pulumi-datarobot 0.0.28

A Pulumi package for creating and managing DataRobot resources.

DataRobot Resource Provider

The DataRobot Resource Provider lets you manage DataRobot resources.

The provider is built on terraform-provider-datarobot.

To use this package, please install the Pulumi CLI first.

Prerequisites

There are a few additional steps to build this Provider locally since it has not been published to the Pulumi Registry yet.

Ensure the following tools are installed and present in your $PATH:

• pulumictl
• Go >= 1.17
• make
• Python (called as python3) and pip. For recent versions of MacOS, the system-installed version is fine.
• NodeJS Active or maintenance version (Node.js Releases). We recommend using nvm to manage NodeJS installations.
• Yarn
• TypeScript
• .NET

You may skip installing any languages that you are not interested in testing in right now. Go is the only language required to test in Pulumi YAML

Although the steps below contain Unix commands, the Provider can also be tested on Windows. It is recommended to use Powershell to run the make commands and the Windows equivalents for cp, export, etc.

Run the Low-Code Monitored RAG Example

1. In a terminal clone the pulumi-datarobot repository:

   git clone https://github.com/datarobot-community/pulumi-datarobot.git
   

2. Build the Provider locally. This step is only required because we have not published the Provider to the Pulumi Registry yet.

   make provider
   

3. Copy the generated binary into your PATH.

   cp bin/pulumi-resource-datarobot $GOPATH/bin 
   

   Depending on your setup, this could require sudo permissions. Windows users may need to rename the pulumi-resource-datarobot binary to pulumi-resource-datarobot.exe to make it executable.

4. Go to the examples/low_code_rag directory.

   cd examples/low_code_rag
   

5. The provider requires an API key set in an environment variable named DATROBOT_API_KEY. Copy the API key from the DataRobot console and create the DATAROBOT_API_KEY environment variable.

   export DATAROBOT_API_KEY=<YOUR_API_KEY>
   

   Where <your API key> is the API key you copied from the DataRobot Console.

6. The example requires Google Cloud service account credentials in order to call the Google Vertex AI API. Follow this guide to create a service account key for your Google account. Download the service account key file and save it as google_credentials.json in the current directory.

7. Create the resources.

   By defualt, Pulumi will prompt you to login to their Pulumi Cloud service to manage the state files. To store and manage the state files locally instead, run pulumi login --local before continuing. This will create the .pulumi directory in your home directory to manage the state.

   pulumi up
   

   Enter a name for your Pulumi stack when prompted. Enter yes when asked if you want to perform the update.

8. Once the creation is complete, navigate to the datarobotChatApplicationUrl to view your chat application.

   Outputs:
       datarobotChatApplicationUrl: {
           value: "<your_chat_application_url>"
       }
   
   Resources:
       + 12 created
   

9. (optional) Delete the resources when you are done.

   pulumi down
   

   Choose your stack name when prompted and enter yes to perform this destroy.

Python Example

Follow these steps to run the Python example. Since the Python SDK is not published yet, it must be built and overriden locally.

1. Install local Python SDK.

   make install_python_sdk
   

2. Navigate to the example directory.

   cd examples/py
   

3. Create the resources.

   pulumi up
   

   If the command fails, try adding the location where the local SDK was installed to your PYTHONPATH. For example:

   export PYTHONPATH="${PYTHONPATH}:/opt/homebrew/lib/python3.12/site-packages"
   

Typescript Example

1. Install local Typescript SDK.

   make install_nodejs_sdk
   

2. Navigate to the example directory.

   cd examples/ts
   

3. Link the local Provider

   npm install
   yarn link @pulumi/datarobot
   

4. Create the resources.

   pulumi up