Support list/tuple inputs for special tokens in MultiSegmentPacker layer #1046
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
mattdangerw
requested changes
May 17, 2023
| @@ -53,12 +53,16 @@ class MultiSegmentPacker(keras.layers.Layer): | |||
|
|
|||
| Args: | |||
| sequence_length: The desired output length. | |||
There was a problem hiding this comment.
let's add the type notes you have in the other layer for consistency
| dtype of the input tensors to the layer. | ||
| end_value: The id(s) or token(s) that is/are to be placed at the end of |
| dtype of the input tensors to the layer. | ||
| end_value: The id(s) or token(s) that is/are to be placed at the end of | ||
| the last input segment (called "[SEP]" for BERT). The dtype much |
| sep_value: The id(s) or token(s) that is/are to be placed at the end of | ||
| every segment, except the last segment (called "[SEP]" for BERT). | ||
| If `None`, `end_value` is used. The dtype much match the dtype of | ||
| the input tensors to the layer. | ||
| pad_value: The id or token that is to be placed into the unused | ||
| positions after the last segment in the sequence | ||
| (called "[PAD]" for BERT). |
There was a problem hiding this comment.
should we add an example below? maybe roberta double sep?
| start_column = tf.fill((batch_size, 1), start_value) | ||
| end_column = tf.fill((batch_size, 1), end_value) | ||
| ones_column = tf.ones_like(start_column, dtype=tf.int32) | ||
| start_values_tensor = tf.repeat( |
There was a problem hiding this comment.
these names are a little confusing start_value is already a tensor. should we co back to _column naming?
start_column, end_column, sep_column?
There was a problem hiding this comment.
We can, but it isn't exactly a column :P. I'll call it start_columns
| ( | ||
| [ | ||
| [ | ||
| "[CLS]", |
There was a problem hiding this comment.
try to come up with a slightly shorter test case that will format the lists to one line
|
/gcbrun |
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.
No description provided.