如何编写预测方法来使用 sklearn 手动测试我的 RNN 随机分类器模型的性能？

问题描述

从 3 位诗人的诗歌数据集中，数据分别被标记为 1,2,3，分别是诗人 1、诗人 2 和诗人 3。 现在我们要从他们的诗歌数据集中对这 3 位诗人进行分类。波纹管，我已经使用 sklearn 分享了我的 RNN 随机分类器模型。现在我的问题是如何编写一个预测方法，我可以手动给出一首诗，它会对诗人的名字进行分类？请帮忙。

代码如下：

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import Pipeline
from sklearn.metrics import accuracy_score
def accuracy_summary(pipeline,X_train,y_train,X_test,y_test):
    poet_fit = pipeline.fit(X_train,y_train)
    y_pred = poet_fit.predict(X_test)
    accuracy = accuracy_score(y_test,y_pred)
    print("accuracy score: {0:.2f}%".format(accuracy*100))
    return accuracy



cv = CountVectorizer()
rf = RandomForestClassifier(class_weight="balanced")
n_features = np.arange(10000,25001,5000)
def nfeature_accuracy_checker(vectorizer=cv,n_features=n_features,stop_words=None,ngram_range=(1,3),classifier=rf):
    result = []
    print(classifier)
    print("\n")
    for n in n_features:
        vectorizer.set_params(stop_words=stop_words,max_features=n,ngram_range=ngram_range)
        checker_pipeline = Pipeline([
            ('vectorizer',vectorizer),('classifier',classifier)
        ])
        print("Test result for {} features".format(n))
        nfeature_accuracy = accuracy_summary(checker_pipeline,y_test)
        result.append((n,nfeature_accuracy))
    return result


from sklearn.feature_extraction.text import TfidfVectorizer
tfidf = TfidfVectorizer(encoding='utf-8')
print("Result for trigram with stop words (Tfidf)\n")
feature_result_tgt = nfeature_accuracy_checker(vectorizer=tfidf,2))



from sklearn.metrics import classification_report
cv = CountVectorizer(max_features=20000,2))
pipeline = Pipeline([
        ('vectorizer',cv),rf)
    ])
poet_fit = pipeline.fit(X_train,y_train)
y_pred = poet_fit.predict(X_test)


print(classification_report(y_test,y_pred,target_names=['poet 1','poet 2','poet 3']))
print('accuracy %s' % accuracy_score(y_pred,y_test))
print(y_pred)

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）