Skip to content

Releases: SeldonIO/seldon-core

v1.11.2

19 Oct 09:22
Compare
Choose a tag to compare

v1.11.2 (2021-10-19)

Full Changelog

Merged Pull Requests:

  • Updating the request logger requirements seldon-deploy-sdk to latest version #3668
  • Fixed KFserving version to v0.6.0 for Alibi Detect Server #3673

v1.11.1

04 Oct 19:20
Compare
Choose a tag to compare

v1.11.1 (2021-10-04)

Full Changelog

Fixed bugs:

  • gRPC broken for non ipv6 systems post 1.9.1 #3616
  • How seldon core analytics is working for model metrics monitoring ??? #3592
  • incomplete meta.requestPath in responses and redundant "tags" field #3477

Closed issues:

  • Python Wrapper does not adhere to Timeout annotations #3613
  • Are you interested in becoming a listed KEDA end-user? #3612
  • Seldon Core v1.11.0 Release #3598
  • what is the best way for receiving/sending data fast? #3449
  • HPA support for autoscaling/v2beta2 API #3143

v1.11.0

22 Sep 08:58
Compare
Choose a tag to compare

v1.11.0

Full Changelog

Fixed bugs:

  • handle bad SemVer for CRD creation #3569
  • leader election RBAC incorrect #3567
  • batch processing with argo failed #3559
  • hello I do not know the composition of jsonΓê⌐Γò¥╞ÆCan you tell me something about it ? #3556
  • docs build failing due to problem with black 20.8b1 wheels #3546
  • Alibi explainer image broken due to numba 0.54 #3540
  • Update Alibi Detect Server to use Alibi Detect v0.7.1 #3481
  • helm 3.5.2 warning on seldon helm charts #2944
  • Support MLFlow models that return pandas DataFrame #2281

Closed issues:

v1.10.0

18 Aug 12:06
279cf3c
Compare
Choose a tag to compare

v1.10.0

Full Changelog

Fixed bugs:

  • sklearn iris model incompatible with latest sklearnserver (1.10.0-dev) #3424
  • It is not possible to add a new inference server. #3415
  • REST Executor Returns wrong Error message in DAG #3411
  • Conda base image is not being pushed to docker hub #3405
  • spec.preserveUnknownFields missing while upgrading seldon-core-operator to 1.9.1 from 1.2.2 using helm chart #3393
  • seldon-puid not included in grpc requests #3389
  • Docs lint is failing due to argo moving repo #3386
  • Seldon Deployment: Dryrun using k8s java Api is not validating all fields #3378
  • batch_processor.py: data will be left unprocessed if the line number in the input file can't be aliquoted by the batch_size #3377
  • Cannot apply Seldon Deployment from Kubernetes Python API #3375
  • click dependencies could not be resolved #3373
  • Can't change REST timeout #3368
  • Deployment giving certificate expired or is not yet valid #3366
  • go mod fails because of invalid character in file name #3354
  • Istio virtualservice created does not whitelist V2 Inference Protocol protobuf names #3352
  • Add functionality to support multiprocessing for Python wrapper GRPC #3334
  • If no-engine=true used only REST/HTTP virtuaservice is created with istio and no GRPC #3329
  • Seldon Explainer Container Crashes Due to GCS Permission Error #3324
  • I deploy tensorflow model using tensorflow2.4.1 and occur error:CUDA error (3): initialization error. #3314
  • Status address URL incorrect for no engine #3312
  • cant build simple-cpp example #3251
  • wrong conda version used in the mlflowserver image #3115
  • Failures on send_feedback_raw path when using proto #2606
  • custom_metrics notebook test is flaky #2570

Closed issues:

