带有计数和 tfidf 矢量化器的管道产生 TypeError: expected string or bytes-like object
Posted
技术标签:
【中文标题】带有计数和 tfidf 矢量化器的管道产生 TypeError: expected string or bytes-like object【英文标题】：Pipeline with count and tfidf vectorizer produces TypeError: expected string or bytes-like object 【发布时间】：2021-03-24 04:27:23 【问题描述】：我有一个如下的语料库 'CCC 0 0 0 X 0 1 0 0 0 0'、'CCC 0 0 0 X 0 1 0 0 0 0'、'CCC 0 0 0 X 0 1 0 0 0 0'、'XX X'、'XX X '，'XX'， 我想使用 count 和 tfidf 矢量化器以及逻辑回归作为分类器。 下面的代码我改编自sklearn的示例。
from pprint import pprint
from time import time
import logging
import pickle
from sklearn.datasets import fetch_20newsgroups
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import GridSearchCV
from sklearn.pipeline import Pipeline
from sklearn.linear_model import LogisticRegression
print(__doc__)
# Display progress logs on stdout
logging.basicConfig(level=logging.INFO,
format='%(asctime)s %(levelname)s %(message)s')
# #############################################################################
# Define a pipeline combining a text feature extractor with a simple
# classifier
pipeline = Pipeline([
('vect', CountVectorizer(analyzer='char',lowercase=False)),
('tfidf', TfidfVectorizer(analyzer='char',lowercase=False)),
('clf', LogisticRegression()),
])
# uncommenting more parameters will give better exploring power but will
# increase processing time in a combinatorial way
parameters =
'vect__max_df': (0.5, 0.75, 1.0),
# 'vect__max_features': (None, 5000, 10000, 50000),
'vect__ngram_range': ((1, 1), (1, 2)), # unigrams or bigrams
# 'tfidf__use_idf': (True, False),
# 'tfidf__norm': ('l1', 'l2'),
'clf__max_iter': (1000,),
'clf__C': (0.00001, 0.000001),
'clf__penalty': ('l2', 'elasticnet'),
# 'clf__max_iter': (10, 50, 80),
if __name__ == "__main__":
# multiprocessing requires the fork to happen in a __main__ protected
# block
# find the best parameters for both the feature extraction and the
# classifier
grid_search = GridSearchCV(pipeline, parameters, n_jobs=-1, verbose=1)
corpus =['C C C 0 0 0 X 0 1 0 0 0 0', 'C C C 0 0 0 X 0 1 0 0 0 0', 'C C C 0 0 0 X 0 1 0 0 0 0', 'X X X', 'X X X',
'X X X', 'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 X',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X',
'C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C X X X X 0 0 0 X 0 X X']
y_train = [0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
print(len(corpus),len(y_train))
print("Performing grid search...")
print("pipeline:", [name for name, _ in pipeline.steps])
print("parameters:")
pprint(parameters)
t0 = time()
#print(type(data.data),type(data.target))
#print(data.data[:1])
#print(data.data[:2])
grid_search.fit(corpus,y_train)
print("done in %0.3fs" % (time() - t0))
print()
print("Best score: %0.3f" % grid_search.best_score_)
print("Best parameters set:")
best_parameters = grid_search.best_estimator_.get_params()
for param_name in sorted(parameters.keys()):
print("\t%s: %r" % (param_name, best_parameters[param_name]))
我的堆栈跟踪如下
Automatically created module for IPython interactive environment
50 50
Performing grid search...
pipeline: ['vect', 'tfidf', 'clf']
parameters:
'clf__C': (1e-05, 1e-06),
'clf__max_iter': (1000,),
'clf__penalty': ('l2', 'elasticnet'),
'vect__max_df': (0.5, 0.75, 1.0),
'vect__ngram_range': ((1, 1), (1, 2))
Fitting 5 folds for each of 24 candidates, totalling 120 fits
[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.
[Parallel(n_jobs=-1)]: Done 120 out of 120 | elapsed: 0.1s finished
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-114-0d47590b1279> in <module>
107 #print(data.data[:2])
108
--> 109 grid_search.fit(corpus,y_train)
110 print("done in %0.3fs" % (time() - t0))
111 print()
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\model_selection\_search.py in fit(self, X, y, groups, **fit_params)
737 refit_start_time = time.time()
738 if y is not None:
--> 739 self.best_estimator_.fit(X, y, **fit_params)
740 else:
741 self.best_estimator_.fit(X, **fit_params)
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\pipeline.py in fit(self, X, y, **fit_params)
348 This estimator
349 """
--> 350 Xt, fit_params = self._fit(X, y, **fit_params)
351 with _print_elapsed_time('Pipeline',
352 self._log_message(len(self.steps) - 1)):
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\pipeline.py in _fit(self, X, y, **fit_params)
313 message_clsname='Pipeline',
314 message=self._log_message(step_idx),
--> 315 **fit_params_steps[name])
316 # Replace the transformer of the step with the fitted
317 # transformer. This is necessary when loading the transformer
E:\anaconda\envs\appliedaicourse\lib\site-packages\joblib\memory.py in __call__(self, *args, **kwargs)
350
351 def __call__(self, *args, **kwargs):
