Conversation
WalkthroughThe documentation for the inference service and model repository features was extensively revised. The updates include restructuring sections, clarifying instructions, adding operational guides, introducing new features such as template management, and expanding parameter descriptions. Titles and section names were updated for clarity, and step-by-step guides were enhanced across both documents. Changes
Poem
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (5)
docs/en/model_inference/model_management/functions/model_repository.mdx (2)
12-13: Minor wording tweak improves clarity.“CLI/Notebook uploads leverage internal network bandwidth.” reads a little abrupt.
Consider:- - CLI/Notebook uploads leverage internal network bandwidth. + - CLI- or Notebook-based uploads use the internal network bandwidth for faster transfer.
60-64: Heading capitalization typo.
## Technical Notesis great, but the subsequent list isn’t capitalized consistently (Default Branch Rules).
Make headings Title-Case throughout or sentence-case throughout to keep style uniform.docs/en/model_inference/inference_service/functions/inference_service.mdx (3)
67-70: Add hyphen & tighten phrasing in the tip block.- Custom publishing inference service requires manual setting of parameters. + Custom-publishing an inference service requires manual parameter configuration.
93-96: Capitalisation of “KServe”.The sentence references the CNCF project. Use the official capitalisation to avoid confusion.
- based on [kserve](https://github.com/kserve/kserve) InferenceService CRD + based on [KServe](https://github.com/kserve/kserve) InferenceService CRD
185-186: Clarify scope of Swagger/code examples.The note is helpful but can be mis-read as a limitation of the platform rather than the runtime.
- In addition, the Swagger tab also only supports access to the inference service published by the mlserver runtime. + Likewise, the **Swagger** tab is generated only for services published with the *mlserver* runtime.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
⛔ Files ignored due to path filters (1)
yarn.lockis excluded by!**/yarn.lock,!**/*.lock
📒 Files selected for processing (3)
docs/en/model_inference/inference_service/functions/inference_service.mdx(8 hunks)docs/en/model_inference/model_management/functions/model_repository.mdx(1 hunks)package.json(1 hunks)
🧰 Additional context used
🪛 LanguageTool
docs/en/model_inference/model_management/functions/model_repository.mdx
[uncategorized] ~9-~9: Possible missing preposition found.
Context: ...t and deployment. ## Advantages 1. Git-native Version Control - Track m...
(AI_HYDRA_LEO_MISSING_OF)
[uncategorized] ~16-~16: This verb does not appear to agree with the subject. Consider using a different form.
Context: ...Seamless Integration* - Directly deploy models from the repository to inference...
(AI_EN_LECTOR_REPLACEMENT_VERB_AGREEMENT)
docs/en/model_inference/inference_service/functions/inference_service.mdx
[uncategorized] ~72-~72: Consider adding a hyphen.
Context: ...ck Publish Inference Service in the upper right corner. ### Step 3 If the **"Publish ...
(TOP_LEFT_CORNER)
[uncategorized] ~95-~95: Consider adding a hyphen.
Context: ...can also click the "YAML" button in the upper right corner when "Publish inference service ...
(TOP_LEFT_CORNER)
🔇 Additional comments (1)
package.json (1)
3-3: Check for breaking changes & lock-file update when bumping@alauda/doom.Jumping three minor versions (0.19 → 0.23) can introduce breaking behaviour in dev-tooling.
- Review the CHANGELOG for
@alauda/doom≥ 0.23.0.- Run
yarn dlx @yarnpkg/doctorto ensure the lock-file has been refreshed (yarn install --mode=update-lockfile).- Re-build the docs locally (
yarn build) to surface any incompatibilities early.
Signed-off-by: Yuan Fang <yuanfang@alauda.io>
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (6)
docs/en/model_inference/model_management/functions/model_repository.mdx (2)
6-6: Improve phrasing of the integration sentenceThe phrase “to bridge model development and deployment” reads awkwardly.
A small tweak improves clarity:-… and integrates with MLOps workflows to bridge model development and deployment. +… and integrates with MLOps workflows to bridge the gap between model development and deployment.
80-81: Tighten the instructional wordingGrammatically, proceed to
Createis vague. A direct action verb reads better:-After defining the necessary information, proceed to `Create` the empty Model Repository. +After defining the necessary information, click **Create** to generate the empty model repository.docs/en/model_inference/inference_service/functions/inference_service.mdx (4)
20-22: Replace “deployment threshold” with clearer wording“Lower the deployment threshold” is unusual phrasing in English. Consider a more idiomatic alternative:
-… This simplifies the model deployment process and lowers the deployment threshold. +… This simplifies the model deployment process and lowers the barrier to entry.