  • Create MLflow example using MLServer #3462
  • Run Kubernetes PodSpec validation #3440
  • automatise generation and upload of example models for pre-packaged model servers #3439
  • Request Logger Update #3421
  • Upgrade Alibi Server to 0.6.0 #3401
  • Add MLServer MLFlow Server to Core #3384
  • Release 1.9.2 #3367
  • Create notebook that outlines steps required to extend all existing secrets to be compatible with rclone #3360
  • Research performance improvements for Python Seldon wrapper and research performance between versions of Seldon Core #3359
  • Update Core Builder to use more recent version of Python #3358
  • Create narrative / documentation around security #3345
  • Add GRPC_THREADS for configuring the number of threads in the Python wrapper (and default to 1) #3333
  • Set GUNICORN_THREADS to 1 by default #3332
  • Release 1.9.1 #3319
  • Python GRPC Server does not adhere to Worker/Thread environment variables #3238
  • Update Benchmarking with Argo Worfklows & Vegeta notebook example #3162
  • 1.8.0 Release #3125
  • Missing appVersion inside Chart.yaml #2737
  • Add integration tests to outlier detector and concept drift components #2681
  • Occasional Latency Spike in Python Nodes of Inference Graph #2656
  • Refactor env var retrieval for model_name / image_name in python wrapper so it's centralised in util #2621
  • Python Wrapper should Handle Exceptions correctly #2338
  • Seldon wrapper image with python 3.8 #1230

v1.9.1

02 Jul 12:46
Compare
Choose a tag to compare

v1.9.1 (2021-06-22)

Full Changelog

Fixed bugs:

  • Help: Unable to run any standard demos on GKE cluster #3305

Closed issues:

  • Seldon Core 1.9.0 release #3287
  • add micro-batching to batch processor #2734

v1.9.0

16 Jun 14:24
Compare
Choose a tag to compare

v1.9.0

Full Changelog

Fixed bugs:

  • seldon-core 1.8.0 helm chart CRD error #3254
  • explainer don't repect the spec.replicas #3241
  • Setting TRACING=0 does not disable Jaeger tracing #3158

Closed issues:

  • Allow Tempo Server Env Override #3282
  • req logger - create elements section for tensorflow protocol #3279
  • Integrate Iter8 #3278
  • add some unit tests for request logger #3270
  • Update OpenAPI folder definitions #3261
  • parsing of categorical and proba in req logger for ndarray #3255
  • Custom name for Seldon deployment instead of metadataname-graph component names #3253
  • Allow V2 Protocol for Alibi Explain Server #3247
  • Usage of route_raw in seldon core 1.1.0 #3236
  • option to skip verify ssl in req logger #3230
  • GPT2-Triton Example: extand to contain load test example #3216
  • Allow multi-model repositories for Tensorflow Serving #3206
  • Allow for overriding Istio VirtualService hosts #3137
  • Run black (nbQA) on notebooks as part of fmt/lint #2885
  • Update request logger to run with gunicorn #2141
  • Progressive Rollout #1805

v1.8.0

20 May 17:29
Compare
Choose a tag to compare

Full Changelog

Implemented enhancements:

  • Removal of mapping type from request logger #3013
  • Improve labelling inconsistencies on seldon-managed k8s resources #2757

Fixed bugs:

  • Java Wrapper /predict API regression #3210
  • Seldon feedback api is not supporting String values #3207
  • SA name hardcoded in seldon-leader-election-role rbac #3168
  • Community Call Calendar in Doc Out of Date #3167
  • V2 Inference Compliance #3156
  • Stuck at "liveness probe failed: HTTP probe failed with statuscode: 403" #3129
  • Failing end to end tests (integration and notebooks) in master #3124
  • GCP Release CRD issue #3114
  • Alibi detect image build fails to build in master #3111
  • Seldon Deployment errors when graph does not have a type field #3105
  • Wrong required variable in documentation #3097
  • storageInitializerImage does not work on Kubernetes 1.18 #3087
  • Deployment issue on AWS #3077
  • Fix python wrapper command line args docs #3069
  • HTTP Port Not Change even after PREDICTIVE_UNIT_SERVICE_PORT set #3035
  • OSS-203: Address CVEs for Java JNI Server Image from Twistlock Reports #2968

Closed issues:

