Many to one and many to many LSTM examples in Keras

by

So:

  1. One-to-one: you could use a Dense layer as you are not processing sequences:

    model.add(Dense(output_size, input_shape=input_shape))

  2. One-to-many: this option is not supported well as chaining models is not very easy in Keras, so the following version is the easiest one:

    model.add(RepeatVector(number_of_times, input_shape=input_shape))
model.add(LSTM(output_size, return_sequences=True))

  3. Many-to-one: actually, your code snippet is (almost) an example of this approach:

    model = Sequential()
model.add(LSTM(1, input_shape=(timesteps, data_dim)))

  4. Many-to-many: This is the easiest snippet when the length of the input and output matches the number of recurrent steps:

    model = Sequential()
model.add(LSTM(1, input_shape=(timesteps, data_dim), return_sequences=True))

  5. Many-to-many when number of steps differ from input/output length: this is freaky hard in Keras. There are no easy code snippets to code that.

EDIT: Ad 5

In one of my recent applications, we implemented something which might be similar to many-to-many from the 4th image. In case you want to have a network with the following architecture (when an input is longer than the output):

                                        O O O
                                        | | |
                                  O O O O O O
                                  | | | | | | 
                                  O O O O O O

You could achieve this in the following manner:

model = Sequential()
model.add(LSTM(1, input_shape=(timesteps, data_dim), return_sequences=True))
model.add(Lambda(lambda x: x[:, -N:, :])) #Select last N from output

Where N is the number of last steps you want to cover (on image N = 3).

From this point getting to:

                                        O O O
                                        | | |
                                  O O O O O O
                                  | | | 
                                  O O O

is as simple as artificial padding sequence of length N using e.g. with 0 vectors, in order to adjust it to an appropriate size.

More Related Contents:

  1. Why binary_crossentropy and categorical_crossentropy give different performances for the same problem?
  2. How to calculate optimal batch size
  3. What is the difference between loss function and metric in Keras? [duplicate]
  4. Keras Text Preprocessing – Saving Tokenizer object to file for scoring
  5. Is deep learning bad at fitting simple non linear functions outside training scope (extrapolating)?
  6. TimeDistributed(Dense) vs Dense in Keras – Same number of parameters
  7. What is the difference between Keras model.evaluate() and model.predict()?
  8. Loss & accuracy – Are these reasonable learning curves?
  9. Common causes of nans during training of neural networks
  10. Keras Sequential model input layer
  11. How to interpret caffe log with debug_info?
  12. Caffe | solver.prototxt values setting strategy
  13. Epoch vs Iteration when training neural networks [closed]
  14. What is `weight_decay` meta parameter in Caffe?
  15. ResNet: 100% accuracy during training, but 33% prediction accuracy with the same data
  16. multi-layer perceptron (MLP) architecture: criteria for choosing number of hidden layers and size of the hidden layer? [closed]
  17. What is the role of TimeDistributed layer in Keras?
  18. How to interpret loss and accuracy for a machine learning model [closed]
  19. Can Keras deal with input images with different size?
  20. Keras: model.evaluate vs model.predict accuracy difference in multi-class NLP task
  21. Caffe: What can I do if only a small batch fits into memory?
  22. How to create caffe.deploy from train.prototxt
  23. How to reduce a fully-connected (`”InnerProduct”`) layer using truncated SVD
  24. keras: how to save the training history attribute of the history object
  25. What is `lr_policy` in Caffe?
  26. Using pre-trained word2vec with LSTM for word generation
  27. Choosing number of Steps per Epoch
  28. How to initialize weights in PyTorch?
  29. How to use fit_generator with multiple inputs
  30. Can I send callbacks to a KerasClassifier?

Leave a Comment