python - 无法在 AWS EMR 上下载 nltk 语料库，对已关闭文件进行 I/O 操作

Question

使用 JupyterLab 打开我的 EMR 集群后。我无法使用 nltk.download() 下载额外的语料库。

代码

nltk.download('wordnet')

错误

I/O operation on closed file
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/site-packages/nltk/downloader.py", line 817, in download
    show('Downloading collection %r' % msg.collection.id)
  File "/usr/local/lib/python3.6/site-packages/nltk/downloader.py", line 783, in show
    subsequent_indent=prefix + prefix2 + ' ' * 4,
  File "/tmp/4461650941863117011", line 534, in write
    super(UnicodeDecodingStringIO, self).write(s)
ValueError: I/O operation on closed file

这是在使用 sc.list_packages() 确认安装了 nltk 之后。

Package                    Version
-------------------------- -------
...
nltk                       3.4.5  
...

并使用 import nltk 导入 nltk。

感觉这个问题是因为我对 EMR 的设置方式缺乏了解。

有什么我应该尝试调试的吗？

更新:

我已经尝试将它安装在引导脚本中，该脚本可以正确安装。

pip install nltk
python -m nltk.downloader wordnet

但是我在尝试使用它时仍然遇到这个错误。

An error occurred while calling o166.showString.
: org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 2.0 failed 4 times, most recent failure: Lost task 0.3 in stage 2.0 (TID 7, ip-172-31-1-163.ca-central-1.compute.internal, executor 3): org.apache.spark.api.python.PythonException: Traceback (most recent call last):
  File "/usr/local/lib/python3.6/site-packages/nltk/corpus/util.py", line 86, in __load
    root = nltk.data.find('{}/{}'.format(self.subdir, zip_name))
  File "/usr/local/lib/python3.6/site-packages/nltk/data.py", line 701, in find
    raise LookupError(resource_not_found)
LookupError: 
**********************************************************************
  Resource wordnet not found.
  Please use the NLTK Downloader to obtain the resource:

  >>> import nltk
  >>> nltk.download('wordnet')

  For more information see: https://www.nltk.org/data.html

  Attempted to load corpora/wordnet.zip/wordnet/

  Searched in:
    - '/home/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/virtualenv_application_1576604798325_0001_0/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/virtualenv_application_1576604798325_0001_0/share/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/virtualenv_application_1576604798325_0001_0/lib/nltk_data'
    - '/usr/share/nltk_data'
    - '/usr/local/share/nltk_data'
    - '/usr/lib/nltk_data'
    - '/usr/local/lib/nltk_data'
**********************************************************************


During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/worker.py", line 377, in main
    process()
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/worker.py", line 372, in process
    serializer.dump_stream(func(split_index, iterator), outfile)
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/serializers.py", line 345, in dump_stream
    self.serializer.dump_stream(self._batched(iterator), stream)
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/serializers.py", line 141, in dump_stream
    for obj in iterator:
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/serializers.py", line 334, in _batched
    for item in iterator:
  File "<string>", line 1, in <lambda>
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/worker.py", line 85, in <lambda>
    return lambda *a: f(*a)
  File "/mnt1/yarn/usercache/livy/appcache/application_1576604798325_0001/container_1576604798325_0001_01_000005/pyspark.zip/pyspark/util.py", line 113, in wrapper
    return f(*args, **kwargs)
  File "<stdin>", line 19, in <lambda>
  File "<stdin>", line 19, in <listcomp>
  File "/usr/local/lib/python3.6/site-packages/nltk/stem/wordnet.py", line 41, in lemmatize
    lemmas = wordnet._morphy(word, pos)
  File "/usr/local/lib/python3.6/site-packages/nltk/corpus/util.py", line 123, in __getattr__
    self.__load()
  File "/usr/local/lib/python3.6/site-packages/nltk/corpus/util.py", line 88, in __load
    raise e
  File "/usr/local/lib/python3.6/site-packages/nltk/corpus/util.py", line 83, in __load
    root = nltk.data.find('{}/{}'.format(self.subdir, self.__name))
  File "/usr/local/lib/python3.6/site-packages/nltk/data.py", line 701, in find
    raise LookupError(resource_not_found)
LookupError: 

更新

我找到了 wordnet 被 shell 脚本下载到的目录，并通过 ssh 进入服务器确认它确实存在。

[nltk_data] Downloading package wordnet to /root/nltk_data...

所以在 jupyter 中我正在检查 nltk.data.path

['/var/lib/livy/nltk_data', '/tmp/1576616653412-0/nltk_data', '/tmp/1576616653412-0/share/nltk_data', '/tmp/1576616653412-0/lib/nltk_data', '/usr/share/nltk_data', '/usr/local/share/nltk_data', '/usr/lib/nltk_data', '/usr/local/lib/nltk_data']

并附加我的新路径。

nltk.data.path.append('/root/nltk_data')
nltk.data.path

我们可以看到它添加了。

['/var/lib/livy/nltk_data', '/tmp/1576616653412-0/nltk_data', '/tmp/1576616653412-0/share/nltk_data', '/tmp/1576616653412-0/lib/nltk_data', '/usr/share/nltk_data', '/usr/local/share/nltk_data', '/usr/lib/nltk_data', '/usr/local/lib/nltk_data', '/root/nltk_data']

但当我尝试调用使用该语料库的函数时，它仍然没有被搜索到。

  Resource wordnet not found.
  Please use the NLTK Downloader to obtain the resource:

  >>> import nltk
  >>> nltk.download('wordnet')

  For more information see: https://www.nltk.org/data.html

  Attempted to load corpora/wordnet.zip/wordnet/

  Searched in:
    - '/home/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576615748346_0001/container_1576615748346_0001_01_000006/virtualenv_application_1576615748346_0001_0/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576615748346_0001/container_1576615748346_0001_01_000006/virtualenv_application_1576615748346_0001_0/share/nltk_data'
    - '/mnt1/yarn/usercache/livy/appcache/application_1576615748346_0001/container_1576615748346_0001_01_000006/virtualenv_application_1576615748346_0001_0/lib/nltk_data'
    - '/usr/share/nltk_data'
    - '/usr/local/share/nltk_data'
    - '/usr/lib/nltk_data'
    - '/usr/local/lib/nltk_data'

这里没有引用/root/nltk_data

Answer 1

鉴于无法更改用于加载 wordnet 的路径(更改 nltk.data.path 并未更改 nltk 查找文件的位置)。

我必须更改它从启动脚本下载到的目录，以匹配 nltk 默认的位置。

启动脚本

sudo pip install nltk
sudo python -m nltk.downloader -d /home/nltk_data wordnet