tmp/pytorch_v1.8.1/build/aten/src/ATen/RegisterMath.cpp:5686
new kernel: registered at /tmp/pytorch/torch_npu/csrc/aten/ops/HasCompatibleShallowCopyType.cpp:37 (function registerKernel)
>>>
二、软件版本:
-- CANN 版本 (e.g., CANN 3.0.x,5.x.x): nnae 5.1.RC
--Tensorflow/Pytorch/MindSpore 版本: PyTorch 1.8.1
--Python 版本 (e.g., Python 3.7.5): Python 3.7.5
-- MindStudio版本 (e.g., MindStudio 2.0.0 (beta3)): 无
--操作系统版本 (e.g., Ubuntu 18.04): Ubuntu 18.04.6(容器)
三、测试步骤:
依次导入torch、torch_npu即可见上述现象
四、日志信息:
见 一 中信息
五、附自定义Dockerfile内容
FROM ubuntu:18.04
ARG HOST_ASCEND_BASE=/usr/local/Ascend
ARG NNAE_PATH=/usr/local/Ascend/nnae/latest
ARG INSTALL_ASCEND_PKGS_SH=install_ascend_pkgs.sh
WORKDIR /tmp
# 更新软件源
RUN apt update && \
apt install -y --no-install-recommends ca-certificates wget && \
cp -a /etc/apt/sources.list /etc/apt/sources.list.bak && \
wget --no-check-certificate -O /etc/apt/sources.list https://repo.huaweicloud.com/repository/conf/Ubuntu-Ports-bionic.list && \
apt update && \
apt upgrade -y
# 更新软件源
RUN apt install -y --no-install-recommends vim sudo bzip2 wget make tar curl g++ pkg-config unzip numactl gfortran patch less\
libopenblas-dev libblas3 liblapack3 liblapack-dev libblas-dev libhdf5-dev libffi-dev libicu60 libxml2 libbz2-dev libssl-dev libfreetype6-dev libpng-dev libgl1-mesa-glx \
gcc git ssh openssh-server htop tmux inetutils-ping && \
apt clean && \
rm -rf /var/lib/apt/lists/*
ENV LD_LIBRARY_PATH=/usr/local/gcc7.3.0/lib64:${LD_LIBRARY_PATH}
# 安装cmake
COPY cmake-3.15.7.tar.gz ./
RUN tar -zxf cmake-3.15.7.tar.gz && cd cmake-3.15.7 && \
./bootstrap && make -j 96 && make install && \
ln -s /usr/local/cmake/bin/cmake /usr/bin/cmake
# 安装python和pip
COPY Python-3.7.5.tar.xz ./
RUN tar -xf Python-3.7.5.tar.xz && cd Python-3.7.5 && ./configure --prefix=/usr/local/python3.7.5 --enable-shared && \
make -j 96 && make install && \
ln -sf /usr/local/python3.7.5/bin/python3 /usr/bin/python && \
ln -sf /usr/local/python3.7.5/bin/python3 /usr/bin/python3 && \
ln -sf /usr/local/python3.7.5/bin/pip3 /usr/bin/pip && \
ln -sf /usr/local/python3.7.5/bin/pip3 /usr/bin/pip3 && \
cd .. && \
rm -rf Python*
ENV LD_LIBRARY_PATH=/usr/local/python3.7.5/lib:$LD_LIBRARY_PATH
ENV PATH=/usr/local/python3.7.5/bin:$PATH
# 配置python pip源
RUN mkdir -p ~/.pip \
&& echo '[global] \n\
index-url=http://pypi.doubanio.com/simple/\n\
trusted-host=pypi.doubanio.com' >> ~/.pip/pip.conf
# HwHiAiUser, hwMindX
RUN useradd -d /home/hwMindX -u 9000 -m -s /bin/bash hwMindX && \
useradd -d /home/HwHiAiUser -u 1000 -m -s /bin/bash HwHiAiUser && \
usermod -a -G HwHiAiUser hwMindX
# python包
RUN pip3 install pip -U && \
pip3 install numpy && \
pip3 install decorator && \
