我正在尝试创建一个bi-lstm crf来在序列建模场景中分配标签 . 我有很多句子，其中的单词在BIO（开始，内部，外部）方案中被标记，我希望深度神经网络从单词序列和分配的标签中学习，具有向前和向后传播和使用条件随机场 .
我正在遵循https://www.depends-on-the-definition.com/sequence-tagging-lstm-crf/中显示的python实现并尝试在R中实现它 .
我达到了对序列进行排序并填充序列并创建X_Train，Y_Train以进入深度学习网络的程度 .
下面的模型的Python实现（来自上面的链接）
input = Input(shape=(max_len,))
model = Embedding(input_dim=n_words + 1, output_dim=20,
input_length=max_len, mask_zero=True)(input) # 20-dim embedding
model = Bidirectional(LSTM(units=50, return_sequences=True,
recurrent_dropout=0.1))(model) # variational biLSTM
model = TimeDistributed(Dense(50, activation="relu"))(model) # a dense layer as suggested by neuralNer
crf = CRF(n_tags) # CRF layer
out = crf(model) # output
model = Model(input, out)
model.compile(optimizer="rmsprop", loss=crf.loss_function, metrics=[crf.accuracy])
history = model.fit(X_tr, np.array(y_tr), batch_size=32, epochs=5,
validation_split=0.1, verbose=1)
HOW IS THIS TO BE IMPLEMENTED IN R ? 到目前为止我的内容如下，但似乎不正确 .
input <- layer_input(shape = c(max_len,)) ## --- Does not seem correct --
output <- input %>%
layer_embedding(input_dim = n_words + 1, output_dim = 20, input_length=max_len, mask_zero = TRUE) %>%
bidirectional(layer_lstm(units=50, return_sequences = TRUE,recurrent_dropout = 0.1)) %>%
time_distributed(layer_dense(units = 50, activation = "relu"))
## -- Dont know how to create a CRF layer
model <- keras_model(input, output)
model %>% compile(
optimizer = "rmsprop",
loss = "categorical_crossentropy"
)
summary(model)
model %>% fit(S1S2S3XTrain, S1S2S3YTrain, epochs = 5, batch_size = 32)
model %>% evaluate(x_train, y_train)
请帮忙 .