用于输入kerastensor（type_spec = tensorsPec（shape =（none，50），

Question

我已经接受了某种培训，最后的jupyter笔记本是：

我理解整个代码，以及如何培训模型。

但是，最后我预测了这样的测试数据集中推文的情绪：

i = random.randint(0, len(test_labels)-1)
print('Sentence:', test_tweets[i])
print('Emotion:', index_to_class[test_labels[i]])
p = model.predict(np.expand_dims(test_seq[i], axis=0))[0]
pred_class = index_to_class[np.argmax(p).astype('uint8')]
print('Predicted Emotion:', pred_class)

这很好。

但是，我想用随机句子来测试模型预测，例如：

sentence = 'I love you more than ever'
print('Sentence:',  sentence)
#print('Emotion:', index_to_class[test_labels[i]])
p = model.predict(np.expand_dims(sentence, axis=0))[0]
pred_class = index_to_class[np.argmax(p).astype('uint8')]
print('Predicted Emotion:', pred_class)

但是我有一个错误：

Sentence: I love you more than ever
WARNING:tensorflow:Model was constructed with shape (None, 50) for input KerasTensor(type_spec=TensorSpec(shape=(None, 50), dtype=tf.float32, name='embedding_input'), name='embedding_input', description="created by layer 'embedding_input'"), but it was called on an input with incompatible shape (None,).

我在这里缺少什么？

Answer 1

您的模型需要整数序列，而不是原始字符串。尝试首先将句子转换为相应的整数序列：

sentence = 'I love you more than ever'
print('Sentence:',  sentence)
#print('Emotion:', index_to_class[test_labels[i]])
sentence = get_sequences(tokenizer, np.expand_dims(sentence, axis=0))
p = model.predict(sentence)[0]
pred_class = index_to_class[np.argmax(p).astype('uint8')]
print('Predicted Emotion:', pred_class)

Sentence: I love you more than ever
Predicted Emotion: joy

Answer 2

只是添加一点：

shape

• np.expand_dims（句子）.shape is （1，），而不是（无，50 ）。
• 对于 batch 大小，应该将其扩展一个维度。

序列

• 输入您的模型是一个填充的数字序列，由令牌器转换。
• 它应该长50。