2.2.3

TensorFlow Serving using TensorFlow 2.2.3.

Major Features and Improvements

Breaking Changes

Bug Fixes and Other Changes

• Internal changes. (commit: d044f1e)
• Explicitly export files needed by other packages (commit: 90cb1c0)
• internal BUILD file cleanup. (commit: 12c0f49)
• Add a SourceAdapter that adds a prefix to StoragePath. (commit: f337623)
• Internal change. (commit: b052238)
• Switch users of tensorflow::Env::Now*() to EnvTime::Now*(). (commit: 8a0895e)
• Update the serving model test to use profiler_client.start_tracing instead of (commit: 495e15b)
• moves tf.app to use compat.v1 (commit: fb3c70d)
• Migrates tf.app to use compat.v1 (commit: e8d5ec5)
• Automated rollback of commit e8d5ec5 (commit: 46fbcb2)
• Add test for flex op. (commit: 83ca96f)
• Automated rollback of commit 83ca96f (commit: b853dfe)
• Add release note for TFS 2.1 (commit: 74107ae)
• Remove SessionBundle usage from regressor/classifier. (commit: b1b6ffd)
• Remove SessionBundle support from Predictor. (commit: 2090d67)
• Replace the error_codes.proto references in tf serving. (commit: ab475bf)
• In ServableStateMonitor, introduce a helper method 'GetAvailableServables' in ServableStateMonitor. (commit: 4dcfa25)
• Automated rollback of commit b853dfe (commit: 1750599)
• Remove unused GetCreator() method. (commit: 114c035)
• Adds performance guide and documentation for TensorBoard integration (commit: f1e4eb2)
• Use SavedModelBundleFactory for session source adapter instead (commit: a59e0dc)
• Update TF Text to 2.0.1 (commit: bd206e4)
• Remove SessionBundleSourceAdapter as we load Session bundles via (commit: d50aa2b)
• Remove option to create session bundle based platform config. (commit: 04d0f7e)
• Cleaning up BUILD visibility for tf_pyclif_proto_library intermediate targets, part 2. (commit: 4774fd8)
• Use SavedModelBundleSourceAdapterConfig instead of (commit: 8ed3cee)
• Automated rollback of commit 4774fd8 (commit: 8aa38bd)
• Fix docs to refer to SavedModelBundleSourceAdatperConfig as (commit: 4de4c7b)
• Update minimum bazel version to 1.2.1. (commit: 1a36026)
• Automated rollback of commit 8aa38bd (commit: befd6c6)
• Drop support for beta gRPC APIs. (commit: 13d01fc)
• API spec for httpserver response-streaming (with flow-control). (commit: fd597f0)
• Use @rules_pkg to replace @bazel_tools to build deb package (commit: 1e7f349)
• Change Python version to PY3. (commit: 7516746)
• Update Python tests in PY3. (commit: 0cf65d2)
• Re-format Python test. (commit: ae4de8b)
• Upgrade bazel version for Dockerfiles. (commit: e507aa1)
• Change dockerfile for PY3. (commit: 7cbd06e)
• Update TF Text to v2.1.1 which removes ICU variable conflict with core TF. (commit: 2fd33e1)
• Reduce contention in FastReadDynamicPtr by sharding the ReadPtrs, by default one per CPU. (commit: d3b374b)
• Ensure that all outstanding ReadPtrs are destroyed before allowing a (commit: e41ee40)
• fix typo in flag (commit: 138d9b7)
• Automated rollback of commit 46fbcb2 (commit: 385e796)
• Fixes outdated instructions in kubernetes tutorial (commit: ee66538)
• Refined the httpserver flow-control API for streaming writes. (commit: e085cb3)
• Update bazel genfile directory after bazel version upgrade. (commit: a328ed6)
• Use package python-distutils-extra instead of python3-distutils for ubuntu16.04. (commit: b9602b3)
• Explicitly disable Tensorflow v2 behaviors for all TF1.x binaries and tests (commit: 3ffd969)
• Internal changes. (commit: fdbbe48)
• Makes clear that version labels are not supported by REST API surface (commit: 9781ed1)
• Fix deps on tensorflow proto targets. (commit: 0b83b7e)
• Allow splitting fields from batched session metadata into individual sessions (commit: caf2a92)
• For google-internal, optimize proto parse code for lazy scenario, by annotate byte-type fields with Cord. (commit: 726ca19)
• Fix tensorflow_serving build when building against a recent hash of TensorFlow (commit: 3d5b44a)
• Allow passing ThreadPoolOptions in various Session implementations. (commit: 2b6212c)
• Fix test broken by CL 301607174: some Session subclasses may not have implementation for Run() with thread pool options. (commit: ff10e69)
• Inline deps variables in BUILD files. (commit: dfc7a5c)
• Update bazel version used in the docker images. (commit: 162f729)
• Bump TF version. (commit: 6d6792c)
• Bump TF version. (commit: 08a1ecf)
• Small performance optimization. (commit: 4196a72)
• Update version for 2.2.0-rc0 release. (commit: ecf3328)
• Update version for 2.2.0-rc1 release. (commit: 6e254ff)
• Update version for 2.2.0-rc2 release. (commit: 3ad9a98)
• Format error strings correctly in JSON response (Fixes #1600). (commit: 1ff4d31)
• Merge pull request #1615 from peakji:master (commit: 22635f6)
• Fix broken GetModelMetadata request processing (#1612) (commit: 55c4037)
• Fix icu dependency checksum. (commit: 322c9ee)
• Support Python 3.7 in tensorflow-serving-api package (Fixes #1640) (commit: f775bb2)
• Update ICU library to include knowledge of built-in data files. (commit: 774f248)
• Update TF Text regression model to catch errors thrown from within ops. (commit: 11e49b0)
• Adds storage.googleapis.com as the primary download location for the ICU, and resets the sha256 to match this archive. (commit: 028d050)
• Upgrade to CUDA Version 10.1. (commit: 3ab70a7)
• Update TF Text to v2.2.0. (commit: fd98428)
• Update version for 2.2.0 release. (commit: 066c50a)
• Install python3-distutils in gpu docker config. (commit: 145b337)
• Install cuda-nvrtc 10.1 into release tf serving gpu docker image. (commit: 9b74939)
• Specifying compute capabilities in the TF 2.2 .bazelrc file to reduce startup time on GPUs; this is in line with TF 2.3 changes (commit: 54de106)
• Update version for 2.2.2 release. (commit: d0b8fdf)
• Update .bazelrc (commit: 3a9b697)
• Update version for 2.2.3 release. (commit: e89e0f5)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

