尝试在 Keras 中标记文本时出错？

Question

Keras 和深度学习非常新，但我正在遵循在线指南，我正在尝试标记我的文本，以便在我为神经网络创建层时可以访问“形状”以用作“input_shape”。到目前为止，这是我的代码：df = pd.read_csv(pathname, encoding = "ISO-8859-1")df = df[['content_cleaned', 'meaningful']]df = df.sample(frac=1)#Transposed columns into numpy arrays X = np.asarray(df[['content_cleaned']])y = np.asarray(df[['meaningful']])#Split into training and testing setX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=21) # Create tokenizertokenizer = Tokenizer(num_words=100) #No row has more than 100 words.#Tokenize the predictors (text)X_train = np.concatenate(tokenizer.sequences_to_matrix(int(X_train), mode="binary"))X_test = np.concatenate(tokenizer.sequences_to_matrix(int(X_test), mode="binary"))#Convert the labels to the binaryencoder = LabelBinarizer()encoder.fit(y_train) y_train = encoder.transform(y_train)y_test = encoder.transform(y_test)错误突出显示：X_train = tokenizer.sequences_to_matrix(int(X_train), mode="binary")错误信息是：TypeError: only length-1 arrays can be converted to Python scalars任何人都可以发现我的错误并可能为此提供解决方案吗？我对此很陌生，无法解决此问题。我希望能够调用“X_train.shape”，以便在创建网络层时将其输入到 input_shape 中。任何帮助都会很棒！

Answer 1

您正在尝试将 numpy 数组转换为 python 整数，这当然是不可能的，并且会给您错误（该错误与 Keras 无关）。您真正想要做的是dtype将该 numpy 数组的更改为int. 请尝试以下操作：

X_train.astype(np.int32)

代替 int(X_train)