Feature/large doc summarisation #617
Merged
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
…UCE summarisation steps
fixing merge conflict when mergin main
gecBurton
reviewed
Jun 20, 2024
gecBurton
reviewed
Jun 20, 2024
gecBurton
reviewed
Jun 20, 2024
Resolving merge conflicts from pulling in changes from main
lmwilkigov
approved these changes
Jun 20, 2024
| @@ -90,7 +89,7 @@ def make_document_context(input_dict): | |||
| documents += chunks | |||
|
|
|||
| # right now, can only handle a single document so we manually truncate | |||
| max_tokens = 20_000 # parameterise later | |||
| max_tokens = (env.ai.max_tokens,) | |||
There was a problem hiding this comment.
Should we maybe call this something more specific? env.ai.summary_map_chunk_size?
jamesrichards4
approved these changes
Jun 21, 2024
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Outdated suggestions cannot be applied.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.
Context
We want to add the abilty to summarise large documents and multiple documents, with more tokens than can be 'stuffed' into the context window of an LLM.
Changes proposed in this pull request
Added a new
build_map_reduce_summary_chainthat uses Langchain Expression Language (LCEL) to perform a map and then a reduce to achieve large document summarisation.Guidance to review
The map_reduce summarisation is now the default for all summarisation. Will map and reduce everything.
Relevant links
Things to check