mirror of
https://github.com/hanzo-ml/kserve.git
synced 2026-10-30 20:53:49 +00:00
a0f1cc8384
Signed-off-by: Jagadeesh J <[email protected]> * Feat: Changes on Torchserve repo for KFServing to Kserve migration Signed-off-by: Jagadeesh J <[email protected]> * update service envelope env variable in serving runtime Signed-off-by: Suresh Nakkeran <[email protected]> * fix: move envvar to constants Signed-off-by: Jagadeesh J <[email protected]> * fix: nil pointer error Signed-off-by: Suresh Nakkeran <[email protected]> * fix: torchserve image with kserve envelope Signed-off-by: Jagadeesh J <[email protected]> * fix: torchserve runtime image Signed-off-by: Jagadeesh J <[email protected]> * fix: v1 example storage uri - add service_envelope deprecation warining to readme - update torchserve image Signed-off-by: Jagadeesh J <[email protected]> Co-authored-by: Suresh Nakkeran <[email protected]>
13 lines
485 B
YAML
13 lines
485 B
YAML
# For example, specify a “concurrency target” of “10”, the autoscaler will try to make sure that every replica receives on average 10 requests at a time.
|
|
# A target is always evaluated against a specified metric.
|
|
apiVersion: "serving.kserve.io/v1beta1"
|
|
kind: "InferenceService"
|
|
metadata:
|
|
name: "torchserve"
|
|
annotations:
|
|
autoscaling.knative.dev/target: "10"
|
|
spec:
|
|
predictor:
|
|
pytorch:
|
|
storageUri: gs://kfserving-examples/models/torchserve/image_classifier/v1
|