Skip to content

Commit

Permalink
Add OpenAI and Astra DB configuration examples
Browse files Browse the repository at this point in the history
  • Loading branch information
@eolivelli
eolivelli committed Jul 13, 2023
1 parent 92af675 commit fbd4b1e
Showing 1 changed file with 56 additions and 18 deletions.
74 changes: 56 additions & 18 deletions pulsar-ai-tools/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,12 +17,12 @@ Computes an embeddings vector for a given text. The embeddings vector is added t

Parameters:

| Name | Description |
|------------------|---------------------------------------------------------------------------------------------------------------------|
| model | The model to use, it depends on the provider |
| compute-service | The embeddings provider (openai, huggingface...) |
| text | Template for the text to compute the embeddings on. You can use the Mustache syntax (for example {{value.field1}} ) |
| embeddings-field | The name of the field to add or update (for example value.embeddingsvector) |
| Name | Description |
|------------------|----------------------------------------------------------------------------------------------------------------------------------------------------|
| model | The model to use, it depends on the provider |
| compute-service | The embeddings provider (openai, huggingface...) |
| text | Template for the text to compute the embeddings on. You can use the [Mustache](https://mustache.github.io/) syntax (for example {{value.field1}} ) |
| embeddings-field | The name of the field to add or update (for example value.embeddingsvector) |


### Query
Expand All @@ -48,16 +48,54 @@ Calls a Completion AI model to generate a text completion.

Parameters:

| Name | Description |
|-------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| output-field | The name of the field to add or update (for example value.embeddingsvector) |
| messages | Template for the messages, it is an array. You can use Mustache in order to fill in the template with the fields of the message. |
| model | ID of the model to use. See the [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility) table for details on which models work with the Chat API. |
| temperature | What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. |
| top-p | An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. |
| stop | Up to 4 sequences where the API will stop generating further tokens. |
| max-tokens | The maximum number of [tokens](https://platform.openai.com/tokenizer) to generate in the chat completion. |
| presence-penalty | Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. |
| frequency-penalty | Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. |
| user | A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids). |
| Name | Description |
|-------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| output-field | The name of the field to add or update (for example value.embeddingsvector) |
| messages | Template for the messages, it is an array. You can use [Mustache](https://mustache.github.io/) in order to fill in the template with the fields of the message. |
| model | ID of the model to use. See the [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility) and [Azure OpenAI Service models](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/models) tables for details on which models work with the Chat API. |
| temperature | What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. |
| top-p | An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. |
| stop | Up to 4 sequences where the API will stop generating further tokens. |
| max-tokens | The maximum number of [tokens](https://platform.openai.com/tokenizer) to generate in the chat completion. |
| presence-penalty | Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. |
| frequency-penalty | Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. |
| user | A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids). |


## Configuring the OpenAI provider

In order to configure the OpenAI provider, you need to add an "openai" section in the configuration file, next to "steps".
Currently we support onle the 'openai' and the 'azure' providers.

```json

{
"openai": {
"accessKey": "YOUR_ACCESS_KEY"
"endpoint": "https://api.openai.com/v1"
"provider": "openai"
},
"steps": []
}
```

## Configuring a DataSource for the Query Step
In order to connect to a Datasource, that will be used for the `query` steps you have to put a `datasource` section in the configuration file.

```json

{
"datasource": {
"type": "astra",
"username": "YOUR-CLIENT-ID",
"password": "YOUR-TOKEN",
"secureBundle": "base64:XXXXXXXX"
},
"steps": []
}
```

Currently, we support only the 'astra' type, that is to connect to DataStax Astra DB.
In the secureBundle option you have to put the base64 encoded content of the secure bundle file that you can download from the Astra DB UI.


0 comments on commit fbd4b1e

Please sign in to comment.