mirror of
https://github.com/hanzo-ml/kserve.git
synced 2026-10-30 20:53:49 +00:00
Jagadeesh JandGitHub
358e57225d
* feat: v2 protocol support for torchserve Signed-off-by: Dan Sun <[email protected]> Feat: Add unit test for torchserve predictor - Add e2e test Signed-off-by: Dan Sun <[email protected]> Feat: Add readme doc for v2 support - fix lint error - add grpc sample yaml - add tensor input generation script - fix model archiver to support v2 protocol Signed-off-by: Dan Sun <[email protected]> Feat: Add gRPC client - fix lint error - update readme for gRPC Signed-off-by: Dan Sun <[email protected]> feat: Add custom handlers for v2 api Signed-off-by: Dan Sun <[email protected]> Update torchserve image in test overlay Signed-off-by: Dan Sun <[email protected]> Feat: Add readme doc for v2 support - fix lint error - add grpc sample yaml - add tensor input generation script - fix model archiver to support v2 protocol Signed-off-by: Dan Sun <[email protected]> Update test_transformer.py Signed-off-by: Dan Sun <[email protected]> Add e2e test for grpc torchserve Signed-off-by: Dan Sun <[email protected]> Update test configuration Signed-off-by: Dan Sun <[email protected]> Support torchserve runtime for v2 protocol Signed-off-by: Dan Sun <[email protected]> Fix storage uri for v2 example Signed-off-by: Dan Sun <[email protected]> Add grpc debug and retry config Signed-off-by: Dan Sun <[email protected]> fix: torchserve gRPC test Signed-off-by: Jagadeesh J <[email protected]> * fix: skip gRPC test Signed-off-by: Jagadeesh J <[email protected]>
14 lines
578 B
Properties
14 lines
578 B
Properties
inference_address=http://0.0.0.0:8085
|
|
management_address=http://0.0.0.0:8085
|
|
metrics_address=http://0.0.0.0:8082
|
|
grpc_inference_port=7070
|
|
grpc_management_port=7071
|
|
enable_metrics_api=true
|
|
metrics_format=prometheus
|
|
number_of_netty_threads=4
|
|
job_queue_size=10
|
|
enable_envvars_config=true
|
|
install_py_dep_per_model=false
|
|
model_store=/mnt/models/model-store
|
|
model_snapshot={"name":"startup.cfg","modelCount":1,"models":{"bert":{"1.0":{"defaultVersion":true,"marName":"BERTSeqClassification.mar","minWorkers":1,"maxWorkers":5,"batchSize":2,"maxBatchDelay":500,"responseTimeout":60}}}}
|