75-76: Use “disabled” instead of “not clickable”“Not clickable” sounds informal and may confuse translators. Suggest:
-If the **"Publish Inference Service"** button is not clickable, +If the **Publish Inference Service** button is disabled,
93-94: Streamline the auto-scaling explanationMinor wording clean-up for brevity & clarity:
-At this time, if a request comes, the inference service can automatically start and load the model cached in the PVC. +When a new request arrives, the service automatically restarts and loads the cached model from the PVC.
98-121: Clarify instance count vs. auto-scaling parametersThe table lists both Auto scaling (boolean) and Number of instances (required) without explaining their interaction. Readers may wonder which value is honored when auto-scaling is enabled.
Recommendation:
- Rename the field to Initial number of replicas.
- Add a note explaining it serves as the initial/min-replica count when auto-scaling is on.
-| Number of instances | Required, The number of instances running the inference service.| +| Initial number of replicas | Required. When auto-scaling is enabled, this acts as the minimum replica count.|
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
docs/en/model_inference/inference_service/functions/inference_service.mdx(7 hunks)docs/en/model_inference/model_management/functions/model_repository.mdx(2 hunks)
🧰 Additional context used
🪛 LanguageTool
docs/en/model_inference/inference_service/functions/inference_service.mdx
[uncategorized] ~72-~72: Consider adding a hyphen.
Context: ...ck Publish Inference Service in the upper right corner. ### Step 3: Configure Model Me...
(TOP_LEFT_CORNER)
[uncategorized] ~95-~95: Consider adding a hyphen.
Context: ...can also click the "YAML" button in the upper right corner when "Publish inference service ...
(TOP_LEFT_CORNER)
docs/en/model_inference/model_management/functions/model_repository.mdx
[uncategorized] ~9-~9: Possible missing preposition found.
Context: ...t and deployment. ## Advantages 1. Git-native Version Control - Track m...
(AI_HYDRA_LEO_MISSING_OF)
[uncategorized] ~16-~16: This verb does not appear to agree with the subject. Consider using a different form.
Context: ... Seamless Integration - Directly deploy models from the repository to inference...
(AI_EN_LECTOR_REPLACEMENT_VERB_AGREEMENT)
| ### Step 1: Define Repository Metadata | ||
|
|
||
| | Parameters | Description | | ||
| | :----- | :--------------------------------- | | ||
| | Name | Required, Model Repository Name. | | ||
| | Labels | Custom tags for categorization and search. (e.g., "CV", "NLP", "production") | | ||
| | Description | Detailed explanation of the repository's purpose, model type, or usage guidelines. | | ||
|
|
There was a problem hiding this comment.
🛠️ Refactor suggestion
Add the missing “Visibility” field to the metadata table
The UI exposes a visibility option (Private / Shared / Public), yet the table omits it. Without documentation, users may overlook the setting and lock themselves out of cross-tenant sharing.
| Name | Required, Model Repository Name. |
+| Visibility | Repository exposure level: **Private**, **Shared**, or **Public**. Determines who can see / clone the repo. |
| Labels | Custom tags for categorization and search. (e.g., "CV", "NLP", "production") |📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| ### Step 1: Define Repository Metadata | |
| | Parameters | Description | | |
| | :----- | :--------------------------------- | | |
| | Name | Required, Model Repository Name. | | |
| | Labels | Custom tags for categorization and search. (e.g., "CV", "NLP", "production") | | |
| | Description | Detailed explanation of the repository's purpose, model type, or usage guidelines. | | |
| ### Step 1: Define Repository Metadata | |
| | Parameters | Description | | |
| | :----- | :--------------------------------- | | |
| | Name | Required, Model Repository Name. | | |
| | Visibility | Repository exposure level: **Private**, **Shared**, or **Public**. Determines who can see / clone the repo. | | |
| | Labels | Custom tags for categorization and search. (e.g., "CV", "NLP", "production") | | |
| | Description | Detailed explanation of the repository's purpose, model type, or usage guidelines. | |
🤖 Prompt for AI Agents
In docs/en/model_inference/model_management/functions/model_repository.mdx
around lines 70 to 77, the metadata table is missing the "Visibility" field that
describes the repository's visibility options (Private / Shared / Public). Add a
new row to the table with "Visibility" as the parameter and a description
explaining the visibility settings and their impact on cross-tenant sharing to
ensure users are aware of this important configuration.
1 participant
已发未合的pr jira
Summary by CodeRabbit