Add support for bulk embedding in Ollama #735
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
…g :text OR :input.
The example req/response looks like this:
```
curl http://localhost:11434/api/embed -d '{
"model": "all-minilm",
"input": ["Why is the sky blue?", "Why is the grass green?"]
}'
{
"model": "all-minilm",
"embeddings": [[
0.010071029, -0.0017594862, 0.05007221, 0.04692972, 0.054916814,
0.008599704, 0.105441414, -0.025878139, 0.12958129, 0.031952348
],[
-0.0098027075, 0.06042469, 0.025257962, -0.006364387, 0.07272725,
0.017194884, 0.09032035, -0.051705178, 0.09951512, 0.09072481
]]
}
```
lib/langchain/llm/ollama.rb
Outdated
| model: defaults[:embeddings_model_name], | ||
| input: [], |
There was a problem hiding this comment.
@bricolage What if we keep the same method interface but instead accept both text: "string" and text: [...]; similar to how OpenAI here takes either a string or an array of strings.
You could also also just always wrap the text below and pass it as an array:
parameters = {
model: model,
input: Array(text)
}There was a problem hiding this comment.
Oh that's nicer. Will do!
lib/langchain/llm/ollama.rb
Outdated
| @@ -219,7 +219,8 @@ def embed( | |||
| ) | |||
| parameters = { | |||
| prompt: text, | |||
There was a problem hiding this comment.
@chalmagean I think we can just remove this prompt: parameter, right?
|
@chalmagean Mind adding "Add support for bulk embedding in Ollama" to CHANGELOG.md? Thank you! |
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Outdated suggestions cannot be applied.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.
Description
As of v0.3.4 Ollama supports bulk embedding. This PR adds support for bulk embeddings via a new
input:keyword paramChanges made
input:keyword param to the#embedmethodtext:keyword param optional (caveat: this will no longer throw an error if you omit it)#embeddingsWhy this change is needed
input:keyword param with an array of strings#embedcurrently requires thetext:keyword param, which is unused when doing bulk embeddingOpen topics
#embedbut pass neithertext:norinput:and it won't throw an error. Is there a preferred way to validate inputs when not relying on keyword argument defaults?