broken, Haiming, Minglotus-6, rhaertel80, Shyam Saladi, Tensorflow-Copybara

2.6.0-rc0

TensorFlow Serving using TensorFlow 2.6.0-rc0.

Major Features and Improvements

No major features or improvements.

Breaking Changes

No Breaking changes.

Bug Fixes and Other Changes

• Enable tensor output filters with batching. (commit: 448dbe1)
• Update tf.io import in warmup example doc. (commit: 6579d2d)
• Rollback incompatible C++17 changes. (commit: ba0fa72)
• Roll forward with std::optional -> absl::optional. (commit: 009dac6)
• Replace STL algorithm call with a container method (performance-inefficient-algorithm). (commit: f5bc099)
• Internal change. (commit: 61ca084)
• Remove unused lambda capture. (commit: c80db5a)
• Use absl::StrContains instead of find() != npos. (commit: 9b3d4d7)
• Remove unused "using" decl. (commit: ffcc4a1)
• Resize tensors if the dimensions of the tflite and tensorflow inputs mismatch, even if the number of elements are the same (commit: 8293f44)
• Make Internal Change. (commit: 12bbced)
• Move status_proto to public visible apis/ (it being used by public API protos) (commit: 7f894c7)
• Move core/logging.proto -> apis/logging.proto (commit: 37c64b8)
• Update TF Text to v2.5.0. (commit: 48e5a9e)
• Fix argument name vs comment mismatch (commit: c9bd671)
• cleanup unused vars+functions. (commit: 4ce9c6b)
• Internal change. (commit: f295831)
• Internal change. (commit: ddc3421)
• Remove unused lambda captures in loader_harness_test.cc (commit: bb34b1d)
• Avoid using 'version' after it was moved. (commit: 23b0755)
• Adding python targets for config protos (commit: 757c3a6)
• Add basic batch scheduler for tflite models to improve batch parallelism. (commit: 0ffd608)
• Remove experimental tags from uses of gRPC C++ callback API. (commit: 97aa2db)
• Remove experimental tags from uses of gRPC C++ callback API. (commit: b355023)
• Reserve Unavailable error to TF communication ops only. (commit: db9aca1)
• Add the flag thread_pool_factory_config_file to model server and fix a typo. (commit: efc445f)
• Add new --use_alts_credentials flag, to enable building secure credentials using Google ALTS. (commit: ceba636)
• Enable HTTP PATCH support in evhttp_server (commit: 6cbc4a9)
• Make Internal Change (commit: 6a54e2a)
• Update version for 2.6.0-rc0 release. (commit: ae853e4)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

