如何保存模型 python3

Question

**我正在尝试保存模型以便在 Web 应用程序中使用它，但出现此错误 **

X = []

sentences = list(review_df['text'])

for sen in sentences:

X.append(clean_text(sen))

y = review_df['Label']

y = np.array(list(map(lambda x: 1 if x=="fake" else 0, y)))

#使用递归神经网络 (LSTM) 进行文本分类

from keras.layers.recurrent import LSTM

model = Sequential()

embedding_layer = Embedding(vocab_size, 100, weights=[embedding_matrix], input_length=maxlen , 

trainable=False)

model.add(embedding_layer)

model.add(LSTM(128))

model.add(Dense(1, activation='sigmoid'))

model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['acc'])

print(model.summary())

#训练模型

history = model.fit(X_train, y_train, batch_size=128, epochs=6, verbose=1, validation_split=0.2)

score = model.evaluate(X_test, y_test, verbose=1) 

#打印模型结果

print("Test Score:", score[0])

print("Test Accuracy:", score[1])

#对单个实例进行预测

instance = X[57]

print(instance)

instance = tokenizer.texts_to_sequences(instance)

flat_list = []

for sublist in instance:

for item in sublist:

flat_list.append(item)

flat_list = [flat_list]

instance = pad_sequences(flat_list, padding='post', maxlen=maxlen)

model.predict(instance)

#保存模型

import pickle

with open('model.pkl', 'wb') as f:

pickle.dump(model, f)

当我尝试保存模型时出现此错误： TypeError: can't pickle _thread.RLock objects 有没有解决此错误的想法

Answer 1

尝试这样做：model_json = model.to_json()with open("my_model.json", "w") as json_file:    json_file.write(model_json)

Answer 2

您可以直接保存keras模型： model.save('path/xyz.h5') 建议将模型保存为H5格式并使用model.save() 您也可以使用以下命令加载该格式model.load()