search

yscacaca / DeepSense

Deepsense: a unified deep learning framework for time-series mobile sensing data processing.
196 stars 71 forks source link

Does this kind of convolution preserve the time time distributed convolution ? #11

Closed shamanez closed 7 years ago

shamanez commented 7 years ago

In the keras there's TimeDistributed wrapper where we can use to do convolution for each time step. So I think by feeding the tensor with Batch Size , Sequence Length , .. etc we can have the same property. Am I right ?

yscacaca commented 7 years ago

The TimeDistributed wrapper should have the same property we want. You probably can set keras.layers.convolutional.Conv1D with the TimeDistributed wrapper.

shamanez commented 7 years ago

Yeah. So main thing is we need to feed a tensor with sequence .