Skip to content

Commit 4c7356b

Browse files
opensearch-trigger-bot[bot]zane-neo
opensearch-trigger-bot[bot]
and
zane-neo
authored
Add bedrock multimodal build-in function usage example in doc (#3073) (#3128)
* Add bedrock multimodel build-in function usage example in ddoc Signed-off-by: zane-neo <[email protected]> * Add multimodal starting support version in doc Signed-off-by: zane-neo <[email protected]> * Address comment issues Signed-off-by: zane-neo <[email protected]> --------- Signed-off-by: zane-neo <[email protected]> (cherry picked from commit 9ce1244) Co-authored-by: zane-neo <[email protected]>
1 parent 4f01193 commit 4c7356b

File tree

1 file changed

+38
-2
lines changed

1 file changed

+38
-2
lines changed

‎docs/remote_inference_blueprints/bedrock_connector_titan_multimodal_embedding_blueprint.md‎

Lines changed: 38 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -17,8 +17,11 @@ PUT /_cluster/settings
1717

1818
## 2. Create connector for Amazon Bedrock:
1919

20-
If you are using self-managed Opensearch, you should supply AWS credentials:
20+
If you are using self-managed Opensearch, you should supply AWS credentials.
21+
You have two different approaches to specify the pre&post process function and request body in the API:
2122

23+
**Use build-in function**
24+
> You can use this starting from OpenSearch 2.16
2225
```json
2326
POST /_plugins/_ml/connectors/_create
2427
{
@@ -46,6 +49,21 @@ POST /_plugins/_ml/connectors/_create
4649
"content-type": "application/json",
4750
"x-amz-content-sha256": "required"
4851
},
52+
"request_body": "{\"inputText\": \"${parameters.inputText:-null}\", \"inputImage\": \"${parameters.inputImage:-null}\"}",
53+
"pre_process_function": "connector.pre_process.bedrock.multimodal_embedding",
54+
"post_process_function": "connector.post_process.bedrock.embedding"
55+
}
56+
]
57+
}
58+
```
59+
**Use painless script**
60+
```json
61+
POST /_plugins/_ml/connectors/_create
62+
{
63+
... //same above
64+
"actions": [
65+
{
66+
... // same above
4967
"request_body": "{ \"inputText\": \"${parameters.inputText:-null}\", \"inputImage\": \"${parameters.inputImage:-null}\" }",
5068
"pre_process_function": "\n StringBuilder parametersBuilder = new StringBuilder(\"{\");\n if (params.text_docs.length > 0 && params.text_docs[0] != null) {\n parametersBuilder.append(\"\\\"inputText\\\":\");\n parametersBuilder.append(\"\\\"\");\n parametersBuilder.append(params.text_docs[0]);\n parametersBuilder.append(\"\\\"\");\n \n if (params.text_docs.length > 1 && params.text_docs[1] != null) {\n parametersBuilder.append(\",\");\n }\n }\n \n \n if (params.text_docs.length > 1 && params.text_docs[1] != null) {\n parametersBuilder.append(\"\\\"inputImage\\\":\");\n parametersBuilder.append(\"\\\"\");\n parametersBuilder.append(params.text_docs[1]);\n parametersBuilder.append(\"\\\"\");\n }\n parametersBuilder.append(\"}\");\n \n return \"{\" +\"\\\"parameters\\\":\" + parametersBuilder + \"}\";",
5169
"post_process_function": "\n def name = \"sentence_embedding\";\n def dataType = \"FLOAT32\";\n if (params.embedding == null || params.embedding.length == 0) {\n return null;\n }\n def shape = [params.embedding.length];\n def json = \"{\" +\n \"\\\"name\\\":\\\"\" + name + \"\\\",\" +\n \"\\\"data_type\\\":\\\"\" + dataType + \"\\\",\" +\n \"\\\"shape\\\":\" + shape + \",\" +\n \"\\\"data\\\":\" + params.embedding +\n \"}\";\n return json;\n "
@@ -55,8 +73,11 @@ POST /_plugins/_ml/connectors/_create
5573
```
5674

5775
If using the AWS Opensearch Service, you can provide an IAM role arn that allows access to the bedrock service.
58-
Refer to this [AWS doc](https://docs.aws.amazon.com/opensearch-service/latest/developerguide/ml-amazon-connector.html)
76+
Refer to this [AWS doc](https://docs.aws.amazon.com/opensearch-service/latest/developerguide/ml-amazon-connector.html)
77+
You have two different approaches to specify the pre&post process function and request body in the API:
5978

79+
**Use build-in function**
80+
> You can use this starting from OpenSearch 2.16
6081
```json
6182
POST /_plugins/_ml/connectors/_create
6283
{
@@ -82,6 +103,21 @@ POST /_plugins/_ml/connectors/_create
82103
"content-type": "application/json",
83104
"x-amz-content-sha256": "required"
84105
},
106+
"request_body": "{\"inputText\": \"${parameters.inputText:-null}\", \"inputImage\": \"${parameters.inputImage:-null}\"}",
107+
"pre_process_function": "connector.pre_process.bedrock.multimodal_embedding",
108+
"post_process_function": "connector.post_process.bedrock.embedding"
109+
}
110+
]
111+
}
112+
```
113+
**Use painless script**
114+
```json
115+
POST /_plugins/_ml/connectors/_create
116+
{
117+
... //same above
118+
"actions": [
119+
{
120+
... //same above
85121
"request_body": "{ \"inputText\": \"${parameters.inputText:-null}\", \"inputImage\": \"${parameters.inputImage:-null}\" }",
86122
"pre_process_function": "\n StringBuilder parametersBuilder = new StringBuilder(\"{\");\n if (params.text_docs.length > 0 && params.text_docs[0] != null) {\n parametersBuilder.append(\"\\\"inputText\\\":\");\n parametersBuilder.append(\"\\\"\");\n parametersBuilder.append(params.text_docs[0]);\n parametersBuilder.append(\"\\\"\");\n \n if (params.text_docs.length > 1 && params.text_docs[1] != null) {\n parametersBuilder.append(\",\");\n }\n }\n \n \n if (params.text_docs.length > 1 && params.text_docs[1] != null) {\n parametersBuilder.append(\"\\\"inputImage\\\":\");\n parametersBuilder.append(\"\\\"\");\n parametersBuilder.append(params.text_docs[1]);\n parametersBuilder.append(\"\\\"\");\n }\n parametersBuilder.append(\"}\");\n \n return \"{\" +\"\\\"parameters\\\":\" + parametersBuilder + \"}\";",
87123
"post_process_function": "\n def name = \"sentence_embedding\";\n def dataType = \"FLOAT32\";\n if (params.embedding == null || params.embedding.length == 0) {\n return null;\n }\n def shape = [params.embedding.length];\n def json = \"{\" +\n \"\\\"name\\\":\\\"\" + name + \"\\\",\" +\n \"\\\"data_type\\\":\\\"\" + dataType + \"\\\",\" +\n \"\\\"shape\\\":\" + shape + \",\" +\n \"\\\"data\\\":\" + params.embedding +\n \"}\";\n return json;\n "

0 commit comments

Comments
 (0)