https://www.kaggle.com/anunnikrishnan/keras-classifier-based-spam-classification
基于此笔记本，预测功能的外观如何，将一条短信作为测试输入？
sms = " You won a free giftcard "
max_words = 1000
max_len = 150
tok = Tokenizer(num_words=max_words)
tok.fit_on_texts(sms)
sequences = tok.texts_to_sequences(text)
sms_matrix = sequence.pad_sequences(sequences,maxlen=max_len)
prediction = model.predict(sms_matrix)
结果看起来像这样
array([[0.00395709],
[0.00090399],
[0.00528631],
[0.00549346],
[0.00347543],
[0.00090399],
[0.00751105],
[0.00549346],
[0.00736764],
[0.00549346],
[0.00568002],
[0.00696304],
[0.00551164],
[0.00551164],
[0.00549346],
[0.0014942 ],
[0.00137451],
[0.00568 ],
[0.01121344],
[0.01121931],
[0.00736762],
[0.00696306],
[0.00619166]], dtype=float32)
我在以二进制方式解释这些结果时遇到了麻烦。