github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-24 15:43:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

experimental: Fix divide by 0 error (#25439)

Browse Source 
Within the semantic chunker, when calling `_threshold_from_clusters`
there is the possibility for a divide by 0 error if the
`number_of_chunks` is equal to the length of `distances`.

Fix simply implements a check if these values match to prevent the error
and enable chunking to continue.
This commit is contained in:
Luke 2024-08-16 00:46:30 +10:00 committed by GitHub
parent ba167dc158
commit 66e30efa61
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
libs/experimental/langchain_experimental/text_splitter.py
View File

@ -180,7 +180,11 @@ class SemanticChunker(BaseDocumentTransformer):
x = max(min(self.number_of_chunks, x1), x2)
# Linear interpolation formula
y = y1 + ((y2 - y1) / (x2 - x1)) * (x - x1)
if x2 == x1:
y = y2
else:
y = y1 + ((y2 - y1) / (x2 - x1)) * (x - x1)
y = min(max(y, 0), 100)
return cast(float, np.percentile(distances, y))