Python 3.x 由于类型错误，模型无法预测，在UTF-8编码的乌尔都语数据集上应用模型

我试图在乌尔都语数据集上运行一个算法，我有一个英语的逻辑回归模型，但当我试图在乌尔都语UTF-8编码数据集上使用它时，遇到了错误

我已经在这个问题上应用了下面给出的代码

import pandas as pd
import re
from nltk.corpus import stopwords
import nltk
import logging
from gensim.models import word2vec
import numpy as np 
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.preprocessing import StandardScaler
from sklearn.feature_extraction.text import TfidfVectorizer
from scipy.sparse import hstack
from scipy.sparse import coo_matrix
from tqdm import tqdm
from scipy import sparse
import numpy 
#import score
import re, nltk, scipy

#import gensim
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier,BaggingClassifier,VotingClassifier,AdaBoostClassifier,GradientBoostingClassifier
from sklearn.model_selection import KFold

dataset1 = pd.read_csv("fakenews.csv", encoding = 'UTF-8')
ds = dataset1.copy()
print(ds.shape)
df = ds.copy()

df = ds.copy()
df.drop('FileName',axis=1,inplace = True)
df.drop('label',axis=1,inplace = True)

da = ds.copy()
da.drop('FileName',axis=1,inplace = True)
da.drop('title',axis=1,inplace = True)
da.drop('text',axis=1,inplace = True)

def extract_word_overlap(title, text):
    word_overlap = []
    for i, (title, body) in tqdm(enumerate(zip(title, text))):
        #preprocess_headline = preprocess(headline)
        #preprocess_body = preprocess(body)
        features = len(set(title).intersection(text)) / float(len(set(title).union(text)))
        word_overlap.append(features)
        word_overlap_sparse = scipy.sparse.coo_matrix(numpy.array(word_overlap)) 

    return word_overlap_sparse

def combine_features(tfidf_vectors, word_overlap):
    combined_features =  sparse.bmat([[tfidf_vectors, word_overlap.T]])
    return combined_features

from sklearn.model_selection import train_test_split   
x_train, x_test,y_train,y_test = train_test_split(df,da, test_size = 0.3,random_state=42)

training_bodies = x_train['text']
training_headlines = x_train['title']

test_bodies = x_test['text']
test_headlines = x_test['title']

print("\t-Extracting tfidf vectors..")

body_vectorizer = TfidfVectorizer(ngram_range=(1, 2))
bodies_tfidf = body_vectorizer.fit_transform(training_bodies)

headline_vectorizer = TfidfVectorizer(ngram_range=(1, 2))
headlines_tfidf = headline_vectorizer.fit_transform(training_headlines)


bodies_tfidf_test = body_vectorizer.transform(test_bodies)
headlines_tfidf_test = headline_vectorizer.transform(test_headlines)

training_tfidf = scipy.sparse.hstack([bodies_tfidf, headlines_tfidf])
test_tfidf = scipy.sparse.hstack([bodies_tfidf_test, headlines_tfidf_test])



training_overlap = extract_word_overlap(training_headlines, training_bodies)
test_overlap = extract_word_overlap(test_headlines, test_bodies)
training_features = combine_features(training_tfidf,  training_overlap)    
test_features = combine_features(test_tfidf, test_overlap)



print("[3] Fitting model..")
print("\t-Logistic Regression")

lr = LogisticRegression(C = 1.0, class_weight='balanced', solver="lbfgs", max_iter=150) 


y_pred = lr.fit(training_features, y_train).predict(test_features)


from sklearn import metrics
from sklearn.metrics import classification_report, accuracy_score, f1_score
print('classification report:')
print(classification_report(y_test, y_pred))
score = metrics.accuracy_score(y_test, y_pred)

print("Accuracy of Regression:   %0.3f" % score)

print('Macro f1:', f1_score(y_test, y_pred, average='macro'))
print("[4] Evaluating model..")
score.report_score(y_test, y_pred)
print(confusion_matrix(y_test, y_pred))
"""

---

我希望模型能预测，但它给了我一个类型错误。TypeError:“使用

dtype

将每列作为字符串读取，同时检查是否缺少/balnk值。@Shijith没有找到缺少/空白值。您能检查

文本

列吗，它是否有

“

在数据中？可能是个问题there@Shijith各种标点符号已被删除，再次检查过的文本列没有“在数据中。您能包含完整的回溯吗，这样就可以清楚地知道是哪一行导致了错误？”？显然，字符串和数字之间存在比较，但代码太多，无法按原样发现错误。请使用

dtype

将每列作为字符串读取，同时检查是否缺少/balnk值。@Shijith没有发现缺少/空白值。您能检查

文本

列吗，它有

“

在数据中？可能是个问题there@Shijith所有的标点符号都被删除了，再次检查过了。数据中的文本列没有“错误”。你能不能包含完整的回溯，这样就可以清楚地知道是哪一行导致了错误？显然，字符串和数字之间存在比较，但代码太多了，无法按原样发现错误。

761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
  y = column_or_1d(y, warn=True)
Traceback (most recent call last):


  File "<ipython-input-1-800030f783cf>", line 1, in <module>
    runfile('C:/Users/door/Desktop/af/pycode.py', wdir='C:/Users/door/Desktop/af')

  File "C:\Users\door\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 704, in runfile
    execfile(filename, namespace)

  File "C:\Users\door\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 108, in execfile
    exec(compile(f.read(), filename, 'exec'), namespace)

  File "C:/Users/door/Desktop/af/pycode.py", line 101, in <module>
    y_pred = lr.fit(training_features, y_train).predict(test_features)

  File "C:\Users\door\Anaconda3\lib\site-packages\sklearn\linear_model\logistic.py", line 1286, in fit
    check_classification_targets(y)


  File "C:\Users\door\Anaconda3\lib\site-packages\sklearn\utils\multiclass.py", line 168, in check_classification_targets
    y_type = type_of_target(y)

  File "C:\Users\door\Anaconda3\lib\site-packages\sklearn\utils\multiclass.py", line 287, in type_of_target
    if (len(np.unique(y)) > 2) or (y.ndim >= 2 and len(y[0]) > 1):

  File "C:\Users\door\Anaconda3\lib\site-packages\numpy\lib\arraysetops.py", line 233, in unique
    ret = _unique1d(ar, return_index, return_inverse, return_counts)

  File "C:\Users\door\Anaconda3\lib\site-packages\numpy\lib\arraysetops.py", line 281, in _unique1d
    ar.sort()

TypeError: '<' not supported between instances of 'float' and 'str'

FileName    title   text    label
0001a   میں استعفیٰ نہیں دے رہا   فیاض الحسن چوہان  "صوبائی وزیر اطلاعات فیاض الحسن چوہان نے کہا ہے کہ میں استعفیٰ نہیں دے رہا اور نہ مجھ سے استعفیٰ مانگا گیا ہے  مجھے محتاط بیان دینے کا کہا گیا ہے 
اور میں نے اپنی بیان میں ہندو مذہب یا ہندو برادری کو نہیں بلکہ بھارتی فوج اور بھارتی میڈیا کو مخاطب کیا تھا"    Agree