Matt Watson
Matt Watson
This is mostly done, we just need some changes for beam search @abheesht17 is currently working on. Once beam search gains support, we can close this.
Thanks for opening! Definitely this is a good thing to work on, getting efficient function tracing support for sequence generation is important. There's actually two variants to this requirement. -...
Thank you very much for digging into this! Bear with us for a bit here as we figure out an approach we would like to take with data augmentation. We...
If you are looking for something to pick up in the meantime, opened a couple issues tagged with "good first issue" where our design is fully defined. Will try to...
Yes, we have been discussing this. I've been trying to capture some design requirements we have. Here's what I have so far... 1) Augmentation functionality should be exposed as layers...
@aflah02 the augmentations should be applied as operations, but applied before tokenization. A lot of discussion we've had was around 3) above. The layer transformations need to be expressible as...
@aflah02 yeah, I would say maybe rather than EDA maybe we should start with designing a layer for synonym replacement? It's a strict subset of EDA as a whole, we...
@abheesht17 does this make sense to you? Would you like to add this as a minor improvement?
There's a lot more we could do here (e.g. tf.distribute testing), but a simple CPU integration test is probably a good start. @chenmoneygithub assigning you as I think you were...