--> 352 return self.func(*args, **kwargs)
353
354 def call_and_shelve(self, *args, **kwargs):
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\pipeline.py in _fit_transform_one(transformer, X, y, weight, message_clsname, message, **fit_params)
726 with _print_elapsed_time(message_clsname, message):
727 if hasattr(transformer, 'fit_transform'):
--> 728 res = transformer.fit_transform(X, y, **fit_params)
729 else:
730 res = transformer.fit(X, y, **fit_params).transform(X)
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\feature_extraction\text.py in fit_transform(self, raw_documents, y)
1857 """
1858 self._check_params()
-> 1859 X = super().fit_transform(raw_documents)
1860 self._tfidf.fit(X)
1861 # X is already a transformed view of raw_documents so
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\feature_extraction\text.py in fit_transform(self, raw_documents, y)
1218
1219 vocabulary, X = self._count_vocab(raw_documents,
-> 1220 self.fixed_vocabulary_)
1221
1222 if self.binary:
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\feature_extraction\text.py in _count_vocab(self, raw_documents, fixed_vocab)
1129 for doc in raw_documents:
1130 feature_counter =
-> 1131 for feature in analyze(doc):
1132 try:
1133 feature_idx = vocabulary[feature]
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\feature_extraction\text.py in _analyze(doc, analyzer, tokenizer, ngrams, preprocessor, decoder, stop_words)
108 doc = ngrams(doc, stop_words)
109 else:
--> 110 doc = ngrams(doc)
111 return doc
112
E:\anaconda\envs\appliedaicourse\lib\site-packages\sklearn\feature_extraction\text.py in _char_ngrams(self, text_document)
255 """Tokenize text_document into a sequence of character n-grams"""
256 # normalize white spaces
--> 257 text_document = self._white_spaces.sub(" ", text_document)
258
259 text_len = len(text_document)
TypeError: expected string or bytes-like object
我单独运行了 tfidf 矢量化器，得到以下结果
vectorizer = TfidfVectorizer(analyzer='char',lowercase=False,ngram_range=(6, 6))
X = vectorizer.fit_transform(corpus)
print(vectorizer.get_feature_names())
print(X.shape)
print(X)
结果
<class 'list'>
[' 0 0 0', ' 0 0 X', ' 0 1 0', ' 0 X 0', ' 0 X X', ' 1 0 0', ' C 0 0', ' C C 0', ' C C C', ' C C X', ' C X X', ' X 0 0', ' X 0 1', ' X 0 X', ' X X 0', ' X X X', '0 0 0 ', '0 0 X ', '0 1 0 ', '0 X 0 ', '1 0 0 ', 'C 0 0 ', 'C C 0 ', 'C C C ', 'C C X ', 'C X X ', 'X 0 0 ', 'X 0 1 ', 'X 0 X ', 'X X 0 ', 'X X X ']
(50, 31)
(0, 20) 0.31810783213188626
(0, 5) 0.31810783213188626
(0, 18) 0.31810783213188626
(0, 2) 0.31810783213188626
(0, 27) 0.31810783213188626
(0, 12) 0.31810783213188626
(0, 19) 0.16116825632411622
(0, 3) 0.16116825632411622
(0, 17) 0.16116825632411622
(0, 1) 0.11378963445554637
(0, 16) 0.22757926891109273
(0, 0) 0.3413689033666391
(0, 21) 0.17370780684495662
(0, 6) 0.17370780684495662
(0, 22) 0.17370780684495662
(0, 7) 0.17370780684495662
(0, 23) 0.11378963445554637
(1, 20) 0.31810783213188626
(1, 5) 0.31810783213188626
(1, 18) 0.31810783213188626
...
...
...
(49, 1) 0.01436413072356797
(49, 16) 0.01436413072356797
(49, 0) 0.01436413072356797
(49, 23) 0.6894782747312626
我的问题
为什么独立矢量化器可以工作，但是当放置在 Gridsearch 使用的管道中时，我得到类型错误
【问题讨论】：
【参考方案1】：默认情况下，CountVectorizer 和 TfidfVectorizer 都需要一个可以是字符串或字节类型的项目序列。在您的管道中，CountVectorizer 接收语料库并使用 scipy.sparse.csr_matrix 向 TfidfVectorizer 输出计数的稀疏表示。由于 TfidfVectorizer 的输入不是预期的类型，因此您会收到类型错误“TypeError：预期的字符串或类似字节的对象”。如果您使用其中一个但不是两个矢量化器，您的管道就可以工作。例如，
pipeline = Pipeline([
#('vect', CountVectorizer(analyzer='char',lowercase=False)),
('tfidf', TfidfVectorizer(analyzer='char',lowercase=False)),
('clf', LogisticRegression())
])
# uncommenting more parameters will give better exploring power but will
# increase processing time in a combinatorial way
parameters =
#'vect__max_df': (0.5, 0.75, 1.0),
# 'vect__max_features': (None, 5000, 10000, 50000),
#'vect__ngram_range': [(1, 1), (1, 2)], # unigrams or bigrams
'tfidf__use_idf': [True, False],
'tfidf__norm': ['l1', 'l2'],
'clf__max_iter': [1000],
'clf__C': [0.00001, 0.000001],
'clf__penalty': ['l2'],
# 'clf__max_iter': (10, 50, 80),
产生以下输出：
50 50
Performing grid search...
pipeline: ['tfidf', 'clf']
parameters:
'clf__C': [1e-05, 1e-06],
'clf__max_iter': [1000],
'clf__penalty': ['l2'],
'tfidf__norm': ['l1', 'l2'],
'tfidf__use_idf': [True, False]
Fitting 5 folds for each of 8 candidates, totalling 40 fits
[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.
done in 0.347s
Best score: 0.680
Best parameters set:
clf__C: 1e-05
clf__max_iter: 1000
clf__penalty: 'l2'
tfidf__norm: 'l1'
tfidf__use_idf: True
[Parallel(n_jobs=-1)]: Done 40 out of 40 | elapsed: 0.2s finished
【讨论】：
以上是关于带有计数和 tfidf 矢量化器的管道产生 TypeError: expected string or bytes-like object的主要内容，如果未能解决你的问题，请参考以下文章