Merge branch 'main' into release-preview-health-ai-models

Author: v-alje

articles/ai-services/document-intelligence/versioning/v3-1-migration-guide.md

@@ -131,7 +131,7 @@ Starting from v3.0, [Document Intelligence REST API](../quickstarts/get-started-
 
 ## Changes to the REST API endpoints
 
- The v3.1 REST API combines the analysis operations for layout analysis, prebuilt models, and custom models into a single pair of operations by assigning **`documentModels`**  and  **`modelId`** to the layout analysis (../prebuilt-layout) and prebuilt models.
+ The v3.1 REST API combines the analysis operations for layout analysis, prebuilt models, and custom models into a single pair of operations by assigning **`documentModels`**  and  **`modelId`** to the [layout analysis](../prebuilt/layout.md) and prebuilt models.
 
 ### POST request
 

articles/ai-services/openai/how-to/batch.md

@@ -157,71 +157,6 @@ Yes. Similar to other deployment types, you can create content filters and assoc
 
 Yes, from the quota page in the Studio UI. Default quota allocation can be found in the [quota and limits article](../quotas-limits.md#global-batch-quota).
 
-### How do I tell how many tokens my batch request contains, and how many tokens are available as quota?
-
-The `2024-10-01-preview` REST API adds two new response headers:
-
-* `deployment-enqueued-tokens` - A approximate token count for your jsonl file calculated immediately after the batch request is submitted. This value represents an estimate based on the number of characters and is not the true token count.
-* `deployment-maximum-enqueued-tokens` The total available enqueued tokens available for this global batch model deployment.
-
-These response headers are only available when making a POST request to begin batch processing of a file with the REST API. The language specific client libraries do not currently return these new response headers. To return all response headers you can add `-i` to the standard REST request.
-
-```http
-curl -i -X POST https://YOUR_RESOURCE_NAME.openai.azure.com/openai/batches?api-version=2024-10-01-preview \
-  -H "api-key: $AZURE_OPENAI_API_KEY" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "input_file_id": "file-abc123",
-    "endpoint": "/chat/completions",
-    "completion_window": "24h"
-  }'
-```
-
-```output
-HTTP/1.1 200 OK
-Content-Length: 619
-Content-Type: application/json; charset=utf-8
-Vary: Accept-Encoding
-Request-Context: appId=
-x-ms-response-type: standard
-deployment-enqueued-tokens: 139
-deployment-maximum-enqueued-tokens: 740000
-Strict-Transport-Security: max-age=31536000; includeSubDomains; preload
-X-Content-Type-Options: nosniff
-x-aml-cluster: vienna-swedencentral-01
-x-request-time: 2.125
-apim-request-id: c8bf4351-c6f5-4bfe-9a79-ef3720eca8af
-x-ms-region: Sweden Central
-Date: Thu, 17 Oct 2024 01:45:45 GMT
-
-{
-  "cancelled_at": null,
-  "cancelling_at": null,
-  "completed_at": null,
-  "completion_window": "24h",
-  "created_at": 1729129545,
-  "error_file_id": null,
-  "expired_at": null,
-  "expires_at": 1729215945,
-  "failed_at": null,
-  "finalizing_at": null,
-  "id": "batch_c8dd49a7-c808-4575-9957-b188cd0dd642",
-  "in_progress_at": null,
-  "input_file_id": "file-f89384af0082485da43cb26b49dc25ce",
-  "errors": null,
-  "metadata": null,
-  "object": "batch",
-  "output_file_id": null,
-  "request_counts": {
-    "total": 0,
-    "completed": 0,
-    "failed": 0
-  },
-  "status": "validating",
-  "endpoint": "/chat/completions"
-}
-```
-
 ### What happens if the API doesn't complete my request within the 24 hour time frame?
 
 We aim to process these requests within 24 hours; we don't expire the jobs that take longer. You can cancel the job anytime. When you cancel the job, any remaining work is cancelled and any already completed work is returned. You'll be charged for any completed work.

articles/ai-services/openai/how-to/code-interpreter.md

@@ -6,16 +6,16 @@ services: cognitive-services
 manager: nitinme
 ms.service: azure-ai-openai
 ms.topic: how-to
-ms.date: 05/20/2024
-author: mrbullwinkle
-ms.author: mbullwin
+ms.date: 10/15/2024
+author: aahill
+ms.author: aahi
 recommendations: false
 
 ---
 
 # Azure OpenAI Assistants Code Interpreter (Preview)
 
-Code Interpreter allows the Assistants API to write and run Python code in a sandboxed execution environment.  With Code Interpreter enabled, your Assistant can run code iteratively to solve more challenging code, math, and data analysis problems. When your Assistant writes code that fails to run, it can iterate on this code by modifying and running different code until the code execution succeeds.
+Code Interpreter allows the Assistants API to write and run Python code in a sandboxed execution environment. With Code Interpreter enabled, your Assistant can run code iteratively to solve more challenging code, math, and data analysis problems. When your Assistant writes code that fails to run, it can iterate on this code by modifying and running different code until the code execution succeeds.
 
