问题描述
我有一个我一直在运行的自定义 gcp-ai-platform 训练作业脚本，我最后一次运行它是在 1 周前。但是，今天相同的训练作业和脚本失败并出现错误。
我试图隔离这个问题，当 Pandas 试图从谷歌云存储读取我的训练集 csv 时，我的代码坏了。我的人工智能平台训练作业和谷歌云桶属于同一个项目。
tf = "gs://bucket_name/train.csv"
train_df = pd.read_csv(tf)
The replica master 0 exited with a non-zero status of 1.
Traceback (most recent call last):
[...]
File "/opt/conda/lib/python3.7/importlib/__init__.py",line 127,in import_module
return _bootstrap._gcd_import(name[level:],package,level)
File "<frozen importlib._bootstrap>",line 1006,in _gcd_import
File "<frozen importlib._bootstrap>",line 983,in _find_and_load
File "<frozen importlib._bootstrap>",line 967,in _find_and_load_unlocked
File "<frozen importlib._bootstrap>",line 677,in _load_unlocked
File "<frozen importlib._bootstrap_external>",line 728,in exec_module
File "<frozen importlib._bootstrap>",line 219,in _call_with_frames_removed
File "/opt/conda/lib/python3.7/site-packages/fsspec/__init__.py",line 42,in <module>
entry_points = entry_points()
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 893,in entry_points
return SelectableGroups.load(eps).select(**params)
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 331,in load
ordered = sorted(eps,key=by_group)
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 891,in <genexpr>
dist.entry_points for dist in unique(distributions())
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 517,in entry_points
return EntryPoints._from_text_for(self.read_text('entry_points.txt'),self)
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 244,in _from_text_for
return cls(ep._for(dist) for ep in cls._from_text(text))
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",in <genexpr>
return cls(ep._for(dist) for ep in cls._from_text(text))
File "/opt/conda/lib/python3.7/site-packages/importlib_Metadata/__init__.py",line 255,in <genexpr>
for name,value in values
ValueError: not enough values to unpack (expected 2,got 1)
我也注意到训练作业初始化时的日志有差异，额外的日志行如下：
Using mount point: /gcs
opening GCS connection...
Set up root directory for all accessible buckets
Mounting file system "gcsfuse"
File system has been successfully mounted.
我不确定这些将如何改变 pandas read_csv 的方式。请帮忙。谢谢。
解决方法
暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！
如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。
小编邮箱:dio#foxmail.com (将#修改为@）