如何修复关于 TensorFlow 预测的 TypeError? [英] How to fix TypeError with respect to TensorFlow prediction?

问题描述

我一直在尝试使用 TensorFlow 执行神经网络预测，评估工作正常，但是当我将相同的数据放入预测时，它会出现类型错误.

I've been trying to perform neural network predictions with TensorFlow and the evaluation works fine but when I put the same data into prediction it gives a type error.

training_data 和 test_data 中的数据都是 2d numpy 整数数组，training_labels 和 test_labels 中的数据是 1d numpy 整数数组.

The data in training_data and test_data are both 2d numpy arrays of integers and training_labels and test_labels are 1d numpy arrays of integers.

model = keras.Sequential([
  keras.layers.Dense(24, activation=tf.nn.selu),
  keras.layers.Dense(10, activation=tf.nn.tanh),
  keras.layers.Dense(5, activation=tf.nn.selu),
  keras.layers.Dense(5, activation=tf.nn.softmax)
])


model.compile(optimizer='SGD',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(training_data, training_labels, epochs=10)

test_loss,test_acc = model.evaluate(test_data, test_labels)

prediction = model.predict(test_data)

当取出预测线时，代码按预期工作，但现在给出以下错误消息:

When taking out the prediction line the code works as expected but it now gives the following error message:

Traceback (most recent call last):
  File "learner.py", line 132, in <module>
    print("Prediction: " + model.predict(test_data))
TypeError: ufunc 'add' did not contain a loop with signature matching types dtype('S32') dtype('S32') dtype('S32')

我已经确保所有数据都是整数，所以我不确定为什么会出现类型冲突.

I've made sure that all the data is integers so I'm not sure why there's type conflict.

推荐答案

test_data 使用 predict 时应与 type(training_data[0]) 的数据类型相同，并且返回类型为(training_labels[0])的数据类型

test_data should be the same datatype as type(training_data[0]) when using predict, and it will return a datatype of type(training_labels[0])

此外，

print("Prediction: " + model.predict(test_data))

必须

print("Prediction: " + str(model.predict(test_data)))

这篇关于如何修复关于 TensorFlow 预测的 TypeError?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！