This repository was archived by the owner on Jul 1, 2023. It is now read-only.
Add an Embedding layer #257
Merged
Conversation
This file contains hidden or 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
rxwei
reviewed
Jun 18, 2019
Add space between colon and protocol name. Co-Authored-By: Richard Wei <[email protected]>
Correct misspelling. Co-Authored-By: Richard Wei <[email protected]>
Remove unneeded line break in documentation. Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Yep, the Swift for TensorFlow team plans to implement default generic arguments at some point if it's not yet done by the Swift community. |
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
Co-Authored-By: Richard Wei <[email protected]>
… into layer/embedding
rxwei
approved these changes
Jun 18, 2019
|
@rxwei Glad to help out! Sorry for all the tiny mistakes. I'll be much more careful on any future PRs I submit. |
jekbradbury
approved these changes
Jun 18, 2019
| @frozen | ||
| public struct EmbeddingInput: Differentiable { | ||
| /// Sequences of indices that will be passed into the layer. | ||
| @noDerivative public var indices: Tensor<Int32> |
There was a problem hiding this comment.
It's absolutely possible to have an embedding layer with more than 2^32 embeddings (and therefore one that requires Int64 indices), but it's fine to not worry about that for now.
There was a problem hiding this comment.
gathering(atIndices:alongAxis:) is not yet generic over BinaryInteger. To fix this, we need to start from there.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Add support for embeddings.
#54
It might be a good idea to find a type alias for this layer to avoid having to supply a useless generic type argument, e.g.
Embedding<Float>, since aTensorFlowFloatingPointis not the true type (Int) of input into the layer. Maybe Swift will soon implement default generic parameters?