 > [!IMPORTANT]
 > Code Interpreter has [additional charges](https://azure.microsoft.com/pricing/details/cognitive-services/openai-service/) beyond the token based fees for Azure OpenAI usage. If your Assistant calls Code Interpreter simultaneously in two different threads, two code interpreter sessions are created. Each session is active by default for one hour.
@@ -28,7 +28,7 @@ Code Interpreter allows the Assistants API to write and run Python code in a san
 
 The [models page](../concepts/models.md#assistants-preview) contains the most up-to-date information on regions/models where Assistants and code interpreter are supported.
 
-We recommend using assistants with the latest models to take advantage of the new features, as well as the larger context windows, and more up-to-date training data.
+We recommend using assistants with the latest models to take advantage of the new features, larger context windows, and more up-to-date training data.
 
 ### API Versions
 
@@ -69,7 +69,7 @@ We recommend using assistants with the latest models to take advantage of the ne
 
 ### File upload API reference
 
-Assistants use the [same API for file upload as fine-tuning](/rest/api/azureopenai/files/upload?view=rest-azureopenai-2024-02-15-preview&tabs=HTTP&preserve-view=true). When uploading a file you have to specify an appropriate value for the [purpose parameter](/rest/api/azureopenai/files/upload?view=rest-azureopenai-2024-02-15-preview&tabs=HTTP&preserve-view=true#purpose).
+Assistants use the [same API for file upload as fine-tuning](/rest/api/azureopenai/files/upload?view=rest-azureopenai-2024-02-15-preview&tabs=HTTP&preserve-view=true). When uploading a file, you have to specify an appropriate value for the [purpose parameter](/rest/api/azureopenai/files/upload?view=rest-azureopenai-2024-02-15-preview&tabs=HTTP&preserve-view=true#purpose).
 
 ## Enable Code Interpreter
 
@@ -136,7 +136,7 @@ assistant = client.beta.assistants.create(
   instructions="You are an AI assistant that can write code to help answer math questions.",
   model="gpt-4-1106-preview",
   tools=[{"type": "code_interpreter"}],
-  file_ids=[file.id]
+  tool_resources={"code interpreter":{"file_ids":[file.id]}}
 )
 ```
 
@@ -161,7 +161,11 @@ curl https://YOUR_RESOURCE_NAME.openai.azure.com/openai/assistants?api-version=2
       { "type": "code_interpreter" }
     ],
     "model": "gpt-4-1106-preview",
-    "file_ids": ["assistant-123abc456"]
+    "tool_resources"{
+      "code interpreter": {
+          "file_ids": ["assistant-1234"]
+      }
+    }
   }'
 ```
 

articles/ai-services/speech-service/high-definition-voices.md

@@ -47,7 +47,7 @@ Here's a comparison of features between Azure AI Speech HD voices, Azure OpenAI
 | **Deployment options**  | Cloud only | Cloud only | Cloud, embedded, hybrid, and containers. |
 | **Real-time or batch synthesis**  | Real-time only  | Real-time and batch synthesis  | Real-time and batch synthesis |
 | **Latency**  | Less than 300 ms | Greater than 500 ms | Less than 300 ms  |
-| **Sample rate of synthesized audio** | 8, 16, 22.05, 24, 44.1, and 48 kHz  | 8, 16, 24, and 48 kHz | 8, 16, 22.05, 24, 44.1, and 48 kHz |
+| **Sample rate of synthesized audio** | 8, 16, 24, and 48 kHz  | 8, 16, 24, and 48 kHz | 8, 16, 24, and 48 kHz |
 | **Speech output audio format** | opus, mp3, pcm, truesilk |  opus, mp3, pcm, truesilk  |  opus, mp3, pcm, truesilk  |
 
