Skip to content

Update ollama-chat.adoc #2644

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Update ollama-chat.adoc #2644

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
594300e
Update ollama-chat.adoc
shahbazaamir Apr 5, 2025
dd0a3cc
Added environment variable to support persisting of models to improv…
shahbazaamir Apr 6, 2025
41f508f
Merge pull request #1 from shahbazaamir/main
shahbazaamir Apr 6, 2025
c164492
Added Test cases to compare embeddings similarity
shahbazaamir Apr 6, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/BaseOllamaIT.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,9 @@ public abstract class BaseOllamaIT {
private static final boolean SKIP_CONTAINER_CREATION = Boolean
.parseBoolean(System.getenv().getOrDefault("OLLAMA_WITH_REUSE", "false"));

protected static final boolean KEEP_ADDITIONAL_MODELS = Boolean
.parseBoolean(System.getenv().getOrDefault("PERSIST_MODELS", "false"));

private static OllamaContainer ollamaContainer;

private static final ThreadLocal<OllamaApi> ollamaApi = new ThreadLocal<>();
Expand Down
4 changes: 2 additions & 2 deletions models/spring-ai-ollama/src/test/java/org/springframework/ai/ollama/OllamaChatModelIT.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,8 +77,8 @@ void autoPullModelTest() {
.content();

assertThat(joke).isNotEmpty();

modelManager.deleteModel(ADDITIONAL_MODEL);
if (!KEEP_ADDITIONAL_MODELS)
modelManager.deleteModel(ADDITIONAL_MODEL);
}

@Test
Expand Down
69 changes: 67 additions & 2 deletions .../spring-ai-ollama/src/test/java/org/springframework/ai/ollama/OllamaEmbeddingModelIT.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@
package org.springframework.ai.ollama;

import java.util.List;

import java.util.stream.Collectors;
import org.junit.jupiter.api.Test;

import org.springframework.ai.embedding.EmbeddingRequest;
Expand All @@ -33,6 +33,7 @@
import org.springframework.boot.test.context.SpringBootTest;
import org.springframework.context.annotation.Bean;

import static org.junit.jupiter.api.Assertions.assertTrue;
import static org.assertj.core.api.Assertions.assertThat;

@SpringBootTest
Expand Down Expand Up @@ -80,15 +81,79 @@ void autoPullModelAtStartupTime() {
assertThat(embeddingResponse.getResults()).hasSize(2);
assertThat(embeddingResponse.getResults().get(0).getIndex()).isEqualTo(0);
assertThat(embeddingResponse.getResults().get(0).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(0).getOutput().length).isEqualTo(384);
assertThat(embeddingResponse.getResults().get(1).getIndex()).isEqualTo(1);
assertThat(embeddingResponse.getResults().get(1).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(1).getOutput().length).isEqualTo(384);
assertThat(embeddingResponse.getMetadata().getModel()).contains(ADDITIONAL_MODEL);
assertThat(embeddingResponse.getMetadata().getUsage().getPromptTokens()).isEqualTo(4);
assertThat(embeddingResponse.getMetadata().getUsage().getTotalTokens()).isEqualTo(4);

assertThat(this.embeddingModel.dimensions()).isEqualTo(768);
if (!KEEP_ADDITIONAL_MODELS)
modelManager.deleteModel(ADDITIONAL_MODEL);
}

@Test
void embeddingsSimilarityComparision() {
var model = "all-minilm";
assertThat(this.embeddingModel).isNotNull();

var modelManager = new OllamaModelManager(this.ollamaApi);
assertThat(modelManager.isModelAvailable(ADDITIONAL_MODEL)).isTrue();

EmbeddingResponse embeddingResponse = this.embeddingModel
.call(new EmbeddingRequest(List.of("apple", "banana", "orange", "train", "bus", "taxi"),
OllamaOptions.builder().model(model).truncate(false).build()));

assertThat(embeddingResponse.getResults()).hasSize(6);
assertThat(embeddingResponse.getResults().get(0).getIndex()).isEqualTo(0);
assertThat(embeddingResponse.getResults().get(0).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(1).getIndex()).isEqualTo(1);
assertThat(embeddingResponse.getResults().get(1).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(2).getIndex()).isEqualTo(2);
assertThat(embeddingResponse.getResults().get(2).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(3).getIndex()).isEqualTo(3);
assertThat(embeddingResponse.getResults().get(3).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(4).getIndex()).isEqualTo(4);
assertThat(embeddingResponse.getResults().get(4).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getResults().get(5).getIndex()).isEqualTo(5);
assertThat(embeddingResponse.getResults().get(5).getOutput()).isNotEmpty();
assertThat(embeddingResponse.getMetadata().getModel()).contains(ADDITIONAL_MODEL);

List<Double> similarities = embeddingResponse.getResults()
.stream()
.map(e -> cosineSimilarity(embeddingResponse.getResults().get(0).getOutput(), e.getOutput()))
.collect(Collectors.toList());
assertTrue(similarities.get(1) > similarities.get(3));
assertTrue(similarities.get(1) > similarities.get(4));
assertTrue(similarities.get(1) > similarities.get(5));
assertTrue(similarities.get(2) > similarities.get(3));
assertTrue(similarities.get(2) > similarities.get(4));
assertTrue(similarities.get(2) > similarities.get(5));
List<Double> similarities2 = embeddingResponse.getResults()
.stream()
.map(e -> cosineSimilarity(embeddingResponse.getResults().get(3).getOutput(), e.getOutput()))
.collect(Collectors.toList());
assertTrue(similarities2.get(4) > similarities2.get(0));
assertTrue(similarities2.get(4) > similarities2.get(1));
assertTrue(similarities2.get(4) > similarities2.get(2));
assertTrue(similarities2.get(5) > similarities2.get(0));
assertTrue(similarities2.get(5) > similarities2.get(1));
assertTrue(similarities2.get(5) > similarities2.get(2));
assertThat(this.embeddingModel.dimensions()).isEqualTo(768);
if (!KEEP_ADDITIONAL_MODELS)
modelManager.deleteModel(ADDITIONAL_MODEL);
}

modelManager.deleteModel(ADDITIONAL_MODEL);
private double cosineSimilarity(float[] vecA, float[] vecB) {
double dot = 0.0, magA = 0.0, magB = 0.0;
for (int i = 0; i < 384; i++) {
dot += vecA[i] * vecB[i];
magA += vecA[i] * vecA[i];
magB += vecB[i] * vecB[i];
}
return dot / (Math.sqrt(magA) * Math.sqrt(magB));
}

@SpringBootConfiguration
Expand Down
7 changes: 7 additions & 0 deletions spring-ai-docs/src/main/antora/modules/ROOT/pages/api/chat/ollama-chat.adoc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,13 @@ ollama pull hf.co/<username>/<model-repository>

Alternatively, you can enable the option to download automatically any needed model: xref:auto-pulling-models[Auto-pulling Models].

To see which models are currently available on your system, run

[source,shellscript]
----
ollama list
----

== Auto-configuration

[NOTE]
Expand Down