search

viorik / ConvLSTM

Spatio-temporal video autoencoder with convolutional LSTMs
290 stars 85 forks source link

Input question #21

Open BransonMarvin opened 7 years ago

BransonMarvin commented 7 years ago

Hello,

Thanks for sharing your implementation, I have one question. Is it possible to use a single tensor as input (instead of a table of tensors) ? If it is, then SeqLen x BatchSize x Featsize ( Featsize = NumFeatMaps x Width x Height) is the correct dimension order?

Thanks!

vikram-gupta commented 6 years ago

Yes, you can. Wrap the ConvLSTM in a sequencer.