Keras : How should I prepare input data for RNN?

by

If you only want to predict the output using the most recent 5 inputs, there is no need to ever provide the full 600 time steps of any training sample. My suggestion would be to pass the training data in the following manner:

             t=0  t=1  t=2  t=3  t=4  t=5  ...  t=598  t=599
sample0      |---------------------|
sample0           |---------------------|
sample0                |-----------------
...
sample0                                         ----|
sample0                                         ----------|
sample1      |---------------------|
sample1           |---------------------|
sample1                |-----------------
....
....
sample6751                                      ----|
sample6751                                      ----------|

The total number of training sequences will sum up to

(600 - 4) * 6752 = 4024192    # (nb_timesteps - discarded_tailing_timesteps) * nb_samples

Each training sequence consists of 5 time steps. At each time step of every sequence you pass all 13 elements of the feature vector. Subsequently, the shape of the training data will be (4024192, 5, 13).

This loop can reshape your data:

input = np.random.rand(6752,600,13)
nb_timesteps = 5

flag = 0

for sample in range(input.shape[0]):
    tmp = np.array([input[sample,i:i+nb_timesteps,:] for i in range(input.shape[1] - nb_timesteps + 1)])

    if flag==0:
        new_input = tmp
        flag = 1

    else:
        new_input = np.concatenate((new_input,tmp))

More Related Contents:

  1. TensorFlow and person recognition in video stream
  2. Multivariate LSTM with missing values
  3. Obtaining output of an Intermediate layer in TensorFlow/Keras
  4. Keras LSTM input dimension setting
  5. Keras split train test set when using ImageDataGenerator
  6. In Keras, what exactly am I configuring when I create a stateful `LSTM` layer with N `units`?
  7. How to display custom images in TensorBoard using Keras?
  8. Choosing number of Steps per Epoch
  9. Tensorflow – ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type float)
  10. What is the meaning of the word logits in TensorFlow? [duplicate]
  11. Proper way to feed time-series data to stateful LSTM?
  12. What is the meaning of the “None” in model.summary of KERAS?
  13. Specify connections in NN (in keras)
  14. What are the differences between all these cross-entropy losses in Keras and TensorFlow?
  15. expected ndim=3, found ndim=2
  16. Passing output of a CNN to BILSTM
  17. Does model.compile() initialize all the weights and biases in Keras (tensorflow backend)?
  18. Keras Loss Function with Additional Dynamic Parameter
  19. Removing then Inserting a New Middle Layer in a Keras Model
  20. LSTM module for Caffe
  21. Create keras callback to save model predictions and targets for each batch during training
  22. ValueError: Shapes (None, 1) and (None, 2) are incompatible
  23. Keras, Tensorflow : Merge two different model output into one
  24. Implementing contrastive loss and triplet loss in Tensorflow
  25. Confusion about keras Model: __call__ vs. call vs. predict methods
  26. ValueError: Tensor must be from the same graph as Tensor with Bidirectinal RNN in Tensorflow
  27. How do I plot a Keras/Tensorflow subclassing API model?
  28. Strange behaviour of the loss function in keras model, with pretrained convolutional base
  29. Checkpointing keras model: TypeError: can’t pickle _thread.lock objects
  30. How to replace (or insert) intermediate layer in Keras model?

Leave a Comment