nniuzft

2.1.4: Merge pull request #1876 from tensorflow/minglotus-6-patch-1

Release 2.1.4

Major Features and Improvements

Breaking Changes

Bug Fixes and Other Changes

• Fix link for TFRecord in Saved Model Warmup documentation. (commit: 127a112)
• Be able to discard aspired-versions request from SourceRouter (commit: 10e4987)
• Fix copying of string tensor outputs by explicitly copying each (commit: 9544077)
• Adds integration with TensorBoard profiler service (commit: 91adea9)
• Migrate from std::string to tensorflow::tstring. (commit: e24571a)

Thanks to our Contributors

This release contains contributions from many people at Google.

2.5.1

(Please note: Due to build issues, we had to skip 2.5.0 version number for this release)

TensorFlow Serving using TensorFlow 2.5.0

Major Features and Improvements

• Upgrade to CUDA 11.2. (commit: 1975e3e)
• Update TF Text to v2.4.3. (commit: ccfb606)
• Experimental support for XLA/CPU models. (commit: 3c1b2b3)
• Add metrics to the Prometheus API (commit: dfb41f1)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)

Breaking Changes

• No Breaking changes

Bug Fixes and Other Changes

• Fix typo in REQUIRED_PACKAGES for grpcio (commit: b9ed0f8)
• update resnet_k8s.yaml file (commit: e7b7b33)
• Fix a compile warning thrown by gcc-9 (commit: 38a017d)
• Update json_tensor.cc (commit: a0a9d14)
• Add TfLiteInterpreterPool to make concurrent use of TfliteSession better (commit: d9efa43)
• Enable download of TF Serving sources at arbitrary commit for CPU docker image. (commit: de1ab9e)
• Updated tests to newer API (commit: 30dd2fe)
• When GRPC messages come to TF Serving GRPC server, server will create a new threads to handle each message. (commit: ac0eb73)
• Add dedicated aliases field to ModelServerConfig. (commit: 358f7d1)
• Update docker command line to work with GPUs (Fixes #1768). (commit: b41a28b)
• option to disable grpc over http (commit: f087290)
• use RecordRuntimeLatency (commit: d353931)
• Improve error message for file not exists. (commit: 78d47f7)
• Transition TensorFlow Serving to TensorFlow's new WORKSPACE protocol. (commit: 50a7ef3)
• fix http_rest_api_handler_test (commit: 53da4f8)
• add status label in request_latency (commit: 0483edb)
• fix timing of request_latency (commit: e20c91e)
• fix http_rest_api_handler_test (commit: 6254411)
• Clarifying object values in REST requests to include B64 encoding and similar key:value pair objects. (commit: 0536678)
• Register custom TfLite ParseExample and add benchmark (commit: 20fe3ca)
• Update docker.md (commit: 7ebcd15)
• Pre-allocate memory for certain vectors where the size is known. (commit: e208b6e)
• Updating serving_basic for adjusting serving_basic.md file and making it up2date with TF2.x - including: (commit: cea306a)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)
• Fixing MKL builds due to missing 'build_with_openmp' option (commit: 0ed23df)
• Implement batch parallelism for tflite sessions (commit: fec1d5d)
• Use LOG_EVERY_N_SEC insted of LOG with local static time variable. (commit: c6b3936)
• Fix TensorFlow Serving build with MKL+OpenMP (commit: ddad074)
• add cors headers (commit: d65914b)
• Remove extra build options for TF Serving (commit: 63acc95)
• Remove hashtable custom op dependencies (commit: bb51722)
• allow http OPTIONS request and add default OPTIONS request handling (commit: 6287cb4)
