You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/AIFoundryDeployment.md
+30-30Lines changed: 30 additions & 30 deletions
Original file line number
Diff line number
Diff line change
@@ -4,68 +4,68 @@ Please follow the steps below to configure the Prompt flow endpoint in App servi
4
4
## Step 1: OpenAI Foundry Project
5
5
1. Launch the [AI Foundry](https://ai.azure.com/) and select `project` under Jump into a project in Azure AI Foundry, If not found click on View all projects from the top menu.
3. Click on `Zip File` radio button. Then click on `Browse` to select the file `DraftFlow.zip` from the cloned/downloaded GitHub repository folder. The file will be located at `<Your Download Folder Path>/ResearchAssistant/Deployment/scripts/ai_hub_scripts/flows/`.
21
+
3. Click on `Zip File` radio button. Then click on `Browse` to select the file `DraftFlow.zip` from the cloned/downloaded GitHub repository folder. The file will be located at `<Your Download Folder Path>/infra/scripts/aihub_scripts/flows/`.
22
22
23
-

23
+

24
24
25
25
4. Once the DraftFlow.zip file is uploaded, change folder name to `DraftFlow` and Select Flow type as `Chat Flow` and click on `Upload` button.
26
26
27
-

27
+

28
28
29
29
30
30
5. Click on `Start compute session` and click on `Start compute session` from the drop-down list. It can take few minutes for the runtime to start.
6. Click on `Deploy` button once it is enabled. Enter a unique name for Endpoint Name field.
35
35
>IMPORTANT: This name has to be unique across all endpoints in your Azure subscription.
36
36
37
37
You can leave the Deployment name, Virtual machine type, and Instace count as populated and click on `Review + Create`. Optionally you can choose a different VM type/size and increase/decrease the Instance count as needed. Then review details and click on `Create` in the next screen.
7. It will take few minutes for the flow to be validated and deployed. Click on `Models + endpoints` from left menu. You might only see the Default_AzureOpenAI deployments in the page until the deployment is completed. Please wait and click on `Refresh` after few minutes.
4. Select the first option `Microsoft Entra Id` from the drop-down list. If `Create new app registration` is disabled, go to [Step 4a](https://github.com/microsoft/Build-your-own-AI-Assistant-Solution-Accelerator/blob/PSL-Authenticate-Doc/ResearchAssistant/Deployment/AIStudioDeployment.md#step-4a-creating-a-new-app-registration)
98
+
4. Select the first option `Microsoft Entra Id` from the drop-down list. If `Create new app registration` is disabled, go to [Step 4a](https://github.com/microsoft/Build-your-own-AI-Assistant-Solution-Accelerator/blob/PSL-Authenticate-Doc/docs/AIStudioDeployment.md#step-4a-creating-a-new-app-registration)
4. Provide the `Name`, select supported account types as `Accounts in this organizational directory only(Contoso only - Single tenant)`, select platform as `Web`, enter/select the `URL` and register.
8. Enter the `web app URL` (Provide the app service name in place of XXXX) and Save. Then go back to [Step 4](https://github.com/microsoft/Build-your-own-AI-Assistant-Solution-Accelerator/blob/PSL-Authenticate-Doc/ResearchAssistant/Deployment/AIStudioDeployment.md#step-4-add-authentication-in-azure-app-service-configuration) and follow from _Point 4_ choose `Pick an existing app registration in this directory` from the Add an Identity Provider page and provide the newly registered App Name.
135
+
8. Enter the `web app URL` (Provide the app service name in place of XXXX) and Save. Then go back to [Step 4](https://github.com/microsoft/Build-your-own-AI-Assistant-Solution-Accelerator/blob/PSL-Authenticate-Doc/docs/AIStudioDeployment.md#step-4-add-authentication-in-azure-app-service-configuration) and follow from _Point 4_ choose `Pick an existing app registration in this directory` from the Add an Identity Provider page and provide the newly registered App Name.
5. Navigate to the local folder where you cloned/downloaded this repository and select notebooks shown in screenshot from path `<Your Download Folder Path>/ResearchAssistant/Deployment/scripts/fabric_scripts/`.
5. Navigate to the local folder where you cloned/downloaded this repository and select notebooks shown in screenshot from path `<Your Download Folder Path>/infra/scripts/fabric_scripts/`.
7. You can Click on each notebook to see the code. These notebooks can be scheduled to run periodically if you plan to upload new documents to the storage account.
Once the flow has been ran successfully, the metrics will be displayed showing a 1-5 score of each respective metric. From here, you can click into the evaluation flow to get a better understanding of the scores.
Copy file name to clipboardExpand all lines: docs/PromptFlowSafetyEvaluation.md
+2-2Lines changed: 2 additions & 2 deletions
Original file line number
Diff line number
Diff line change
@@ -1,7 +1,7 @@
1
1
# Safety Evaluation Flow Guide
2
2
Please follow the steps below to generate Adversarial data and set up the evaluation flow.
3
3
4
-
1. Open the notebook `simulate_and_evaluate_flow.ipynb` in path `<Your Download Folder Path>/ResearchAssistant/Deployment/scripts/evaluation_scripts/` in Visual Studio Code or your choice of notebook editor.
4
+
1. Open the notebook `simulate_and_evaluate_flow.ipynb` in path `<Your Download Folder Path>//infra/scripts/evaluation_scripts/` in Visual Studio Code or your choice of notebook editor.
5
5
6
6
2. Update the parameters like Key Vault Name, AI Studio Project Name and AI Studio Deployment URL, Key and Name parameters and run all the cells in the notebook.
7
7
@@ -11,7 +11,7 @@ Please follow the steps below to generate Adversarial data and set up the evalua
11
11
12
12
5. Click on the flow created from the notebook to see the results.
arguments: '${baseUrl} ${keyVaultName} ${solutionName} ${resourceGroupName} ${subscriptionId} ${solutionLocation}'// Specify any arguments for the script
25
25
timeout: 'PT1H'// Specify the desired timeout duration
26
26
retentionInterval: 'PT1H'// Specify the desired retention interval
0 commit comments