ClearML/clearml-serving
1
0
Fork 0
mirror of https://github.com/clearml/clearml-serving synced 2025-06-26 18:16:00 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Documentation

Browse Source
This commit is contained in:
allegroai 2022-03-21 17:54:57 +02:00
parent c83403c34b
commit 7e9c44e73d
7 changed files with 30 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
examples/ensemble/readme.md
View File

@ -24,8 +24,11 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_ensemble_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_ensemble_auto`
3. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_ensemble" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
3. If you already have the `clearml-serving` docker-compose running, it might take it a minute or two to sync with the new endpoint.
Or you can run the clearml-serving container independently `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_ensemble" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.

8
examples/keras/readme.md
View File

@ -30,11 +30,9 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_auto`
3. Run the Triton Engine `docker run -v ~/clearml.conf:/root/clearml.conf -p 8001:8001 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving-triton:latest`
4. Configure the Triton Engine IP on the Serving Service (if running on k8s, the gRPC ingest of the triton container)
`clearml-serving --id <service_id> config --triton-grpc-server <local_ip_here>:8001`
5. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8001:8001 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
6. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_keras" -H "accept: application/json" -H "Content-Type: application/json" -d '{"url": "https://camo.githubusercontent.com/8385ca52c9cba1f6e629eb938ab725ec8c9449f12db81f9a34e18208cd328ce9/687474703a2f2f706574722d6d6172656b2e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031372f30372f6465636f6d707265737365642e6a7067"}'`
3. Make sure you have the `clearml-serving` `docker-compose-triton.yml` (or `docker-compose-triton-gpu.yml`) running, it might take it a minute or two to sync with the new endpoint.
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_keras" -H "accept: application/json" -H "Content-Type: application/json" -d '{"url": "https://camo.githubusercontent.com/8385ca52c9cba1f6e629eb938ab725ec8c9449f12db81f9a34e18208cd328ce9/687474703a2f2f706574722d6d6172656b2e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031372f30372f6465636f6d707265737365642e6a7067"}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.

6
examples/lightgbm/readme.md
View File

@ -26,9 +26,11 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_auto`
3. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
3. If you already have the `clearml-serving` docker-compose running, it might take it a minute or two to sync with the new endpoint.
4. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_lgbm" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2, "x2": 3, "x3": 4}'`
Or you can run the clearml-serving container independently `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_lgbm" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2, "x2": 3, "x3": 4}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.

6
examples/pipeline/readme.md
View File

@ -17,9 +17,11 @@ Training a scikit-learn model (see example/sklearn)
3. Create pipeline model endpoint:
`clearml-serving --id <service_id> model add --engine custom --endpoint "test_model_pipeline" --preprocess "examples/pipeline/preprocess.py"`
4. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. If you already have the `clearml-serving` docker-compose running, it might take it a minute or two to sync with the new endpoint.
5. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_pipeline" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
Or you can run the clearml-serving container independently `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
5. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_pipeline" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
> **_Notice:_** You can also change the serving service while it is already running!

8
examples/pytorch/readme.md
View File

@ -34,11 +34,9 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_pytorch_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_pytorch_auto`
3. Run the Triton Engine `docker run -v ~/clearml.conf:/root/clearml.conf -p 8001:8001 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving-triton:latest`
4. Configure the Triton Engine IP on the Serving Service (if running on k8s, the gRPC ingest of the triton container)
`clearml-serving --id <service_id> config --triton-grpc-server <local_ip_here>:8001`
5. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8001:8001 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
6. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_pytorch" -H "accept: application/json" -H "Content-Type: application/json" -d '{"url": "https://camo.githubusercontent.com/8385ca52c9cba1f6e629eb938ab725ec8c9449f12db81f9a34e18208cd328ce9/687474703a2f2f706574722d6d6172656b2e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031372f30372f6465636f6d707265737365642e6a7067"}'`
3. Make sure you have the `clearml-serving` `docker-compose-triton.yml` (or `docker-compose-triton-gpu.yml`) running, it might take it a minute or two to sync with the new endpoint.
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_pytorch" -H "accept: application/json" -H "Content-Type: application/json" -d '{"url": "https://camo.githubusercontent.com/8385ca52c9cba1f6e629eb938ab725ec8c9449f12db81f9a34e18208cd328ce9/687474703a2f2f706574722d6d6172656b2e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031372f30372f6465636f6d707265737365642e6a7067"}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.

8
examples/sklearn/readme.md
View File

@ -24,8 +24,12 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_sklearn_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_sklearn_auto`
3. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_sklearn" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
3. If you already have the `clearml-serving` docker-compose running, it might take it a minute or two to sync with the new endpoint.
Or you can run the clearml-serving container independently `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_sklearn" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.

7
examples/xgboost/readme.md
View File

@ -25,8 +25,11 @@ Or add Canary endpoint
`clearml-serving --id <service_id> model canary --endpoint "test_model_xgb_auto" --weights 0.1 0.9 --input-endpoint-prefix test_model_xgb_auto`
4. Run the clearml-serving container `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
5. Test new endpoint: `curl -X POST "http://127.0.0.1:8080/serve/test_model_xgb" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2, "x2": 3, "x3": 4}'`
3. If you already have the `clearml-serving` docker-compose running, it might take it a minute or two to sync with the new endpoint.
Or you can run the clearml-serving container independently `docker run -v ~/clearml.conf:/root/clearml.conf -p 8080:8080 -e CLEARML_SERVING_TASK_ID=<service_id> clearml-serving:latest`
4. Test new endpoint (do notice the first call will trigger the model pulling, so it might take longer, from here on, it's all in memory): `curl -X POST "http://127.0.0.1:8080/serve/test_model_xgb" -H "accept: application/json" -H "Content-Type: application/json" -d '{"x0": 1, "x1": 2, "x2": 3, "x3": 4}'`
> **_Notice:_** You can also change the serving service while it is already running!
This includes adding/removing endpoints, adding canary model routing etc.