I have read it in a paper https://cs230.stanford.edu/projects_spring_2018/reports/8291197.pdf, convLSTM 3x3-256-2 means convLSTM with 3x3 filter size, 256 hidden states, and 2 layers.
But the original https://keras.io/api/layers/recurrent_layers/conv_lstm2d/ do not show any argument regarding hidden states and layers.
Please elaborate this text, and how do we write it in the python code?
Regards
Remal Al-Gounmeein
See these links:
https://paperswithcode.com/paper/convolutional-lstm-network-a-machine-learning/review/
http://proceedings.mlr.press/v80/jang18a/jang18a-supp.pdf
0 votes
0 thanks
Amin Honarmandi Shandiz
hi Tariq Sm ,
In my Side of View it means two ConvLSTM layers, which it means you can use two of that source code with each 256 filters and 3*3 kernel size.
(good luck)
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing
More Tariq Sm's questions See All
Similar questions and discussions