Question

我正在尝试对产品评论进行一些情绪分析，并且因为让我的模型读取字数词典而被绊倒

import pandas as pd  
import numpy as np   
from sklearn import linear_model, model_selection, metrics

products = pd.read_csv('data.csv')

def count_words(s):
   d = {}
   wl = str(s).split()
   for w in wl:
       d[w] = wl.count(w)
   return d

products['word_count'] = products['review'].apply(count_words)

products = products[products['rating'] != 3]
products['sentiment'] = (products['rating'] >= 4) * 1 

train_data, test_data = model_selection.train_test_split(products, test_size = 0.2, random_state=0)

sentiment_model = linear_model.LogisticRegression()
sentiment_model.fit(X = train_data['word_count'], y =train_data['sentiment'])

当我运行最后一行时，我收到以下错误：

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-51-0c3f47af3a6e> in <module>()
----> 1 sentiment_model.fit(X = train_data['word_count'], y = 
train_data['sentiment'])

C:\ProgramData\anaconda_3\lib\site-packages\sklearn\linear_model\logistic.py 
in fit(self, X, y, sample_weight)
   1171 
   1172         X, y = check_X_y(X, y, accept_sparse='csr', dtype=np.float64,
-> 1173                          order="C")
   1174         check_classification_targets(y)
   1175         self.classes_ = np.unique(y)

C:\ProgramData\anaconda_3\lib\site-packages\sklearn\utils\validation.py in 
check_X_y(X, y, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, multi_output, ensure_min_samples, ensure_min_features, y_numeric, warn_on_dtype, estimator)
    519     X = check_array(X, accept_sparse, dtype, order, copy, force_all_finite,
    520                     ensure_2d, allow_nd, ensure_min_samples,
--> 521                     ensure_min_features, warn_on_dtype, estimator)
    522     if multi_output:
    523         y = check_array(y, 'csr', force_all_finite=True, ensure_2d=False,

C:\ProgramData\anaconda_3\lib\site-packages\sklearn\utils\validation.py in check_array(array, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, warn_on_dtype, estimator)
    380                                       force_all_finite)
    381     else:
--> 382         array = np.array(array, dtype=dtype, order=order, copy=copy)
    383 
    384         if ensure_2d:

TypeError: float() argument must be a string or a number, not 'dict'

似乎该模型将字典作为x变量而不是字典中的条目。我想我需要将字典解压缩到数组（？）但是没有运气这样做。

更新：以下是运行word_count和定义情绪后的产品 products.head()

Answer 1

尝试

X = train_data['word_count'].values()

如果您正在寻找的话，这应该返回train_data['word_count']中每个项目的字数（数字）列表。

Answer 2

如果您只想更正错误，请先使用train_data['word_count']上的DictVectorizer将其转换为可接受的格式，即形状[n_samples, n_features]的数组。

在sentiment_model.fit()：

之前将以下内容添加到您的代码中

from sklearn.feature_extraction import DictVectorizer
dictVectorizer = DictVectorizer()

train_data_dict = dictVectorizer.fit_transform(train_data['word_count'])

然后像这样调用sentiment_model.fit（）：

sentiment_model.fit(X = train_data_dict, y =train_data['sentiment'])

注意： - 而不是实现自己的计数单词方法，我建议您使用CountVectorizer。

from sklearn.feature_extraction.text import CountVectorizer countVec = CountVectorizer() train_data_vectorizer = countVec.fit_transform(train_data['review']) sentiment_model.fit(X = train_data_vectorizer, y =train_data['sentiment'])

在Python中解压缩字典以进行Logistic回归

2 个答案: