This is an automated email from the ASF dual-hosted git repository.
jamesnetherton pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/camel-quarkus-examples.git
commit 1b2eefd2c1eec5bae4a7e836fa427a42bcc71ba2
Author: aldettinger <aldettin...@gmail.com>
AuthorDate: Tue Oct 29 10:54:25 2024 +0100
Data extract switch to granite3
---
data-extract-langchain4j/README.adoc | 8 ++++----
.../src/main/resources/application.properties | 2 +-
.../org/acme/extraction/OllamaTestResource.java | 15 ++++++++++++--
..._chat-495066d1-9278-4e4b-b8f6-a1c2fa296779.json | 24 ----------------------
..._chat-52961581-e5b1-4309-b62f-1c6e1e0008eb.json | 24 ----------------------
..._chat-5c1f926c-0480-41e9-9ca7-a93e17919e99.json | 24 ----------------------
..._chat-75e11d9e-fdc3-4873-847d-982daec53233.json | 24 ++++++++++++++++++++++
..._chat-861b9f1b-91c6-4206-8c8b-7e393c33d16c.json | 24 ++++++++++++++++++++++
..._chat-b837ed7e-cb2a-43f4-84a9-8522b0c4d34d.json | 24 ++++++++++++++++++++++
9 files changed, 90 insertions(+), 79 deletions(-)
diff --git a/data-extract-langchain4j/README.adoc
b/data-extract-langchain4j/README.adoc
index 00ea497..f0854e7 100644
--- a/data-extract-langchain4j/README.adoc
+++ b/data-extract-langchain4j/README.adoc
@@ -15,7 +15,7 @@ In this example, we'll convert those text conversations into
Java Objects that c
image::schema.png[]
In order to achieve this extraction, we'll need a
https://en.wikipedia.org/wiki/Large_language_model[Large Language Model (LLM)]
that natively supports JSON output.
-Here, we arbitrarily choose https://ollama.com/library/codellama[codellama]
served through https://ollama.com/[ollama].
+Here, we choose https://ollama.com/library/granite3-dense[granite3-dense]
served through https://ollama.com/[ollama] as it seems compute friendly and
under Apache V2 license.
In order to request inference to the served model, we'll use the high-level
LangChain4j APIs like
https://docs.langchain4j.dev/tutorials/ai-services[AiServices].
More precisely, we'll setup the
https://docs.quarkiverse.io/quarkus-langchain4j/dev/index.html[Quarkus
LangChain4j extension] to register an AiService bean.
Finally, we'll invoke the AiService extraction method via the
https://camel.apache.org/camel-quarkus/latest/reference/extensions/bean.html[Camel
Quarkus bean extension] .
@@ -26,7 +26,7 @@ Let's start a container to serve the LLM with Ollama, in a
first shell type:
[source,shell]
----
-docker run --rm -it -v cqex-data-extract-ollama:/root/.ollama -p 11434:11434
--name cqex-data-extract-ollama ollama/ollama:0.3.12
+docker run --rm -it -v cqex-data-extract-ollama:/root/.ollama -p 11434:11434
--name cqex-data-extract-ollama ollama/ollama:0.4.0-rc5
----
After a moment, a log like below should be output:
@@ -40,7 +40,7 @@ Then, download the codellama model, in a second shell type:
[source,shell]
----
-docker exec -it cqex-data-extract-ollama ollama pull codellama
+docker exec -it cqex-data-extract-ollama ollama pull granite3-dense
----
After a moment, log like below should be output:
@@ -106,7 +106,7 @@ At the end, we are provided with a Plain Old Java Object
(POJO) handling the ext
"customerSatisfied": "true",
"customerName": "Sarah London",
"customerBirthday": "10 July 1986",
- "summary": "Declare an accident on main vehicle and receive reimbursement
for expenses."
+ "summary": "The customer, Sarah London, called to declare an accident on
her main vehicle and was informed that all expenses related to the accident
would be reimbursed."
}
----
diff --git a/data-extract-langchain4j/src/main/resources/application.properties
b/data-extract-langchain4j/src/main/resources/application.properties
index 633c2db..da692fb 100644
--- a/data-extract-langchain4j/src/main/resources/application.properties
+++ b/data-extract-langchain4j/src/main/resources/application.properties
@@ -20,7 +20,7 @@ quarkus.banner.enabled = false
# Configure Quarkus LangChain4j that handle interactions with the Large
Language Model
quarkus.langchain4j.ollama.base-url = http://localhost:11434
quarkus.langchain4j.ollama.timeout = 3m
-quarkus.langchain4j.ollama.chat-model.model-id = codellama
+quarkus.langchain4j.ollama.chat-model.model-id = granite3-dense
quarkus.langchain4j.ollama.chat-model.format = json
quarkus.langchain4j.ollama.chat-model.temperature = 0
# Uncomment lines below to log Ollama client requests and responses
diff --git
a/data-extract-langchain4j/src/test/java/org/acme/extraction/OllamaTestResource.java
b/data-extract-langchain4j/src/test/java/org/acme/extraction/OllamaTestResource.java
index b9cebd6..08ef690 100644
---
a/data-extract-langchain4j/src/test/java/org/acme/extraction/OllamaTestResource.java
+++
b/data-extract-langchain4j/src/test/java/org/acme/extraction/OllamaTestResource.java
@@ -23,10 +23,12 @@ import com.github.tomakehurst.wiremock.WireMockServer;
import io.quarkus.test.common.QuarkusTestResourceLifecycleManager;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
+import org.testcontainers.containers.Container.ExecResult;
import org.testcontainers.containers.output.Slf4jLogConsumer;
import org.testcontainers.ollama.OllamaContainer;
import static java.lang.String.format;
+import static java.lang.System.currentTimeMillis;
import static org.eclipse.microprofile.config.ConfigProvider.getConfig;
@@ -34,7 +36,7 @@ public class OllamaTestResource implements
QuarkusTestResourceLifecycleManager {
private static final Logger LOG =
LoggerFactory.getLogger(OllamaTestResource.class);
- private static final String OLLAMA_IMAGE = "ollama/ollama:0.3.12";
+ private static final String OLLAMA_IMAGE = "ollama/ollama:0.4.0-rc5";
private static final int OLLAMA_SERVER_PORT = 11434;
private static final String MODE_MOCK = "mock";
@@ -84,7 +86,16 @@ public class OllamaTestResource implements
QuarkusTestResourceLifecycleManager {
String ollamaModelId =
getConfig().getValue("quarkus.langchain4j.ollama.chat-model.model-id",
String.class);
- ollamaContainer.execInContainer("ollama", "pull",
ollamaModelId);
+ ExecResult result = ollamaContainer.execInContainer("ollama",
"pull", ollamaModelId);
+ long pullBegin = currentTimeMillis();
+ while ((currentTimeMillis() - pullBegin < 10000)
+ && (result.getStderr() == null ||
!result.getStderr().contains("success"))) {
+ LOG.info("Will retry ollama pull after sleeping 250ms");
+
+ Thread.sleep(250);
+
+ result = ollamaContainer.execInContainer("ollama", "pull",
ollamaModelId);
+ }
baseUrl = format(BASE_URL_FORMAT, ollamaContainer.getHost(),
ollamaContainer.getMappedPort(OLLAMA_SERVER_PORT));
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-495066d1-9278-4e4b-b8f6-a1c2fa296779.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-495066d1-9278-4e4b-b8f6-a1c2fa296779.json
deleted file mode 100644
index ba1c2d3..0000000
---
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-495066d1-9278-4e4b-b8f6-a1c2fa296779.json
+++ /dev/null
@@ -1,24 +0,0 @@
-{
- "id" : "495066d1-9278-4e4b-b8f6-a1c2fa296779",
- "name" : "api_chat",
- "request" : {
- "url" : "/api/chat",
- "method" : "POST",
- "bodyPatterns" : [ {
- "equalToJson" : "{\n \"model\" : \"codellama\",\n \"messages\" : [ {\n
\"role\" : \"assistant\",\n \"content\" :
\"{\\n\\\"customerSatisfied\\\": false,\\n\\\"customerName\\\": \\\"John
Doe\\\",\\n\\\"customerBirthday\\\": \\\"2001-11-01\\\",\\n\\\"summary\\\":
\\\"Insurance company failed to notify customer of automatic cancellation of
full reimbursement option and only provided half reimbursement for
accident.\\\"\\n}\"\n }, {\n \"role\" : \"user\",\n \"content\" : [...]
- "ignoreArrayOrder" : true,
- "ignoreExtraElements" : true
- } ]
- },
- "response" : {
- "status" : 200,
- "body" :
"{\"model\":\"codellama\",\"created_at\":\"2024-08-28T16:54:19.439835677Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
true,\\n\\\"customerName\\\": \\\"Kate Boss\\\",\\n\\\"customerBirthday\\\":
\\\"1999-08-13\\\",\\n\\\"summary\\\": \\\"Customer was unable to find their
insurance contract and had to go through a process of updating their name on
the contract. The operator provided assistance throughout the process, ensuring
that the cu [...]
- "headers" : {
- "Date" : "Wed, 28 Aug 2024 16:54:19 GMT",
- "Content-Type" : "application/json; charset=utf-8"
- }
- },
- "uuid" : "495066d1-9278-4e4b-b8f6-a1c2fa296779",
- "persistent" : true,
- "insertionIndex" : 4
-}
\ No newline at end of file
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-52961581-e5b1-4309-b62f-1c6e1e0008eb.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-52961581-e5b1-4309-b62f-1c6e1e0008eb.json
deleted file mode 100644
index 9a626a0..0000000
---
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-52961581-e5b1-4309-b62f-1c6e1e0008eb.json
+++ /dev/null
@@ -1,24 +0,0 @@
-{
- "id" : "52961581-e5b1-4309-b62f-1c6e1e0008eb",
- "name" : "api_chat",
- "request" : {
- "url" : "/api/chat",
- "method" : "POST",
- "bodyPatterns" : [ {
- "equalToJson" : "{\n \"model\" : \"codellama\",\n \"messages\" : [ {\n
\"role\" : \"user\",\n \"content\" : \"Extract information about a
customer from the text delimited by triple backticks: ```Operator: Hello, how
may I help you ?\\nCustomer: Hello, I'm calling because I need to declare an
accident on my main vehicle.\\nOperator: Ok, can you please give me your name
?\\nCustomer: My name is Sarah London.\\nOperator: Could you please give me
your birth date ?\\nCustomer: 1 [...]
- "ignoreArrayOrder" : true,
- "ignoreExtraElements" : true
- } ]
- },
- "response" : {
- "status" : 200,
- "body" :
"{\"model\":\"codellama\",\"created_at\":\"2024-08-28T16:52:55.508750951Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
true,\\n\\\"customerName\\\": \\\"Sarah
London\\\",\\n\\\"customerBirthday\\\": \\\"1986-07-10\\\",\\n\\\"summary\\\":
\\\"Declare an accident on main vehicle and receive reimbursement for
expenses.\\\"\\n}\"},\"done_reason\":\"stop\",\"done\":true,\"total_duration\":22737688834,\"load_duration\":848753776,\"prompt_eval_
[...]
- "headers" : {
- "Date" : "Wed, 28 Aug 2024 16:52:55 GMT",
- "Content-Type" : "application/json; charset=utf-8"
- }
- },
- "uuid" : "52961581-e5b1-4309-b62f-1c6e1e0008eb",
- "persistent" : true,
- "insertionIndex" : 6
-}
\ No newline at end of file
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-5c1f926c-0480-41e9-9ca7-a93e17919e99.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-5c1f926c-0480-41e9-9ca7-a93e17919e99.json
deleted file mode 100644
index 9ed4cfe..0000000
---
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-5c1f926c-0480-41e9-9ca7-a93e17919e99.json
+++ /dev/null
@@ -1,24 +0,0 @@
-{
- "id" : "5c1f926c-0480-41e9-9ca7-a93e17919e99",
- "name" : "api_chat",
- "request" : {
- "url" : "/api/chat",
- "method" : "POST",
- "bodyPatterns" : [ {
- "equalToJson" : "{\n \"model\" : \"codellama\",\n \"messages\" : [ {\n
\"role\" : \"assistant\",\n \"content\" :
\"{\\n\\\"customerSatisfied\\\": true,\\n\\\"customerName\\\": \\\"Sarah
London\\\",\\n\\\"customerBirthday\\\": \\\"1986-07-10\\\",\\n\\\"summary\\\":
\\\"Declare an accident on main vehicle and receive reimbursement for
expenses.\\\"\\n}\"\n }, {\n \"role\" : \"user\",\n \"content\" :
\"Extract information about a customer from the text delimited by trip [...]
