在ml-engine上使用文件结构训练模型时,我正在尝试安装软件包:
/ml_engine
setup.py
/trainer
__init__.py
task.py
model.py
我已经运行:python setup.py sdist
创建一个包,并将该包上传到存储桶中。
最后,
gcloud alpha ml-engine versions create m_0_03 \
--model {model_name} \
--origin gs://{model_location} \
--runtime-version 1.10 \
--framework SCIKIT_LEARN \
--python-version 3.5 \
--package-uris gs://{package_location} \
--model-class trainer.model \
但是,它给出了一个错误:
Creating version (this might take a few minutes)......
.......................................................................................................................................................................................................................................................failed.
ERROR: (gcloud.alpha.ml-engine.versions.create) Bad model detected with error:
"Failed to load model: User-provided package dist.tar.gz failed to install: Command '['python-default', '-m', 'pip', 'install', '--target=/tmp/custom_lib', '--no-cache-dir', '/tmp/custom_code/dist.tar.gz']' returned non-zero exit status 1 (Error code: 0)"
我尝试下载相同的软件包,并成功在本地安装。
答案 0 :(得分:0)
Google当前在创建自定义ML模型版本时遇到问题。