Skip to content

Commit

Permalink
chore: Adding Documentation for OpenAI client (#134)
Browse files Browse the repository at this point in the history 
* chore: add documentation for OpenAI client

* chore:minor updates

* chore:Add documentation for embedding client

* chore: switch to the correct readme

* Update packages/foundation-models/README.md

Co-authored-by: Matthias Kuhr <[email protected]>

* chore: cleanup root readme

* chore: minor fix

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* chore: progress till now

* chore:progress till now

* chore: address review comments

* chore: minor cleanup

* chore: address review comments

* chore: address review comments

* chore: minor fixes

* Update README.md

Co-authored-by: Matthias Kuhr <[email protected]>

* chore: review comments

* Update packages/foundation-models/README.md

Co-authored-by: Matthias Kuhr <[email protected]>

* chore: address review comments

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* fix: Changes from lint

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* Update packages/foundation-models/README.md

Co-authored-by: Zhongpin Wang <[email protected]>

* chore: address review comments

---------

Co-authored-by: Matthias Kuhr <[email protected]>
Co-authored-by: Zhongpin Wang <[email protected]>
Co-authored-by: cloud-sdk-js <[email protected]>
  • Loading branch information
@KavithaSiva @MatKuhr @ZhongpinWang
4 people authored Sep 19, 2024
1 parent 18dfd16 commit 1f7c639
Show file tree
Hide file tree
Showing 2 changed files with 142 additions and 16 deletions.
29 changes: 15 additions & 14 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@

# SAP Cloud SDK for AI

Integrate chat completion into your business applications with SAP Cloud SDK for GenAI Hub. Leverage the Generative AI Hub of SAP AI Core to make use of templating, grounding, data masking, content filtering and more. Setup your SAP AI Core instance with SAP Cloud SDK for AI Core.
Integrate chat completion into your business applications with SAP Cloud SDK for AI. Leverage the generative AI hub of SAP AI Core to make use of templating, grounding, data masking, content filtering and more. Setup your SAP AI Core instance with SAP Cloud SDK for AI Core.

## Disclaimer ⚠️

Expand All @@ -27,6 +27,16 @@ This project is currently in an experimental state. All functionality and naming

This project publishes multiple packages and is managed using [pnpm](https://pnpm.io/)

### @sap-ai-sdk/orchestration

This package incorporates generative AI [orchestration](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/orchestration) capabilities into your AI activities in SAP AI Core and SAP AI Launchpad.

#### Installation

```
$ npm install @sap-ai-sdk/orchestration
```

### @sap-ai-sdk/ai-api

This package provides tools to manage your scenarios and workflows in SAP AI Core.
Expand All @@ -36,7 +46,7 @@ This package provides tools to manage your scenarios and workflows in SAP AI Cor
- Deploy inference endpoints for your trained models.
- Register custom Docker registries, sync AI content from your own git repositories, and register your own object storage for training data and model artifacts.

To install the AI Core package in your project, run:
#### Installation

```
$ npm install @sap-ai-sdk/ai-api
Expand All @@ -46,28 +56,19 @@ $ npm install @sap-ai-sdk/ai-api

This package incorporates generative AI foundation models into your AI activities in SAP AI Core and SAP AI Launchpad.

To install the Gen AI Hub package in your project, run:
```
$ npm install @sap-ai-sdk/foundation-models
```

### @sap-ai-sdk/orchestration
#### Installation

This package incorporates generative AI orchestration capabilities into your AI activities in SAP AI Core and SAP AI Launchpad.

To install the Gen AI Hub package in your project, run:
```
$ npm install @sap-ai-sdk/orchestration
$ npm install @sap-ai-sdk/foundation-models
```

## SAP Cloud SDK for AI Sample Project

We have created a sample project demonstrating the different clients' usage of the SAP Cloud SDK for AI for TypeScript/JavaScript. The [project README](./sample-code/README.md) outlines the set-up needed to build and run it locally.


## Support, Feedback, Contribution

This project is open to feature requests/suggestions, bug reports etc. via [GitHub issues](https://github.com/SAP/ai-sdk-js/issues).
This project is open to feature requests/suggestions, bug reports etc. via [GitHub issues](https://github.com/SAP/ai-sdk-js/issues).

Contribution and feedback are encouraged and always welcome. For more information about how to contribute, the project structure, as well as additional contribution information, see our [Contribution Guidelines](CONTRIBUTING.md).

Expand Down
129 changes: 127 additions & 2 deletions packages/foundation-models/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,134 @@ $ npm install @sap-ai-sdk/foundation-models
- Create a `.env` file in the sample-code directory.
- Add an entry `AICORE_SERVICE_KEY='<content-of-service-key>'`.

## Usage
## Azure OpenAI Client

<!-- Details to be added -->
To make a generative AI model available for use, you need to create a deployment.
You can create a deployment for each model and model version, as well as for each resource group that you want to use with generative AI hub.

After the deployment is complete, you have a `deploymentUrl`, which can be used to access the model.

The Azure OpenAI client allows you to send chat completion or embedding requests to OpenAI models deployed in SAP generative AI hub.

### Prerequisites

- A deployed OpenAI model in SAP generative AI hub.
- You can use the [`DeploymentApi`](../ai-api/README.md#deploymentapi) from `@sap-ai-sdk/ai-api` to deploy a model to SAP generative AI hub. For more information, see [here](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/create-deployment-for-generative-ai-model-in-sap-ai-core).
- `sap-ai-sdk/foundation-models` package installed in your project.

### Usage of Azure OpenAI Chat Client

Use the `AzureOpenAiChatClient` to send chat completion requests to an OpenAI model deployed in SAP generative AI hub.
You can pass the model name as a parameter to the client, the SDK will implicitly fetch the deployment ID for the model from the AI Core service and use it to send the request.

By default, the SDK caches the deployment information, which includes the deployment ID and properties such as the model name and model version, for 5 minutes to prevent performance impact from fetching the deployment information for every request.

