我编写了以下模型函数:
from tensorflow.keras.layers import Dense, LSTM, Dropout, Input, BatchNormalizationfrom tensorflow.keras.models import Modelfrom tensorflow.keras.optimizers import Adamimport tensorflow_addons as tfaimport tensorflow as tfdef get_model(num_features, output_size, output_bias=None): output_bias = tf.keras.initializers.Constant(output_bias) opt = Adam(learning_rate=0.0008) inputs = Input(shape=[None, num_features], dtype=tf.float32, ragged=True) layers = LSTM(32, activation='tanh')( inputs.to_tensor(), mask=tf.sequence_mask(inputs.row_lengths())) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(32, activation='relu')(layers) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(32, activation='relu')(layers) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(output_size, activation='sigmoid', bias_initializer=output_bias)(layers) model = Model(inputs, layers) model.compile(loss=tf.keras.losses.BinaryCrossentropy(), optimizer=opt, metrics=[tfa.metrics.F1Score(num_classes=2)]) model.summary() return model这是模型摘要:
Model: "model_5"__________________________________________________________________________________________________Layer (type) Output Shape Param # Connected to ==================================================================================================input_6 (InputLayer) [(None, None, 11)] 0 __________________________________________________________________________________________________input.row_lengths_5 (InstanceMe (None,) 0 input_6[0][0] __________________________________________________________________________________________________input.to_tensor_5 (InstanceMeth (None, None, 11) 0 input_6[0][0] __________________________________________________________________________________________________tf.sequence_mask_5 (TFOpLambda) (None, None) 0 input.row_lengths_5[0][0] __________________________________________________________________________________________________lstm_5 (LSTM) (None, 32) 5632 input.to_tensor_5[0][0] tf.sequence_mask_5[0][0] __________________________________________________________________________________________________batch_normalization_15 (BatchNo (None, 32) 128 lstm_5[0][0] __________________________________________________________________________________________________dropout_15 (Dropout) (None, 32) 0 batch_normalization_15[0][0] __________________________________________________________________________________________________dense_15 (Dense) (None, 32) 1056 dropout_15[0][0] __________________________________________________________________________________________________batch_normalization_16 (BatchNo (None, 32) 128 dense_15[0][0] __________________________________________________________________________________________________dropout_16 (Dropout) (None, 32) 0 batch_normalization_16[0][0] __________________________________________________________________________________________________dense_16 (Dense) (None, 32) 1056 dropout_16[0][0] __________________________________________________________________________________________________batch_normalization_17 (BatchNo (None, 32) 128 dense_16[0][0] __________________________________________________________________________________________________dropout_17 (Dropout) (None, 32) 0 batch_normalization_17[0][0] __________________________________________________________________________________________________dense_17 (Dense) (None, 1) 33 dropout_17[0][0] ==================================================================================================Total params: 8,161Trainable params: 7,969Non-trainable params: 192__________________________________________________________________________________________________以下是我的数据形状:
print(train_x.shape,train_y.shape)print(val_x.shape,val_y.shape)(52499, None, 11) (52499,)(17500, None, 11) (17500,)在尝试拟合我的模型时，我遇到了以下错误:
model.fit(train_x, train_y, epochs=300, batch_size=500, validation_data=(val_x, val_y))ValueError: Dimension 0 in both shapes must be equal, but are 2 and 1. Shapes are [2] and [1].我无法理解形状有什么问题。
回答：
你的模型看起来没问题。问题是你遇到了一个关于tfa.metrics.F1Score的未解决的issue。对于你的二分类情况，你需要将F1Score的参数改为tfa.metrics.F1Score(num_classes=1, threshold=0.5)。这是一个完整的可工作示例:
from tensorflow.keras.layers import Dense, LSTM, Dropout, Input, BatchNormalizationfrom tensorflow.keras.models import Modelfrom tensorflow.keras.optimizers import Adamimport tensorflow_addons as tfaimport tensorflow as tfdef get_model(num_features, output_size, output_bias=0.001): output_bias = tf.keras.initializers.Constant(output_bias) opt = Adam(learning_rate=0.0008) inputs = Input(shape=[None, num_features], dtype=tf.float32, ragged=True) layers = LSTM(32, activation='tanh')( inputs.to_tensor(), mask=tf.sequence_mask(inputs.row_lengths())) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(32, activation='relu')(layers) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(32, activation='relu')(layers) layers = BatchNormalization()(layers) layers = Dropout(0.05)(layers) layers = Dense(output_size, activation='sigmoid', bias_initializer=output_bias)(layers) model = Model(inputs, layers) model.compile(loss=tf.keras.losses.BinaryCrossentropy(), optimizer=opt, metrics=[tfa.metrics.F1Score(num_classes=1, threshold=0.5)]) model.summary() return modelmodel = get_model(11, 1)rt = tf.RaggedTensor.from_row_splits(values=tf.ones([5, 11], tf.int32), row_splits=[0, 2, 5])model.fit(rt, tf.random.uniform((2,1), maxval=2), epochs=300, batch_size=2, verbose=2)或者，你可以定义自己的F1Score方法，并将其设置为模型中的指标。请参阅这个帖子获取更多信息。