  • Add Tempo Prepackaged Server #3192
  • Adjust outlier examples to use rclone based storage initializer #3189
  • TEST ISSUE TO TEST SYNC TO GITHUB #3185
  • Seldon graph complexity #3184
  • Update Alibi-Detect to latest in Alibi-Detect server #3148
  • Allow any structure in custom field in metadata #3144
  • The /aggregate endpoint is wrongly called if we have a COMBINER with SEND_FEEDBACK method activated #3139
  • Metadata for Transformer #3132
  • Update kustomize usage in core #3127
  • GCP 1.7.0 Release #3103
  • Add support for transformers with arbitrary request/response format (ie. not SeldonMessage) #3096
  • Redhat release for 1.7.0 #3091
  • Add new example on Triton Jupyter Notebook Example with GPT-2 #3080
  • Add raw_data parameter to predict / transform / etc functions in Seldon Client #3079
  • Update SPACY notebooks to be aliged with latest Seldon Core #3072
  • Documentation around supported Alibi Algorithms #3053
  • Add health/ping to api v1 #3046
  • no matches for kind "SeldonDeployment" in version "machinelearning.seldon.io/v1alpha2" #3037
  • seldon-container-engine keeps restarting because readinessProbe failed #3036
  • Add GPU Drift Detection Example #3033
  • SeldonClient: Token Authentication without HTTPS #3032
  • 1.7.0 Release #3011
  • implement rclone-based storage.py equivalent #2942
  • Explore consistent python environments for users that create explainers #2934
  • Is it a good idea to support predicting multiple instances upon one request? #2929
  • Expose log level setting in Helm chart #2919
  • Add default /health/status implementation for models #2899
  • Remove "PERSISTENCE" Redis functionality and documentation #2888
  • Seldon component exit on failure without passing to next component in the seldon graph #2730
  • Metrics Endpoint dose not work with Istio Sidecar #2720
  • Prometheus gauge shown as NaN #2685
  • ability to specify init container on per deployment basis #2611
  • Evaluate alterntatives to Storage.py to reduce dependencies and improve support more data sources #1028

v1.7.0

19 Mar 13:52
a331717
Compare
Choose a tag to compare

v1.7.0

Full Changelog

Fixed bugs:

  • Missing protocol check for KFServing for URL in sdep status #3063
  • environment.yml typo in docs #3052
  • Meta parameter did not passed to next model #3050
  • fix integration and notebook tests #3040
  • python microservice refuses to start: setuptools dep conflict #3038
  • HTTP Port Not Change even after PREDICTIVE_UNIT_SERVICE_PORT set #3035
  • Misaligned documentation for SKLearn pre-packaged model server #3029
  • Remove Mutating Webhook if found in latest operator startup #3024
  • Handle default api status in Seldon protocol in executor and python wrapper #3022
  • Update docs to state GUnicorn is stable feature. #3016
  • Cannot create new SeldonDeployment after seldon-core automatic update from 1.5 to 1.6 #3005
  • Explore re-allowing multiple shadow deployments (for Istio only as Ambassador doesn't support) #2991
  • Files created by controller-gen #2987
  • SeldonPodSpec in SeldonDeployment V1alpha and V1 in seldon v1.4 is not parsing metadata successfully #2983
  • Bug in elasticsearch index of metrics server #2971
  • Address CVEs for MAB Epsilon Greedy & Thompson Sampling Server Image from Twistlock Reports #2969
  • Address CVEs for Alibi Detect Server Image from Twistlock Reports #2967
  • Address CVEs for Alibi Explain Server Image from Twistlock Reports #2965
  • Address CVEs for XGBoost Server Image from Twistlock Reports #2964
  • Address CVEs for SKLearn Server Image from Twistlock Reports #2963
  • Address CVEs for MLFlow Server Image from Twistlock Reports #2962
  • Address CVEs for Storage Initializer Image from Twistlock Reports #2961
  • Address CVEs for Request Logger Image from Twistlock Reports #2960
  • seldon-core-microservice: error: unrecognized arguments: REST #2951
  • Seldon Batch Template Bug? #2943
  • Flaky Operator Unit Test: MLServer Panic #2904
  • Seldon-core-microservice Warning/Error message for changed args #2896
  • cannot overwrite initContainers image: reconcile error #2821
  • Manual scale doesn't work if hpaSpec is set #2816
  • Remove Status section of generated CRD by kubebuilder #2132

Closed issues:

