有人可以帮我找到代码中错误源的解决方案吗？

Question

I am training a BI-LSTM-CRF model for a NER task. I was able to build the model, but when I fit it with the training data, googlecolab throw me an error.

Here is the code for my model below (or here : code for my model ):

input_layer = layers.Input(shape=(MAX_SENTENCE,))

model = layers.Embedding(WORD_COUNT, DENSE_EMBEDDING, embeddings_initializer="uniform", input_length=MAX_SENTENCE)(input_layer)

model = layers.Bidirectional(layers.LSTM(LSTM_UNITS, recurrent_dropout=LSTM_DROPOUT, return_sequences=True))(model)

model = layers.TimeDistributed(layers.Dense(DENSE_UNITS, activation="relu"))(model)

crf_layer = CRF(units=TAG_COUNT)

output_layer = crf_layer(model)

ner_model = Model(input_layer, output_layer)

loss = losses.crf_loss

acc_metric = metrics.crf_accuracy

opt = tf.keras.optimizers.Adam(learning_rate=0.001)

ner_model.compile(optimizer=opt, loss=loss, metrics=[acc_metric])

ner_model.summary()

Then after fitting the model, I get the following error:

AttributeError: 'Tensor' object has no attribute '_keras_history' ([error message for google colab][2])

Here is the list of my dependencies : Dependencies

Can someone help me out?

Answer 1

I'm not sure what module you're using because you didn't provide any but I'm guessing that's tensorflow . Try using lower versions of Python like Python 3.7 and also try installing a specific and compatible version: pip install tensorflow==1.13.1