- "ignoreArrayOrder" : true,
- "ignoreExtraElements" : true
- } ]
- },
- "response" : {
- "status" : 200,
- "body" :
"{\"model\":\"codellama\",\"created_at\":\"2024-08-28T16:53:37.033889726Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
false,\\n\\\"customerName\\\": \\\"John Doe\\\",\\n\\\"customerBirthday\\\":
\\\"2001-11-01\\\",\\n\\\"summary\\\": \\\"Insurance company failed to notify
customer of automatic cancellation of full reimbursement option and only
provided half reimbursement for
accident.\\\"\\n}\"},\"done_reason\":\"stop\",\"done\":true,\" [...]
- "headers" : {
- "Date" : "Wed, 28 Aug 2024 16:53:37 GMT",
- "Content-Type" : "application/json; charset=utf-8"
- }
- },
- "uuid" : "5c1f926c-0480-41e9-9ca7-a93e17919e99",
- "persistent" : true,
- "insertionIndex" : 5
-}
\ No newline at end of file
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-75e11d9e-fdc3-4873-847d-982daec53233.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-75e11d9e-fdc3-4873-847d-982daec53233.json
new file mode 100644
index 0000000..9f95f7b
--- /dev/null
+++
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-75e11d9e-fdc3-4873-847d-982daec53233.json
@@ -0,0 +1,24 @@
+{
+ "id" : "75e11d9e-fdc3-4873-847d-982daec53233",
+ "name" : "api_chat",
+ "request" : {
+ "url" : "/api/chat",
+ "method" : "POST",
+ "bodyPatterns" : [ {
+ "equalToJson" : "{\n \"model\" : \"granite3-dense\",\n \"messages\" :
[ {\n \"role\" : \"assistant\",\n \"content\" :
\"{\\n\\\"customerSatisfied\\\": true,\\n\\\"customerName\\\": \\\"Sarah
London\\\",\\n\\\"customerBirthday\\\": \\\"1986-07-10\\\",\\n\\\"summary\\\":
\\\"The customer, Sarah London, called to declare an accident on her main
vehicle and was informed that all expenses related to the accident would be
reimbursed.\\\"\\n}\"\n }, {\n \"role\" : \"user\",\n [...]
+ "ignoreArrayOrder" : true,
+ "ignoreExtraElements" : true
+ } ]
+ },
+ "response" : {
+ "status" : 200,
+ "body" :
"{\"model\":\"granite3-dense\",\"created_at\":\"2024-10-29T14:22:39.979850929Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
false,\\n\\\"customerName\\\": \\\"John Doe\\\",\\n\\\"customerBirthday\\\":
\\\"2001-11-01\\\",\\n\\\"summary\\\": \\\"The customer, John Doe, called to
express dissatisfaction with the insurance company's reimbursement policy and
was informed that the full reimbursement option had been automatically
cancelled.\\\" [...]
+ "headers" : {
+ "Date" : "Tue, 29 Oct 2024 14:22:39 GMT",
+ "Content-Type" : "application/json; charset=utf-8"
+ }
+ },
+ "uuid" : "75e11d9e-fdc3-4873-847d-982daec53233",
+ "persistent" : true,
+ "insertionIndex" : 5
+}
\ No newline at end of file
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-861b9f1b-91c6-4206-8c8b-7e393c33d16c.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-861b9f1b-91c6-4206-8c8b-7e393c33d16c.json
new file mode 100644
index 0000000..7829e7a
--- /dev/null
+++
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-861b9f1b-91c6-4206-8c8b-7e393c33d16c.json
@@ -0,0 +1,24 @@
+{
+ "id" : "861b9f1b-91c6-4206-8c8b-7e393c33d16c",
+ "name" : "api_chat",
+ "request" : {
+ "url" : "/api/chat",
+ "method" : "POST",
+ "bodyPatterns" : [ {
+ "equalToJson" : "{\n \"model\" : \"granite3-dense\",\n \"messages\" :
[ {\n \"role\" : \"assistant\",\n \"content\" :
\"{\\n\\\"customerSatisfied\\\": false,\\n\\\"customerName\\\": \\\"John
Doe\\\",\\n\\\"customerBirthday\\\": \\\"2001-11-01\\\",\\n\\\"summary\\\":
\\\"The customer, John Doe, called to express dissatisfaction with the
insurance company's reimbursement policy and was informed that the full
reimbursement option had been automatically cancelled.\\\"\\n}\"\n [...]
+ "ignoreArrayOrder" : true,
+ "ignoreExtraElements" : true
+ } ]
+ },
+ "response" : {
+ "status" : 200,
+ "body" :
"{\"model\":\"granite3-dense\",\"created_at\":\"2024-10-29T14:22:53.209341109Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
true,\\n\\\"customerName\\\": \\\"Kate Boss\\\",\\n\\\"customerBirthday\\\":
\\\"1999-08-13\\\",\\n\\\"summary\\\": \\\"The customer, Kate Boss, called to
inquire about a proof of insurance after an accident and the operator helped
her find her contract and arranged for the proof to be sent to the police
station.\\\ [...]
+ "headers" : {
+ "Date" : "Tue, 29 Oct 2024 14:22:53 GMT",
+ "Content-Type" : "application/json; charset=utf-8"
+ }
+ },
+ "uuid" : "861b9f1b-91c6-4206-8c8b-7e393c33d16c",
+ "persistent" : true,
+ "insertionIndex" : 4
+}
\ No newline at end of file
diff --git
a/data-extract-langchain4j/src/test/resources/mappings/api_chat-b837ed7e-cb2a-43f4-84a9-8522b0c4d34d.json
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-b837ed7e-cb2a-43f4-84a9-8522b0c4d34d.json
new file mode 100644
index 0000000..ddfc598
--- /dev/null
+++
b/data-extract-langchain4j/src/test/resources/mappings/api_chat-b837ed7e-cb2a-43f4-84a9-8522b0c4d34d.json
@@ -0,0 +1,24 @@
+{
+ "id" : "b837ed7e-cb2a-43f4-84a9-8522b0c4d34d",
+ "name" : "api_chat",
+ "request" : {
+ "url" : "/api/chat",
+ "method" : "POST",
+ "bodyPatterns" : [ {
+ "equalToJson" : "{\n \"model\" : \"granite3-dense\",\n \"messages\" :
[ {\n \"role\" : \"user\",\n \"content\" : \"Extract information about a
customer from the text delimited by triple backticks: ```Operator: Hello, how
may I help you ?\\nCustomer: Hello, I'm calling because I need to declare an
accident on my main vehicle.\\nOperator: Ok, can you please give me your name
?\\nCustomer: My name is Sarah London.\\nOperator: Could you please give me
your birth date ?\\nCustom [...]
+ "ignoreArrayOrder" : true,
+ "ignoreExtraElements" : true
+ } ]
+ },
+ "response" : {
+ "status" : 200,
+ "body" :
"{\"model\":\"granite3-dense\",\"created_at\":\"2024-10-29T14:22:27.462514101Z\",\"message\":{\"role\":\"assistant\",\"content\":\"{\\n\\\"customerSatisfied\\\":
true,\\n\\\"customerName\\\": \\\"Sarah
London\\\",\\n\\\"customerBirthday\\\": \\\"1986-07-10\\\",\\n\\\"summary\\\":
\\\"The customer, Sarah London, called to declare an accident on her main
vehicle and was informed that all expenses related to the accident would be
reimbursed.\\\"\\n}\"},\"done_reason\":\"stop\", [...]
+ "headers" : {
+ "Date" : "Tue, 29 Oct 2024 14:22:27 GMT",
+ "Content-Type" : "application/json; charset=utf-8"
+ }
+ },
+ "uuid" : "b837ed7e-cb2a-43f4-84a9-8522b0c4d34d",
+ "persistent" : true,
+ "insertionIndex" : 6
+}
\ No newline at end of file
