Python 3.x 使用LSTM训练的模型只能预测所有样本的相同值
我有一个包含4000行和两列的数据集。第一列包含一些句子,第二列包含一些数字。 大约有4000个句子,它们被100个不同的数字分类。例如:Python 3.x 使用LSTM训练的模型只能预测所有样本的相同值,python-3.x,machine-learning,keras,deep-learning,lstm,Python 3.x,Machine Learning,Keras,Deep Learning,Lstm,我有一个包含4000行和两列的数据集。第一列包含一些句子,第二列包含一些数字。 大约有4000个句子,它们被100个不同的数字分类。例如: Sentences Codes Google headquarters is in California 87390 Steve Jobs was a great man 70214 Steve Jobs
Sentences Codes
Google headquarters is in California 87390
Steve Jobs was a great man 70214
Steve Jobs has done great technology innovations 70214
Google pixel is a very nice phone 87390
Microsoft is another great giant in technology 67012
Bill Gates founded Microsoft 67012
类似地,总共有4000行包含这些句子,这些行用100个这样的代码进行分类
我已经尝试了下面的代码,但当我预测时,它预测的是一个相同的值。换句话说,y_pred给出了一个相同值的数组
我能知道代码哪里出错了吗
import pandas as pd
import numpy as np
xl = pd.ExcelFile("dataSet.xlsx")
df = xl.parse('Sheet1')
#df = df.sample(frac=1).reset_index(drop=True)# shuffling the dataframe
df = df.sample(frac=1).reset_index(drop=True)# shuffling the dataframe
X = df.iloc[:, 0].values
Y = df.iloc[:, 1].values
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
import pickle
count_vect = CountVectorizer()
X = count_vect.fit_transform(X)
tfidf_transformer = TfidfTransformer()
X = tfidf_transformer.fit_transform(X)
X = X.toarray()
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_Y = LabelEncoder()
Y = labelencoder_Y.fit_transform(Y)
y = Y.reshape(-1, 1) # Because Y has only one column
onehotencoder = OneHotEncoder(categories='auto')
Y = onehotencoder.fit_transform(y).toarray()
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.2, random_state=0)
inputDataLength = len(X_test[0])
outputDataLength = len(Y[0])
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from keras.layers.embeddings import Embedding
from keras.preprocessing import sequence
from keras.layers import Dropout
# fitting the model
embedding_vector_length = 100
model = Sequential()
model.add(Embedding(outputDataLength,embedding_vector_length, input_length=inputDataLength))
model.add(Dropout(0.2))
model.add(LSTM(outputDataLength))
model.add(Dense(outputDataLength, activation='softmax'))
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())
model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=10, batch_size=20)
y_pred = model.predict(X_test)
invorg = model.inverse_transform(y_test)
y_test = labelencoder_Y.inverse_transform(invorg)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
您使用的是
binary\u交叉熵
即使您有100个
类。这不是正确的做法。对于此任务,您必须使用categorical\u crossentropy
像这样编译你的模型
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
model.compile(loss='classifical\u crossentropy',optimizer='adam',metrics=['accurity'])
另外,您正在使用模型进行预测,并转换为类标签,如下所示
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
由于您的模型是使用softmax激活的,以便获得类别标签,因此您必须找到预测的argmax
例如,如果预测值为[0.2,0.3,0.0005,0.99]
则必须取argmax,这将为您提供输出3
。高概率的类
所以你必须像这样修改预测代码
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
现在,您将在
invorg
中获得实际类标签,并在y\u pred
中获得预测类标签,您正在使用二进制交叉熵
即使您拥有100个
类。这不是正确的做法。对于此任务,您必须使用categorical\u crossentropy
像这样编译你的模型
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
model.compile(loss='classifical\u crossentropy',optimizer='adam',metrics=['accurity'])
另外,您正在使用模型进行预测,并转换为类标签,如下所示
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
由于您的模型是使用softmax激活的,以便获得类别标签,因此您必须找到预测的argmax
例如,如果预测值为[0.2,0.3,0.0005,0.99]
则必须取argmax,这将为您提供输出3
。高概率的类
所以你必须像这样修改预测代码
y_pred = model.predict(X_test)
inv = onehotencoder.inverse_transform(y_pred)
y_pred = labelencoder_Y.inverse_transform(inv)
y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)
y_pred = labelencoder_Y.inverse_transform(y_pred)
invorg = np.argmax(y_test, axis=1)
invorg = labelencoder_Y.inverse_transform(invorg)
现在,您将在
invorg
中获得实际类标签,在y\u pred
中获得预测类标签,因此您有100个类。然后你必须使用categorical\u crossentropy
来计算损失。另外,在预测之后,您必须使用预测的argmax
,这样您就有100个类。然后你必须使用categorical\u crossentropy
来计算损失。另外,在预测之后,您必须使用预测的argmax
,我想您可以使用binary\u crossentropy
和分类精度
是的,您可以。所讨论的代码的主要问题是预测。我正在使用的一个热编码呢?y=y.Reformate(-1,1)onehotencoder=onehotencoder(categories='auto')y=onehotencoder.fit_transform(y).toarray()我得到的数组的常量输出(['67012','67012','67012','67012','67012','67012'],dtype='你所说的输出是什么意思?model.predict()
给你这个结果。我想你可以使用二进制交叉熵和分类精度是的,你可以。问题代码的主要问题是预测。我使用的一个热编码呢?y=y。重塑(-1,1)onehotencoder=onehotencoder(categories='auto')y=onehotencoder。拟合变换(y).toarray()我得到的数组(['67012','67012','67012','67012',…,'67012','67012','67012','67012'])的常量输出,dtype='你所说的输出是什么意思。model.predict()
会给你这个结果吗。?