我想用三维数据(Train_value),一维标记数据(Train_label_list)进行LSTM。但是它有错误(ValueError)。我修不好它 from keras.models import Sequentialy_train = y_train.reshape(y_train.shape[0])
mod
我有一个Keras模型,我训练它来预测下一个序列:from= Sequential( Input(shape= (500,5)), #500 arrays like this one -> [0,0,0,0,0]model.fit(corpuswithids, np.asarray(ydata), batch_s
然而,当我尝试以不同的方式对它进行编码时,我会得到一个值错误(参见下面)。知道怎么解决这个问题吗?\backend\tensorflow_backend.py", line 468, in is_keras_tensorValueError: Unexpectedly found an instance of type
当我试图在LSTM模型中添加validation_split时,我得到了以下错误
ValueError: `validation_split` is only supported for Tensorsor NumPy arrays, found: (<tensorflow.python.keras.preprocessing.sequence.TimeseriesGenerator object)train_generator, epochs=50,verbose=2,callbacks=[tensorboard_callback]