  • after run "kubectl get seldondeployments" but got "No resources found." #3010
  • Torchserve support #3002
  • Make Seldon Client REST requests more efficient #3001
  • Support model repositories with Triton Server #2986
  • Dependabot can't evaluate your Python dependency files #2975
  • seldon-batch-processor Install Instruction Missing #2956
  • Release 1.5.2 #2945
  • switch elastic helm chart to opendistro #2912
  • Make custom metrics work with gunicorn reload #2873
  • Create example using alert-manager for thresholds on Alibi Detect servers #2822
  • Allow annotations on Service created by operator #2590

Merged pull requests:

Read more

v1.6.0

03 Feb 14:20
Compare
Choose a tag to compare

Changelog

v1.6.0

Full Changelog

Implemented enhancements:

  • Create a prepackaged model server for PyTorch Models #831

Fixed bugs:

  • IsADirectoryError: [Errno 21] Is a directory: '/mnt/models' #2876
  • error: a container name must be specified for pod #2875
  • MLFlow server-- ModuleNotFoundError: No module named 'prediction' #2874
  • V1 CRD has missing grpcPort and httpPort #2866
  • Broken Link to Documentation Example I'd like to find if it exists #2836
  • Executor does not send feedback to Routers. #2827
  • ArgoCD OutOfSync if SeldonDeployment includes mountpoint #2811
  • Helm failing to fetch https://kubernetes-charts.storage.googleapis.com/ resulting in failing tests #2808
  • send_feedback response is incorrectly managed in seldon_methods.py #2801
  • Upgrading to 1.5.0 causes unexpected error when calling predict endpoint of Python custom model #2786
  • SHAP Breaks Alibi Detect on Python 3.6 due to unpinned Numpy dependency #2767
  • Error when using the R language wrapper #2744
  • Transformers model unable to run with Cuda #2680
  • Allow seldon manager to run as non-root #2631
  • Operator sets HTTPS on the Engine's liveness and ready checks #2586
  • high memory and cpu usage in deployment of xgboost rest #1986

Security fixes:

Closed issues:

  • CVE checks update for redhat image scans #2869
  • Does Seldon Batch Processing Work with Azure Blob Storage? #2858
  • Update engine docs as deprecated #2840
  • Support V2 Protocol in outlier and drift detectors #2831
  • add example of batch processor with rclone #2819
  • Add example of custom init container with rclone #2818
  • remove mutating webhook #2817
  • Handle KFServing V2 Protocol in request logger #2791
  • Create 1.5.1 release with cherrypick #2756
  • Use f-strings in MAB study case examples #2729
  • helm chart imagePullSecrets support to bypass ratelimiting #2694
  • Seldon-core-operator Update for handling namespace #2676
  • docs: No Release Highlights since 1.1.0 #2634
  • Depricate engine (old Java service orchestrator) #2588
  • Add support for Datadog Tracing in the Executor and the Python Wrapper #2436
  • Multi_Archtecture Support #2333
  • Make deployment names configurable #2301
  • java-wrapper-0.2.0 jar is not checked for validity #2180
  • Stateful Model Serving by Saving state to Redis #2138
  • Add documentation on how to extend base prepackaged servers with new images (xgboost, sklearn, etc) #2060
  • Add documentation that dives into the iniContainer #2055
  • Multiplexing or parallel serving of gRPC / REST in Python Wrapper #1968
  • Allow globally configurable docker registry secret for seldon deployments #1923
  • Remove probesonly flag #1856
  • Use custom errors #1841
  • Allow mixed rest/grpc graphs in new golang based executor #1820

Merged pull requests:

Read more

v1.5.1

19 Dec 14:05
Compare
Choose a tag to compare

v1.5.1

Full Changelog

Fixed bugs:

  • SC Operator continues to reconcile objects that are being (foreground) delete #2781
  • Custom metrics not available in Prometheus #2766
  • seldon-batch-processor on seldon-core-s2i-python37 image is not generating any output #2745
  • transport: is not respected on seldondeployment #2540
  • helm install results in wrong configmap #2528

Closed issues:

  • Hard requirement in Tensorflow (API) on GRPCIO 1.32.x breaks Seldon Core #2787
  • istio request timeouts #2727
  • Document how to run python wrapper locally for development #2722
  • Swagger API needs to be upgraded following best practices #2669
  • Authentication support for ELK Logging #2300
  • Support for xgboost4j-spark 0.9 #1395

Merged pull requests: