Keras LSTM input dimension setting

by

For the sake of completeness, here’s what’s happened.

First up, LSTM, like all layers in Keras, accepts two arguments: input_shape and batch_input_shape. The difference is in convention that input_shape does not contain the batch size, while batch_input_shape is the full input shape including the batch size.

Hence, the specification input_shape=(None, 20, 64) tells keras to expect a 4-dimensional input, which is not what you want. The correct would have been just (20,).

But that’s not all. LSTM layer is a recurrent layer, hence it expects a 3-dimensional input (batch_size, timesteps, input_dim). That’s why the correct specification is input_shape=(20, 1) or batch_input_shape=(10000, 20, 1). Plus, your training array should also be reshaped to denote that it has 20 time steps and 1 input feature per each step.

Hence, the solution:

X_train = np.expand_dims(X_train, 2)  # makes it (10000,20,1)
...
model = Sequential()
model.add(LSTM(..., input_shape=(20, 1)))

More Related Contents:

  1. Understanding Keras LSTMs
  2. Keras, How to get the output of each layer?
  3. Tensorflow – ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type float)
  4. How to apply gradient clipping in TensorFlow?
  5. Proper way to feed time-series data to stateful LSTM?
  6. Keras replacing input layer
  7. Keras custom loss function: Accessing current input pattern
  8. expected ndim=3, found ndim=2
  9. Passing output of a CNN to BILSTM
  10. How can I use a pre-trained neural network with grayscale images?
  11. Dimension of shape in conv1D
  12. Removing then Inserting a New Middle Layer in a Keras Model
  13. Custom loss function in Keras
  14. How do I get the weights of a layer in Keras?
  15. Multiple outputs in Keras
  16. ValueError: Input 0 of layer sequential is incompatible with the layer: : expected min_ndim=4, found ndim=3. Full shape received: [8, 28, 28]
  17. How to implement Grad-CAM on a trained network
  18. How to fix “ResourceExhaustedError: OOM when allocating tensor”
  19. How do I plot a Keras/Tensorflow subclassing API model?
  20. Strange behaviour of the loss function in keras model, with pretrained convolutional base
  21. Multi-input Multi-output Model with Keras Functional API
  22. Using a pre-trained word embedding (word2vec or Glove) in TensorFlow
  23. What is the difference between ‘SAME’ and ‘VALID’ padding in tf.nn.max_pool of tensorflow?
  24. TypeError: __array__() takes 1 positional argument but 2 were given (Image classification Keras)
  25. Tensorflow 2.0 – AttributeError: module ‘tensorflow’ has no attribute ‘Session’
  26. Neural Network LSTM input shape from dataframe
  27. Keras: .predict returns percentages instead of classes
  28. Keras Custom loss function to pass arguments other than y_true and y_pred
  29. Unknown initializer: GlorotUniform when loading Keras model
  30. looking for source code of from gen_nn_ops in tensorflow

Leave a Comment