Created
June 12, 2025 10:16
-
-
Save Kerollmops/d2d12ac1898360484638d2b8f2730bb8 to your computer and use it in GitHub Desktop.
A simple example to group documents to embedded, increase the group size to reduce the size of the groups
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
// This is set by the engine for every invocation of this function. | |
let context = #{ group_size: 3, group_id: 1 }; | |
// The _vectors field doesn't even exists. | |
let doc = #{ id: "abc1234", metadata: #{ title: "super title" } }; | |
print(doc); | |
// ---- start ----- | |
fn hash(s) { | |
s.to_chars().reduce(|sum| sum + this.to_int(), 0) | |
} | |
if hash(doc.id) % context.group_size == context.group_id { | |
// Force embeddings regeneration | |
doc._vectors = #{ openai: #{ regenerate: true } }; | |
// Change the template output and trigger the embedder | |
doc.metadata.title = " " + doc.metadata.title; | |
} | |
// ----- end ------ | |
print(doc); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment