Skip to content
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

Add connector blueprint for Azure OpenAI Embedding and Chat model #1874

Merged
merged 3 commits into from
Feb 7, 2024
Merged

Add connector blueprint for Azure OpenAI Embedding and Chat model #1874

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
07a54c6
add: connector blueprint for Azure OpenAI embedding
shengbo-ma Jan 16, 2024
420c8f5
add: connector blueprint for azure openai chat model
shengbo-ma Jan 16, 2024
cdffa93
minor: fix typo
shengbo-ma Jan 16, 2024
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
208 changes: 208 additions & 0 deletions docs/remote_inference_blueprints/azure_openai_connector_chat_blueprint.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,208 @@
# Azure OpenAI connector Blueprint for Chat Completion:

## 1. Add Azure OpenAI Endpoint to Trusted URLs:

```json
PUT /_cluster/settings
{
"persistent": {
"plugins.ml_commons.trusted_connector_endpoints_regex": [
"^https://.*\\.openai\\.azure\\.com/.*$"
]
}
}
```

## 2. Create Connector for Azure OpenAI Chat Model:

Refer to [Azure OpenAI Service REST API reference - Chat Completion](https://learn.microsoft.com/en-us/azure/ai-services/openai/reference#chat-completions).

If you are using self-managed Opensearch, you should supply OpenAI API key:


```json
POST /_plugins/_ml/connectors/_create
{
"name": "<YOUR CONNECTOR NAME>",
"description": "<YOUR CONNECTOR DESCRIPTION>",
"version": "<YOUR CONNECTOR VERSION>",
"protocol": "http",
"parameters": {
"endpoint": "<YOUR RESOURCE NAME>.openai.azure.com/",
"deploy-name": "<YOUR DEPLOYMENT NAME>",
"model": "gpt-4",
"api-version": "<YOUR API VERSION>",
"temperature": 0.7
},
"credential": {
"openAI_key": "<YOUR API KEY>"
},
"actions": [
{
"action_type": "predict",
"method": "POST",
"url": "https://${parameters.endpoint}/openai/deployments/${parameters.deploy-name}/chat/completions?api-version=${parameters.api-version}",
"headers": {
"api-key": "${credential.openAI_key}"
},
"request_body": "{ \"messages\": ${parameters.messages}, \"temperature\": ${parameters.temperature} }"
}
]
}
```

### Sample response:
```json
{
"connector_id": "EapnEY0BpYxvPx3Hxpmp"
}
```

## 3. Create model group:

```json
POST /_plugins/_ml/model_groups/_register
{
"name": "remote_model_group",
"description": "This is an example description"
}
```

### Sample response
```json
{
"model_group_id": "YyvcbYsBjU568JRbdHqv",
"status": "CREATED"
}
```

## 4. Register model to model group & deploy model:

```json
POST /_plugins/_ml/models/_register
{
"name": "azure-openAI-gpt-4",
"function_name": "remote",
"model_group_id": "YyvcbYsBjU568JRbdHqv",
"description": "Azure OpenAI GPT 4",
"connector_id": "EapnEY0BpYxvPx3Hxpmp"
}
```

### Sample response
```json
{
"task_id": "E6ppEY0BpYxvPx3HZZkL",
"status": "CREATED",
"model_id": "FKppEY0BpYxvPx3HZZk0"
}
```

Check if the task is completed
```
GET /_plugins/_ml/tasks/E6ppEY0BpYxvPx3HZZkL
```

When model registration is completed, deploy it
```
POST /_plugins/_ml/models/FKppEY0BpYxvPx3HZZk0/_deploy
```

## 5. Test model
```json
POST /_plugins/_ml/models/FKppEY0BpYxvPx3HZZk0/_predict
{
"parameters": {
"messages": [
{
"role": "system",
"content": "You are a helpful assistant."
},
{
"role": "user",
"content": "Hello!"
}
]
}
}
```

### Sample response

```json
{
"inference_results": [
{
"output": [
{
"name": "response",
"dataAsMap": {
"id": "chatcmpl-8hZDlKJIaFLrC2sGTI2EYBiHQUhdI",
"object": "chat.completion",
"created": 1705394185,
"model": "gpt-35-turbo",
"prompt_filter_results": [
{
"prompt_index": 0,
"content_filter_results": {
"hate": {
"filtered": false,
"severity": "safe"
},
"self_harm": {
"filtered": false,
"severity": "safe"
},
"sexual": {
"filtered": false,
"severity": "safe"
},
"violence": {
"filtered": false,
"severity": "safe"
}
}
}
],
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"role": "assistant",
"content": "Hello! How can I assist you today?"
},
"content_filter_results": {
"hate": {
"filtered": false,
"severity": "safe"
},
"self_harm": {
"filtered": false,
"severity": "safe"
},
"sexual": {
"filtered": false,
"severity": "safe"
},
"violence": {
"filtered": false,
"severity": "safe"
}
}
}
],
"usage": {
"prompt_tokens": 19,
"completion_tokens": 9,
"total_tokens": 28
}
}
}
],
"status_code": 200
}
]
}
```

146 changes: 146 additions & 0 deletions docs/remote_inference_blueprints/azure_openai_connector_embedding_blueprint.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,146 @@
# Azure OpenAI connector blueprint example for embedding model

## 1. Add Azure OpenAI endpoint to trusted URLs:

```json
PUT /_cluster/settings
{
"persistent": {
"plugins.ml_commons.trusted_connector_endpoints_regex": [
"^https://.*\\.openai\\.azure\\.com/.*$"
]
}
}
```

## 2. Create connector for Azure OpenAI embedding model:

Refer to [Azure OpenAI Service REST API reference - Embedding](https://learn.microsoft.com/en-us/azure/ai-services/openai/reference#embeddings).

If you are using self-managed Opensearch, you should supply OpenAI API key:


```json
POST /_plugins/_ml/connectors/_create
{
"name": "<YOUR CONNECTOR NAME>",
"description": "<YOUR CONNECTOR DESCRIPTION>",
"version": "<YOUR CONNECTOR VERSION>",
"protocol": "http",
"parameters": {
"endpoint": "<YOUR RESOURCE NAME>.openai.azure.com/",
"deploy-name": "<YOUR DEPLOYMENT NAME>",
"model": "text-embedding-ada-002",
"api-version": "<YOUR API VERSION>"
},
"credential": {
"openAI_key": "<YOUR API KEY>"
},
"actions": [
{
"action_type": "predict",
"method": "POST",
"url": "https://${parameters.endpoint}/openai/deployments/${parameters.deploy-name}/embeddings?api-version=${parameters.api-version}",
"headers": {
"api-key": "${credential.openAI_key}"
},
"request_body": "{ \"input\": ${parameters.input}}",
"pre_process_function": "connector.pre_process.openai.embedding",
"post_process_function": "connector.post_process.openai.embedding"
}
]
}
```

Sample response:
```json
{
"connector_id": "OyB0josB2yd36FqHy3lO"
}
```

## 3. Create model group:

```json
POST /_plugins/_ml/model_groups/_register
{
"name": "remote_model_group",
"description": "This is an example description"
}
```

Sample response:
```json
{
"model_group_id": "TWR0josByE8GuSOJ629m",
"status": "CREATED"
}
```

## 4. Register model to model group & deploy model:

```json
POST /_plugins/_ml/models/_register
{
"name": "OpenAI embedding model",
"function_name": "remote",
"model_group_id": "TWR0josByE8GuSOJ629m",
"description": "test model",
"connector_id": "OyB0josB2yd36FqHy3lO"
}
```


Sample response:
```json
{
"task_id": "PCB1josB2yd36FqHAXk9",
"status": "CREATED"
}
```
Get model id from task
```json
GET /_plugins/_ml/tasks/PCB1josB2yd36FqHAXk9
```
Deploy model, in this demo the model id is `PSB1josB2yd36FqHAnl1`
```json
POST /_plugins/_ml/models/PSB1josB2yd36FqHAnl1/_deploy
```

## 5. Test model inference

```json
POST /_plugins/_ml/models/PSB1josB2yd36FqHAnl1/_predict
{
"parameters": {
"input": [ "What is the meaning of life?" ]
}
}
```

Response:
```json
{
"inference_results": [
{
"output": [
{
"name": "sentence_embedding",
"data_type": "FLOAT32",
"shape": [
1536
],
"data": [
-0.0043460787,
-0.029653417,
-0.008173223,
...
]
}
],
"status_code": 200
}
]
}
```

Loading