我在尝试为句子分类实现一个卷积神经网络（CNN）；我试图遵循论文中提出的架构。我使用Keras（搭配TensorFlow）来实现这一点。以下是我模型的摘要：
____________________________________________________________________________________________________Layer (type) Output Shape Param # Connected to ====================================================================================================input_4 (InputLayer) (None, 56) 0 ____________________________________________________________________________________________________embedding (Embedding) (None, 56, 300) 6510000 ____________________________________________________________________________________________________dropout_7 (Dropout) (None, 56, 300) 0 ____________________________________________________________________________________________________conv1d_10 (Conv1D) (None, 54, 100) 90100 ____________________________________________________________________________________________________conv1d_11 (Conv1D) (None, 53, 100) 120100 ____________________________________________________________________________________________________conv1d_12 (Conv1D) (None, 52, 100) 150100 ____________________________________________________________________________________________________max_pooling1d_10 (MaxPooling1D) (None, 27, 100) 0 ____________________________________________________________________________________________________max_pooling1d_11 (MaxPooling1D) (None, 26, 100) 0 ____________________________________________________________________________________________________max_pooling1d_12 (MaxPooling1D) (None, 26, 100) 0 ____________________________________________________________________________________________________flatten_10 (Flatten) (None, 2700) 0 ____________________________________________________________________________________________________flatten_11 (Flatten) (None, 2600) 0 ____________________________________________________________________________________________________flatten_12 (Flatten) (None, 2600) 0 ____________________________________________________________________________________________________concatenate_4 (Concatenate) (None, 7900) 0 ____________________________________________________________________________________________________dropout_8 (Dropout) (None, 7900) 0 ____________________________________________________________________________________________________dense_7 (Dense) (None, 50) 395050 ____________________________________________________________________________________________________dense_8 (Dense) (None, 5) 255 ====================================================================================================Total params: 7,265,605.0Trainable params: 7,265,605.0Non-trainable params: 0.0使用给定的架构，我遇到了严重的过拟合问题。以下是我的结果：
我无法理解过拟合的原因，请建议我对架构进行一些更改以避免这种情况。如果您需要更多信息，请告诉我。
源代码：
if model_type in ['CNN-non-static', 'CNN-static']: embedding_wts = train_word2vec( np.vstack((x_train, x_test, x_valid)), ind_to_wrd, num_features = embedding_dim) if model_type == 'CNN-static': x_train = embedding_wts[0][x_train] x_test = embedding_wts[0][x_test] x_valid = embedding_wts[0][x_valid]elif model_type == 'CNN-rand': embedding_wts = Noneelse: raise ValueError("Unknown model type")batch_size = 50filter_sizes = [3,4,5]num_filters = 75dropout_prob = (0.5, 0.8)hidden_dims = 50l2_reg = 0.3# Deciding dimension of input based on the modelinput_shape = (max_sent_len, embedding_dim) if model_type == "CNN-static" else (max_sent_len,)model_input = Input(shape = input_shape)# Static model do not have embedding layerif model_type == "CNN-static": z = Dropout(dropout_prob[0])(model_input)else: z = Embedding(vocab_size, embedding_dim, input_length = max_sent_len, name="embedding")(model_input) z = Dropout(dropout_prob[0])(z)# Convolution layersz1 = Conv1D( filters=num_filters, kernel_size=3, padding="valid", activation="relu", strides=1)(z)z1 = MaxPooling1D(pool_size=2)(z1)z1 = Flatten()(z1)z2 = Conv1D( filters=num_filters, kernel_size=4, padding="valid", activation="relu", strides=1)(z)z2 = MaxPooling1D(pool_size=2)(z2)z2 = Flatten()(z2)z3 = Conv1D( filters=num_filters, kernel_size=5, padding="valid", activation="relu", strides=1)(z)z3 = MaxPooling1D(pool_size=2)(z3)z3 = Flatten()(z3)# Concatenate the output of all convolution layersz = Concatenate()([z1, z2, z3])z = Dropout(dropout_prob[1])(z)# Dense(64, input_dim=64, kernel_regularizer=regularizers.l2(0.01), activity_regularizer=regularizers.l1(0.01))z = Dense(hidden_dims, activation="relu", kernel_regularizer=regularizers.l2(0.01))(z)model_output = Dense(N_category, activation="sigmoid")(z)model = Model(model_input, model_output)model.compile(loss="categorical_crossentropy", optimizer=optimizers.Adadelta(lr=1, decay=0.005), metrics=["accuracy"])model.summary()回答：
