Extending TF Support for RNN & Upsample Layers #314
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Ram81
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
@utsavgarg this PR contains initial work for RNN layer support for TF import.
For handling RNN layers the current logic needed to be tweaked. Difference between other layers and RNN layers is that
RNN layers are composed of multiple basic ops but they don't have layer name common for all the ops or any other pattern in name for combining ops like we do in conv layers.
This PR is based on my observation of each RNN layer ops, it starts with an op named based on TypeofRNNCell+'ZeroState', ex: 'LSTMCellZeroState'. and all the ops following 'rnn' prefix after the ZeroState ops can be used to extract RNN layer parameter.
So I use these properties to assign layer type to multiple ops and use them to extract features of layers.