pip3 install sympy==1.4 && \
pip3 install cffi==1.12.3 && \
pip3 install pyyaml && \
pip3 install pathlib2 && \
pip3 install grpcio && \
pip3 install grpcio-tools && \
pip3 install protobuf && \
pip3 install scipy && \
pip3 install requests && \
pip3 install attrs && \
pip3 install wheel && \
pip3 install Pillow==6.2.2 && \
pip3 install wheel==0.32.1 && \
pip3 install setuptools==49.1.0 && \
pip3 install matplotlib==3.2.2 && \
pip3 install opencv-python==4.5.5.64 && \
pip3 install sklearn==0.0 && \
pip3 install pandas==1.0.5 && \
pip3 install pycocotools==2.0.1 && \
pip3 install tables==3.6.1 && \
pip3 install mmcv==0.2.14 && \
pip3 install lxml==4.5.2 && \
pip3 install easydict==1.9 && \
pip3 install expecttest && \
pip3 install jupyter && \
pip3 install jupyterlab && \
rm -rf /root/.cache/pip
# 拷贝相关文件
COPY . ./
# Ascend包
RUN bash $INSTALL_ASCEND_PKGS_SH
# PyTorch安装
RUN cd pytorch_* && \
python3 setup.py install
# PyTorch_NPU 安装
RUN cd pytorch && \
bash ci/build.sh --python=3.7 && \
pip3 install --upgrade dist/torch_npu*.whl && \
pip3 install torchvision==0.9.1
ENV LD_LIBRARY_PATH=$NNAE_PATH/fwkacllib/lib64/:/usr/local/python3.7.5/lib/python3.7/site-packages/torch:/usr/local/python3.7.5/lib/python3.7/site-packages/torch/lib:/usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/common/:/usr/local/Ascend/driver/lib64/driver/:/usr/local/Ascend/add-ons/:/usr/lib/aarch64_64-linux-gnu:$LD_LIBRARY_PATH
ENV PATH=$PATH:$NNAE_PATH/fwkacllib/ccec_compiler/bin/:$NNAE_PATH/toolkit/tools/ide_daemon/bin/
ENV ASCEND_OPP_PATH=$NNAE_PATH/opp/
ENV OPTION_EXEC_EXTERN_PLUGIN_PATH=$NNAE_PATH/fwkacllib/lib64/plugin/opskernel/libfe.so:$NNAE_PATH/fwkacllib/lib64/plugin/opskernel/libaicpu_engine.so:$NNAE_PATH/fwkacllib/lib64/plugin/opskernel/libge_local_engine.so
ENV PYTHONPATH=$NNAE_PATH/fwkacllib/python/site-packages/:$NNAE_PATH/fwkacllib/python/site-packages/auto_tune.egg/auto_tune:$NNAE_PATH/fwkacllib/python/site-packages/schedule_search.egg:$PYTHONPATH
ENV ASCEND_AICPU_PATH=$NNAE_PATH
# 日志包安装
RUN tar -zxf dllogger-1.0.0.tar.gz
RUN cd dllogger-1.0.0/ && \
python3 setup.py build && \
python3 setup.py install
# 清理工作
RUN rm -f /etc/ascend_install.info && \
rm -rf /tmp/*
USER hwMindX
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。
Please add labels , also you can visit https://gitee.com/ascend/community/blob/master/labels.md to find more.
为了让代码尽快被审核,请您为Issue打上标签,打上标签的Issue可以直接推送给责任人进行审核。
更多的标签可以查看https://gitee.com/ascend/community/blob/master/labels.md
以模型训练相关代码提交为例,如果你提交的是模型训练代码,你可以这样评论:
//train/model
另外你还可以给这个Issue标记类型,例如是bugfix或者是特性需求:
//kind/bug or //kind/feature
恭喜你,你已经学会了使用命令来打标签,接下来就在下面的评论里打上标签吧!
//train/model
正常提示信息,不影响使用
登录 后才可以发表评论