ClearML/clearml-docs
1
0
Fork 0
mirror of https://github.com/clearml/clearml-docs synced 2025-02-07 05:20:07 +00:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Add upgrade notes to ClearML Serving 1.1.0 release notes (#1017)

Browse Source
This commit is contained in:
pollfly 2025-01-26 13:00:56 +02:00 committed by GitHub
parent 949c51345c
commit e4eca94eaa
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 28 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
docs/release_notes/clearml_serving/ver_1_1.md
View File

@ -5,7 +5,7 @@ title: Version 1.1
### ClearML Serving 1.1.0 ### ClearML Serving 1.1.0
:::warning Backwards Compatibility :::warning Backwards Compatibility
This release is not backwards compatible - see notes below on upgrading This release is not backwards compatible. See `Upgrading from v1.0` note below.
::: :::
**Breaking Changes** **Breaking Changes**
@ -17,4 +17,30 @@ This release is not backwards compatible - see notes below on upgrading
* Triton engine support for variable request (matrix) sizes * Triton engine support for variable request (matrix) sizes
* Triton support, fix `--aux-config` to support more configurations elements * Triton support, fix `--aux-config` to support more configurations elements
* Hugging Face Transformer support * Hugging Face Transformer support
* `Preprocess` class as module * `Preprocess` class as module (see `Preprocess Class` note below)
:::note Upgrading from v1.0
1. Take down the serving containers (docker-compose or k8s)
1. Update the clearml-serving CLI: `pip3 install -U clearml-serving`
1. Re-add a single existing endpoint: `clearml-serving model add ...` (press `yes` when asked). This will upgrade the
clearml-serving session definitions
1. Pull latest serving containers (`docker-compose pull ...` or k8s)
1. Re-spin serving containers (docker-compose or k8s)
:::
:::note Preprocess Class
You can now add a `Preprocess` class from a module. The entire module folder will be packaged.
```
preprocess_folder
├── __init__.py # from .sub.some_file import Preprocess
└── sub
└── some_file.py
```
Pass the top-level folder path using `--preprocess`. For example:
```
clearml-serving --id <serving_session_id> model add --preprocess /path/to/preprocess_folder ...
```
:::