You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The current APIs for running models in RTen are annoying to use if a model has many inputs or outputs. The current APIs require a list of (NodeId, Input) tuples to be passed for the input, and a list of NodeIds of outputs. It would however be more ergonomic if:
The inputs/outputs could be specified as names, without requiring the NodeId lookup step first
There was a convenient way to build up the inputs as a hash map
Outputs could easily be retrieved by name rather than index
The current APIs for running models in RTen are annoying to use if a model has many inputs or outputs. The current APIs require a list of
(NodeId, Input)
tuples to be passed for the input, and a list ofNodeId
s of outputs. It would however be more ergonomic if:NodeId
lookup step firstThis issue originally came up when porting the Tacotron2 TTS demo from https://github.com/xd009642/xd-tts/ to RTen. See page 38 of https://github.com/xd009642/xd-tts/blob/master/slides/slides.pdf for an example of why this matters.
The text was updated successfully, but these errors were encountered: