python - python中未使用函数的重新定义

Question

我在同一文件夹中有 2 个 python 文件 utils.py 和 preprocessing.py。

我在 utils.py 文件中定义了两个函数:preprocess_text, text2seq_generator

我需要在 preprocessing.py 中使用这些函数。但我无法导入并警告说重新定义第 19 行未使用的 preprocess_text

代码:

utils.py

from __future__ import print_function
from collections import defaultdict
from collections import Counter
from string import punctuation
#import pdb


def filter_text(text):
    return ''.join([c for c in text if c not in punctuation]).lower()

def preprocess_text(file_path_src, file_path_tar, max_feats):
    f_src = open(file_path_src)
    f_tar = open(file_path_tar)
    vocab = defaultdict(int)
    freq_src = defaultdict(int)
    freq_tar = defaultdict(int)
    sents_src = [line.rstrip() for line in f_src.readlines()]
    sents_tar = [line.rstrip() for line in f_tar.readlines()]

    for sent in sents_src:
        sent = filter_text(sent)
        for word in sent.split():
            freq_src[word] += 1

   for sent in sents_tar:
       for word in sent.split():
           freq_tar[word] += 1

   freq_sorted_src = Counter(freq_src)
   freq_sorted_src = freq_sorted_src.most_common(max_feats)
   freq_sorted_tar = Counter(freq_tar)
   freq_sorted_tar = freq_sorted_tar.most_common(max_feats)

   freq_words_src = map(lambda x:x[0], freq_sorted_src)
   freq_words_tar = map(lambda x:x[0], freq_sorted_tar)

   vocab_src = dict({v:k for k, v in enumerate(freq_words_src)})
   vocab_tar = dict({v:k for k, v in enumerate(freq_words_tar)})

   vocab_src["UNK"] = max_feats
   vocab_src["<s>"] = max_feats + 1
   vocab_src["</s>"] = max_feats + 2

   vocab_tar["UNK"] = max_feats
   vocab_tar["<s>"] = max_feats + 1
   vocab_tar["</s>"] = max_feats + 2
   #pdb.set_trace()
   return vocab_src, vocab_tar, sents_src, sents_tar


def text2seq_generator(vocab_src, vocab_tar, sents_src, sents_tar):
    unk_key = vocab_src["UNK"]
    for sent_src, sent_tar in zip(sents_src, sents_tar):
        seq_src = map(lambda x:vocab_src.get(x, unk_key), filter_text(sent_src).split())
        seq_tar = map(lambda x:vocab_tar.get(x, unk_key), sent_tar.split())
        #pdb.set_trace()
        yield seq_src, seq_tar


 #vs, vt, ss, st = preprocess_text('../data/training.hi-en.en', '../data/training.hi-en.hi', 5000)
 #for x,y in text2seq_generator(vs, vt, ss, st):
 #print(y)

预处理.py文件:

  from keras.preprocessing.text import Tokenizer
  from keras.preprocessing.sequence import pad_sequences
  import codecs
  import pdb
  import numpy as np
  from utils import preprocess_text,text2seq_generator

错误:

  from keras.preprocessing.text import Tokenizer
  from keras.preprocessing.sequence import pad_sequences
  import codecs
  import pdb
  import numpy as np
  from utils import preprocess_text

  Traceback (most recent call last):

   File "<ipython-input-8-335a30ab7871>", line 6, in <module>
from utils import preprocess_text

  ImportError: cannot import name 'preprocess_text'

警告:

   redefinition of unused preprocess_text from line 19

Answer 1

您的 PYTHONPATH 中似乎有另一个名为 utils 的包与当前文件冲突。

您可以使用 importlib 模块从其路径显式加载模块。

将这些行添加到 preprocessing.py(python 3.3 和 3.4):

from importlib.machinery import SourceFileLoader

utils = SourceFileLoader("utils", "/path/to/utils.py").load_module()
utils.preprocess_text(...)

对于 python3.5+，您应该使用:

import importlib.util
spec = importlib.util.spec_from_file_location("utils", "/path/to/utils.py")
utils = importlib.util.module_from_spec(spec)
spec.loader.exec_module(utils)
utils.preprocess_text(...)