-
Notifications
You must be signed in to change notification settings - Fork 21
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[pull] master from kserve:master #173
Commits on Jan 18, 2024
-
Change the default value for enableDirectPvcVolumeMount to true (#3371)
Signed-off-by: jooho <jlee@redhat.com>
Configuration menu - View commit details
-
Copy full SHA for c6589bb - Browse repository at this point
Copy the full SHA c6589bbView commit details
Commits on Jan 21, 2024
-
feat: Automatically generate Helm Chart docs. Fixes #3356 (#3363)
* generate docs Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> * fix Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> * fix Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> --------- Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for b634922 - Browse repository at this point
Copy the full SHA b634922View commit details
Commits on Jan 22, 2024
-
Modified script for include all kserve poetry projects. (#3350)
1. graph poetry project was not included in previous checks. 2. As poetry lock --check is deprecated, used `poetry check --lock`. Signed-off-by: Andrews Arokiam <andrews.arokiam@ideas2it.com>
Configuration menu - View commit details
-
Copy full SHA for c254e70 - Browse repository at this point
Copy the full SHA c254e70View commit details
Commits on Jan 24, 2024
-
RawDeployment support for Inference Graph (#3199)
* initial commit for graph raw deployment Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * addd hpa support as well for inference graph raw deployment Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Just for local Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Just local Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Fix local setup Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Sleep change Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Fix self-signed-ca installation (#3165) Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * refactored Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Fix logging message Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * adding unit tests Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * adding unit tests Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * removed temporary dev env changes Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Rawdeployment mode type Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * restoring from master branch Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Remove affinity test Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * add with resource test Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * added controller test for graph Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * addressed code review comments Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Added copy right statement for new files Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Removed dependency for InferenceGraph kind on componentExtensionSpec which is specific to inference service spec Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * restored Makefile from master branch Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * fixed codescan issue for AtoI function Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * fixed test failures Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Added unit test for hpa reconciler Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Added Ready status when Deployment is available in raw mode Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Removed unused code Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * e2e test for ISVC in raw mode Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * improved e2e tests for inference graph raw deployment Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * fixed python lint errors Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * added annotations for ig spec that triggers raw deployment Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * made names unique b/w runs Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * Add test Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * Adding unit test Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> * fixed e2e test failure for ig. IG is not moving to ready state because of not handling raw deployment mode condition correctly Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * changed marker for graph test Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * corrected test validation for raw deployment mode knative resources Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * correct rebase errors Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * hpa field moved from annotations to inferencegraphspec as fields Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> * commiting make generate output due to InferenceGraphSpec changes Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> --------- Signed-off-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Signed-off-by: Tanvi Thakur <tanvi_thakur@intuit.com> Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> Co-authored-by: Mopuri, Bharath <bharath_mopuri@intuit.com> Co-authored-by: Tanvi Thakur <tanvi_thakur@intuit.com> Co-authored-by: Sivanantham <90966311+sivanantha321@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for cde094d - Browse repository at this point
Copy the full SHA cde094dView commit details
Commits on Jan 25, 2024
-
Add compatibility for Istio CNI plugin (#3316)
When Istio is installed with its CNI plugin, KServe inference services are not capable to start. This is because the storage initializer is an init-container and the network is not available when the CNI plugin is enabled. The typical recommendation to fix the issue is to remove init-containers and move any initialization code to a regular container. This approach would not work well with KServe, because the serving runtimes assume the model is already present on the filesystem and moving the storage initializer as a regular container will cause race conditions (the runtime will succeed loading only if the storage initializer manages to pull the model before the runtime starts). There are alternative approaches documented in https://istio.io/latest/docs/setup/additional-setup/cni/#compatibility-with-application-init-containers. All alternatives have the downside that the traffic won't be captured by Istio and won't benefit from Istio features, which should be OK for KServe storage-initializer case. These changes use the approach for running the storage initializer using the same UserID as the Istio sidecar. The UID is copied from the sidecar container to cover Istio derivatives, like Maistra. Signed-off-by: Edgar Hernández <23639005+israel-hdez@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 39b8a67 - Browse repository at this point
Copy the full SHA 39b8a67View commit details
Commits on Jan 26, 2024
-
Pass missing infer parameters during conversion (#3368)
* Pass missing infer parameters Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> * Add tests Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> * Fix warnings Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> * Add equal magic method for infer types Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> * Avoid inplace modification of parameters in conversion Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com> --------- Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com>
Configuration menu - View commit details
-
Copy full SHA for 1bfba02 - Browse repository at this point
Copy the full SHA 1bfba02View commit details
Commits on Jan 27, 2024
-
feat: Support S3 transfer acceleration (#3305)
* feat: Support S3 transfer acceleration Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> * Add missing field in ISVC yaml and storage.py Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> * fix test Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> * fix default Signed-off-by: Yuan Tang <terrytangyuan@gmail.com> --------- Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Configuration menu - View commit details
-
Copy full SHA for 7a9f7f7 - Browse repository at this point
Copy the full SHA 7a9f7f7View commit details -
Implement HuggingFace model server (#3334)
* Initial huggingface server implementation Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Implement preprocess and predict for huggingface server Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add postprocess Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Update lock file Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Update func signature Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Update encoding Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add token classification test Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add comments Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Fix comments Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add generate endpoint Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Remove vllm async engine file Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Handle optional vllm Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add async generator Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Fix lint Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Generate response streaming support Signed-off-by: Dan Sun <dsun20@bloomberg.net> * rename async generate stream file Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Call generate endpoint directly Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Remove the tritonclient dependency Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add huggingface dockerfile Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Test for triton bert case Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add huggingface test Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add Makefile Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Support gpu device Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Add gnereate parameters Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Fix generate tests Signed-off-by: Dan Sun <dsun20@bloomberg.net> * Run huggingface test Signed-off-by: Dan Sun <dsun20@bloomberg.net> --------- Signed-off-by: Dan Sun <dsun20@bloomberg.net>
Configuration menu - View commit details
-
Copy full SHA for 4bfa60a - Browse repository at this point
Copy the full SHA 4bfa60aView commit details -
fix: Add missing volume mount to transformer container when using mod…
…elcars (#3384) Modelcar: Add mount of shared volume to transformer container Adding the missing mount of the emptyDir volume that holds the symbolic link to the model data in the model sidecar to a transformer container, if it is configured. Signed-off-by: Roland Huß <rhuss@redhat.com>
Configuration menu - View commit details
-
Copy full SHA for 9189831 - Browse repository at this point
Copy the full SHA 9189831View commit details -
align cloudevents/sdk-go dependency (#3387)
chore: The cloudevents/sdk-go was migrated to v2, however, the v1 is still present in the go.mod. It was being used bu the pkg/logger/worker.go file to set a field that was not being in use anymore, thus it was removed Signed-off-by: Spolti <fspolti@redhat.com>
Configuration menu - View commit details
-
Copy full SHA for 8a0df88 - Browse repository at this point
Copy the full SHA 8a0df88View commit details