python - 错误：super(type, obj) obj 必须是 type 的实例或子类型

Question

请协助使用机器学习模型对我的分类代码进行故障排除。它抛出以下错误。代码的主要本质是使用 imdb 数据训练模型，以正确分类正面评论和负面评论。

File "/Users/darlingtonemeagi/Downloads/untitled14.py", line 49, in <module>
    model = models.Sequential()

  File "/anaconda3/lib/python3.5/site-packages/keras/engine/sequential.py", line 87, in __init__
    super(Sequential, self).__init__(name=name)

  File "/anaconda3/lib/python3.5/site-packages/keras/legacy/interfaces.py", line 91, in wrapper
    return func(*args, **kwargs)

  File "/anaconda3/lib/python3.5/site-packages/keras/engine/network.py", line 96, in __init__
    self._init_subclassed_network(**kwargs)

  File "/anaconda3/lib/python3.5/site-packages/keras/engine/network.py", line 300, in _init_subclassed_network
    self._base_init(name=name)

  File "/anaconda3/lib/python3.5/site-packages/keras/engine/network.py", line 110, in _base_init
    self.name = name

  File "/anaconda3/lib/python3.5/site-packages/keras/engine/network.py", line 322, in __setattr__
    super(Network, self).__setattr__(name, value)

TypeError: super(type, obj): obj must be an instance or subtype of type"

我查看了关键代码，但无法发现问题出在哪里。

from keras import models
from keras import layers
from keras.datasets import imdb
import numpy as np

(train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=10000)

#Vectorised the data using Hot-encode technology

def vectorise_sequence(sequences, dimension=10000):
    results = np.zeros((len(sequences), dimension)) #Create zeros matrix of shape len(sequences) and dimension
    for i, sequence in enumerate(sequences):
        results[i, sequence] = 1 # set specific indices of results[i] to 1
    return results

#Vectorise training and test data

x_train = vectorise_sequence(train_data)
x_test = vectorise_sequence(test_data)

#Also vectorising the labels which is also straight forward

y_train = np.asarray(train_labels).astype('float32')
y_test = np.asarray(test_labels).astype('float32')

#The inputs data are vectors while the output data is scalar (0 and 1)

#Set aside a validation set: Data the model has never seen before in other to monitor
#the accuracy of the model

x_val= x_train[:10000]
partial_x_train = x_train[10000:]

y_val = y_train[:10000]
partial_y_train = y_train[10000:]

#Building the Network Model

model = models.Sequential()
model.add(layers.Dense(16, activation='relu', input_size=(10000,)))
model.add(layers.Dense(16, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

#Compiling the network

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

history = model.fit(partial_x_train, partial_y_train, epochs=20,\
                    batch_size=512, validation_data = (x_val,y_val))

history_dict = history.history
print(history_dict.keys())

import matplotlib.pyplot as plt

loss_values = history_dict['loss']
val_loss_values = history_dict['val_loss']

epochs = range(1, 20)
plt.plot(epochs, loss_values, 'bo', label='Training Loss')
plt.plot (epochs, val_loss_values, 'b', label='Validation Loss')
plt.title('Training and Validation Loss')
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.Legend()
plt.show()

预期输出是训练损失和验证损失的图形表示。

标签： pythonkeras