Python版本：3.8
CANN版本：8.0.rc1
操作系统版本：ubuntu_18.04
mindspore 2.3.0rc1

# 拉取镜像 + 修改 /etc/hosts 加入 58.48.42.198 swr.cn-central-221.ovaijisuan.com + 修改 /etc/docker/daemon.json 如下： ```json { "registry-mirrors": [ ...[没有就不写这个字段] ], "insecure-registries": [ "swr.cn-central-221.ovaijisuan.com" ] } ``` + 重启docker ```shell systemctl daemon-reload systemctl restart docker ``` + 拉取镜像 ```shell docker pull swr.cn-central-221.ovaijisuan.com/wh-aicc-fae/openmind_v0.5-mindspore_2.3.0rc1-cann_8.0.rc1-py_3.8-ubuntu_18.04-aarch64:v0.5.1 ``` + 无法正常拉取情况 ```shell wget https://obs-whaicc-fae-public.obs.cn-central-221.ovaijisuan.com/share/openmind/v0.5.1/openmind_v0.5-mindspore_2.3.0rc1-cann_8.0.rc1-py_3.8-ubuntu_18.04-aarch64.tar docker load -i ./openmind_v0.5-mindspore_2.3.0rc1-cann_8.0.rc1-py_3.8-ubuntu_18.04-aarch64.tar ``` # 启动指令示例(自己加一下 【反斜杠】 我写了但不显示) ```shell docker run -it --ipc=host \ --device=/dev/davinci0 \ --device=/dev/davinci1 \ --device=/dev/davinci2 \ --device=/dev/davinci3 \ --device=/dev/davinci4 \ --device=/dev/davinci5 \ --device=/dev/davinci6 \ --device=/dev/davinci7 \ --device=/dev/davinci_manager \ --device=/dev/devmm_svm \ --device=/dev/hisi_hdc \ -e LD_LIBRARY_PATH=/usr/local/Ascend/driver/lib64/driver:/usr/local/Ascend/driver/lib64/common:/usr/local/lib \ -e LD_PRELOAD=/usr/lib/aarch64-linux-gnu/libgomp.so.1:/usr/local/lib/python3.8/dist-packages/torch/lib/../../torch.libs/libgomp-6e1a1d1b.so.1.0.0 \ -e OPENMIND_FRAMEWORK="pt" \ -v /usr/local/Ascend/driver:/usr/local/Ascend/driver \ -v /home/aicc:/home/aicc \ -v /usr/local/Ascend/add-ons/:/usr/local/Ascend/add-ons/ \ -v /usr/local/sbin/npu-smi:/usr/local/sbin/npu-smi \ -v /usr/bin/msnpureport:/usr/bin/msnpureport swr.cn-central-221.ovaijisuan.com/wh-aicc-fae/openmind_v0.5-mindspore_2.3.0rc1-cann_8.0.rc1-py_3.8-ubuntu_18.04-aarch64:v0.5.1 /bin/bash ``` # 加载环境变量 ```shell source /usr/local/Ascend/ascend-toolkit/set_env.sh ``` # 测试代码 + pytorch测试代码 ```python from openmind import pipeline import ssl ssl._create_default_https_context = ssl._create_unverified_context question_answerer = pipeline( "question-answering", model="openmind/roberta_base_squad2_pt", device="npu:0", ) outputs = question_answerer(question="Is Shakespeare British?", context="Yes, Shakespeare is British.") print(outputs) ``` + MindSpore测试代码 1. 修改环境变量: `export OPENMIND_FRAMEWORK="ms"` 2. 测试代码 ```python # Use pipeline from openmind import pipeline pipe = pipeline("text-generation", model="MindSpore-Lab/telechat_7b", framework="ms", trust_remote_code=True) pipe("Give me some advice on how to stay healthy.") ```

在其他AICC使用本镜像时，
1） 在本地arm宿主机通过docker pull 拉取上面镜像到本地（即执行docker pull remote_image_address）
2） 用docker tag 将局点信息和组织名替换成对应版本（即执行 docker tag local_image_address remote_image_address），
3） 用docker push 将修改后的镜像名称推送到局点的swr服务中（即执行docker push remote_image_address）

若您有任何疑虑，您可以在这里反馈您的建议。