```TS
import { AzureOpenAiChatClient } from '@sap-ai-sdk/foundation-models';

const client = new AzureOpenAiChatClient('gpt-35-turbo');
const response = await client.run({
messages: [
{
role: 'user',
content: 'Where is the deepest place on earth located'
}
]
});

const responseContent = response.getContent();

```

Use the following snippet to send a chat completion request with system messages:

```TS
import { AzureOpenAiChatClient } from '@sap-ai-sdk/foundation-models';

const client = new AzureOpenAiChatClient('gpt-35-turbo');
const response = await client.run({
messages: [
{
role: 'system',
content: 'You are a friendly chatbot.'
},
{
role: 'user',
content: 'Hi, my name is Isa'
},
{
role: 'assistant',
content: 'Hi Isa! It is nice to meet you. Is there anything I can help you with today?'
},
{
role: 'user',
content: 'Can you remind me, What is my name?'
}
],
max_tokens: 100,
temperature: 0.0
});

const responseContent = response.getContent();
const tokenUsage = response.getTokenUsage();

logger.info(
`Total tokens consumed by the request: ${tokenUsage.total_tokens}\n` +
`Input prompt tokens consumed: ${tokenUsage.prompt_tokens}\n` +
`Output text completion tokens consumed: ${tokenUsage.completion_tokens}\n`
);

```

It is possible to send multiple messages in a single request.
This feature is useful for providing a history of the conversation to the model.

Pass parameters like `max_tokens` and `temperature` to the request to control the completion behavior.
Refer to `AzureOpenAiChatCompletionParameters` interface for knowing more parameters that can be passed to the chat completion request.

#### Obtaining a Client using Resource Groups

[Resource groups](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/resource-groups?q=resource+group) represent a virtual collection of related resources within the scope of one SAP AI Core tenant.

You can use the deployment ID and resource group as an alternative to obtaining a model by using the model name.

```TS
import { AzureOpenAiChatClient } from '@sap-ai-sdk/foundation-models';

const response = await new AzureOpenAiChatClient({ deploymentId: 'd1234' , resourceGroup: 'rg1234' }).run({
messages: [
{
'role':'user',
'content': 'What is the capital of France?'
}
]
});

```

### Usage of Azure OpenAI Embedding Client

Use the `AzureOpenAiEmbeddingClient` to send embedding requests to an OpenAI model deployed in SAP generative AI hub.
You can pass the model name as a parameter to the client, the SDK will implicitly fetch the deployment ID for the model from the AI Core service and use it to send the request.

By default, the SDK caches the deployment information, which includes the deployment ID and properties such as the model name and model version, for 5 minutes to prevent performance impact from fetching the deployment information for every request.

```TS
import { AzureOpenAiEmbeddingClient } from '@sap-ai-sdk/foundation-models';

const client = new AzureOpenAiEmbeddingClient('text-embedding-ada-002');
const response = await client.run({
input: 'AI is fascinating'
});
const embedding = response.getEmbedding();

```

Like in [Azure OpenAI Chat client](#obtaining-a-client-using-resource-groups), you could also pass the [resource group](https://help.sap.com/docs/sap-ai-core/sap-ai-core-service-guide/resource-groups?q=resource+group) name to the client along with the deployment ID instead of the model name.

```TS
const client = new AzureOpenAiEmbeddingClient({ deploymentId: 'd1234' , resourceGroup: 'rg1234' })
```

## Support, Feedback, Contribution

Expand Down

0 comments on commit 1f7c639

Please sign in to comment.