Release arena v0.9.14

VERSION

@@ -1 +1 @@
-0.9.13
+0.9.14

docs/model/analyze/index.md

@@ -1,5 +1,10 @@
 # Model Analyze Guide
 
+<div style="background-color: #e0f2f4; padding: 10px; border-left: 5px solid #e0f2f4;">
+    <strong>Note</strong><br />
+    This feature is still experimental and may change in a future release without warning.
+</div>
+
 Welcome to the Arena Model Analyze Guide! This guide covers how to use the `arena cli` to profile the model to find performance bottleneck, and how to use tensorrt to optimize the inference performance, you can also benchmark the model to get inference metrics like qps, latency, gpu usage and so on. This page outlines the most common situations and questions that bring readers to this section.
 
 ## Who should use this guide?

docs/model/index.md

@@ -204,6 +204,7 @@ When submitting a training job, you can register a model version at the same tim
 ```shell
 $ arena submit pytorchjob \
   --name=bloom-sft \
+  --namespace=default \
   --gpus=1 \
   --image=registry.cn-hangzhou.aliyuncs.com/acs/deepspeed:v0.9.0-chat \
   --data=training-data:/model \
@@ -260,6 +261,7 @@ Description:
       --model-name my-model \
       --model-source pvc://default/training-data/bloom-560m-sft \
       --name bloom-sft \
+      --namespace=default \
       "cd /model/DeepSpeedExamples/applications/DeepSpeed-Chat/training/step1_supervised_finetuning && bash training_scripts/other_language/run_chinese.sh /model/bloom-560m-sft"
 Tags:
   createdBy: arena
@@ -276,6 +278,7 @@ When submitting a serving job, you can associate it with a model by specifying `
 ```shell
 $ arena serve custom \
     --name=bloom-tgi-inference \
+    --namespace=default \
     --gpus=1 \
     --version=v1 \
     --replicas=1 \
@@ -335,6 +338,7 @@ Description:
       --model-name my-model \
       --model-source pvc://default/training-data/bloom-560m-sft \
       --name bloom-sft \
+      --namespace=default \
       "cd /model/DeepSpeedExamples/applications/DeepSpeed-Chat/training/step1_supervised_finetuning && bash training_scripts/other_language/run_chinese.sh /model/bloom-560m-sft"
 Tags:
   createdBy: arena

docs/releases/v0.9.14.md

@@ -0,0 +1,17 @@
+# Release 0.9.14
+
+Arena now supports model management. You can use the `arena model` subcommand to manage registered model and model versions in mlflow, and associate them with your training jobs and serving jobs.
+For more information, please refer to [Model Manage Guide](https://github.com/kubeflow/arena/blob/master/docs/model/index.md).
+
+### New features
+
+- Add support for MLflow model manage. #1058
+- Add model manage documenation. #1066
+
+### Breaking changes
+
+- Migrate model subcommand to model analyze. #1060
+
+### Misc
+
+- Fix readthedocs build failed. #1069

mkdocs.yml

@@ -17,6 +17,7 @@ nav:
   - SDK:
       - Golang: sdk/go/index.md
   - Release Notes:
+      - v0.9.14: releases/v0.9.14.md
       - v0.9.13: releases/v0.9.13.md
       - v0.9.12: releases/v0.9.12.md
       - v0.9.11: releases/v0.9.11.md