 ## Supported Azure AI Speech HD voices

articles/ai-services/translator/language-support.md

@@ -114,7 +114,7 @@ ms.author: lajanuar
 |Mongolian (Traditional)|`mn-Mong`|✔|✔| | | |
 |Myanmar|`my`|✔|✔| |✔| |
 |Nepali|`ne`|✔|✔| |✔| |
-|Norwegian|`nb`|✔|✔|✔|✔|✔|
+|Norwegian Bokmål|`nb`|✔|✔|✔|✔|✔|
 |Nyanja|`nya`|✔|✔| | | |
 |Odia|`or`|✔|✔|✔|✔| |
 |Pashto|`ps`|✔|✔| |✔| |
@@ -241,7 +241,7 @@ ms.author: lajanuar
 |Mongolian (Traditional)|`mn-Mong`|Yes|No|
 |Myanmar (Burmese)|`my`|Yes|No|
 |Nepali|`ne`|Yes|Yes|
-|Norwegian|`nb`|Yes|Yes|
+|Norwegian Bokmål|`nb`|Yes|Yes|
 |Odia|`or`|Yes|No|
 |Pashto|`ps`|Yes|No|
 |Persian|`fa`|Yes|No|

articles/machine-learning/how-to-deploy-custom-container.md

@@ -226,15 +226,30 @@ blue_deployment = ManagedOnlineDeployment(
 
 ---
 
-There are a few important concepts to notice in this YAML/Python parameter:
+There are a few important concepts to note in this YAML/Python parameter:
 
-#### Readiness route vs. liveness route
+#### Base image
 
-An HTTP server defines paths for both _liveness_ and _readiness_. A liveness route is used to check whether the server is running. A readiness route is used to check whether the server is ready to do work. In machine learning inference, a server could respond 200 OK to a liveness request before loading a model. The server could respond 200 OK to a readiness request only after the model is loaded into memory.
+The base image is specified as a parameter in environment, and `docker.io/tensorflow/serving:latest` is used in this example. As you inspect the container, you can find that this server uses `ENTRYPOINT` to start an entry point script, which takes the environment variables such as `MODEL_BASE_PATH` and `MODEL_NAME`, and exposes ports such as `8501`. These details are all specific information for this chosen server. You can use this understanding of the server, to determine how to define the deployment. For example, if you set environment variables for `MODEL_BASE_PATH` and `MODEL_NAME` in the deployment definition, the server (in this case, TF Serving) will take the values to initiate the server. Likewise, if you set the port for the routes to be `8501` in the deployment definition, the user request to such routes will be correctly routed to the TF Serving server.
 
-For more information about liveness and readiness probes, see the [Kubernetes documentation](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/).
+Note that this specific example is based on the TF Serving case, but you can use any containers that will stay up and respond to requests coming to liveness, readiness, and scoring routes. You can refer to other examples and see how the dockerfile is formed (for example, using `CMD` instead of `ENTRYPOINT`) to create the containers.
 
-Notice that this deployment uses the same path for both liveness and readiness, since TF Serving only defines a liveness route.
+#### Inference config
+
+Inference config is a parameter in environment, and it specifies the port and path for 3 types of the route: liveness, readiness, and scoring route. Inference config is required if you want to run your own container with managed online endpoint.
+
+#### Readiness route vs liveness route
+
+The API server you choose may provide a way to check the status of the server. There are two types of the route that you can specify: _liveness_ and _readiness_. A liveness route is used to check whether the server is running. A readiness route is used to check whether the server is ready to do work. In the context of machine learning inferencing, a server could respond 200 OK to a liveness request before loading a model, and the server could respond 200 OK to a readiness request only after the model is loaded into the memory.
+
+For more information about liveness and readiness probes in general, see the [Kubernetes documentation](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/).
+
+The liveness and readiness routes will be determined by the API server of your choice, as you would have identified when testing the container locally in earlier step. Note that the example deployment in this article uses the same path for both liveness and readiness, since TF Serving only defines a liveness route. Refer to other examples for different patterns to define the routes.
+
+#### Scoring route
+
+The API server you choose would provide a way to receive the payload to work on. In the context of machine learning inferencing, a server would receive the input data via a specific route. Identify this route for your API server as you test the container locally in earlier step, and specify it when you define the deployment to create.
+Note that the successful creation of the deployment will update the scoring_uri parameter of the endpoint as well, which you can verify with `az ml online-endpoint show -n <name> --query scoring_uri`.
 
 #### Locating the mounted model