• Enable aspired version which failed to load to attempt reload. (commit: 2530a33)
• Fixed a compilation error in aspired_versions_manager.cc (commit: 4ca9a4b)
• handle options requests and check if Origin header exists (commit: c512141)
• Add "_r" root event annotation to ProcessBatch events. (commit: e5c3aec)
• Bump minimum bazel version 3.7.2. (commit: 5edcd13)
• Update TensorFlow serving documentation with instructions to bind a host volume. The profiler in the docker will write to this volume, and on the host side TensorBoard will load the profile from the location. (commit: 7c21b22)
• Fix package build due to config move in: (commit: 18dd766)
• Fix memory leak from allocating input tensors (commit: 2f9b6a0)
• Allowing lossy floating point conversions for JSON inputs since JSON has only a single numeric type. This is inline with textproto conversion (and even C++ implicit double -> float conversion). (commit: 57dac6c)
• Adding enable_profiler command line flag. (commit: 7e8720d)
• Add logging in ServerCore. (commit: 623da67)
• Fix zlib. (commit: bc24390)
• Fix broken GPU build by add TF cuda options: (commit: 05377a9)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Abhinav Pundir, Abolfazl Shahbazi, Aurélien Geron, Bairen Yi, gbaned, handong, Hao Ziyu, Junqin Zhang, kiddos, Oliver Sampson, OniB, Runzhen Wang, skawasak, zou000

2.5.0-rc3

TensorFlow Serving using TensorFlow 2.5.0-rc3

Major Features and Improvements

• Upgrade to CUDA 11.2. (commit: 1975e3e)
• Update TF Text to v2.4.3. (commit: ccfb606)
• Experimental support for XLA/CPU models. (commit: 3c1b2b3)
• Add metrics to the Prometheus API (commit: dfb41f1)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)

Breaking Changes

• No Breaking changes

Bug Fixes and Other Changes

• Fix typo in REQUIRED_PACKAGES for grpcio (commit: b9ed0f8)
• update resnet_k8s.yaml file (commit: e7b7b33)
• Fix a compile warning thrown by gcc-9 (commit: 38a017d)
• Update json_tensor.cc (commit: a0a9d14)
• Add TfLiteInterpreterPool to make concurrent use of TfliteSession better (commit: d9efa43)
• Enable download of TF Serving sources at arbitrary commit for CPU docker image. (commit: de1ab9e)
• Updated tests to newer API (commit: 30dd2fe)
• When GRPC messages come to TF Serving GRPC server, server will create a new threads to handle each message. (commit: ac0eb73)
• Add dedicated aliases field to ModelServerConfig. (commit: 358f7d1)
• Update docker command line to work with GPUs (Fixes #1768). (commit: b41a28b)
• option to disable grpc over http (commit: f087290)
• use RecordRuntimeLatency (commit: d353931)
• Improve error message for file not exists. (commit: 78d47f7)
• Transition TensorFlow Serving to TensorFlow's new WORKSPACE protocol. (commit: 50a7ef3)
• fix http_rest_api_handler_test (commit: 53da4f8)
• add status label in request_latency (commit: 0483edb)
• fix timing of request_latency (commit: e20c91e)
• fix http_rest_api_handler_test (commit: 6254411)
• Clarifying object values in REST requests to include B64 encoding and similar key:value pair objects. (commit: 0536678)
• Register custom TfLite ParseExample and add benchmark (commit: 20fe3ca)
• Update docker.md (commit: 7ebcd15)
• Pre-allocate memory for certain vectors where the size is known. (commit: e208b6e)
• Updating serving_basic for adjusting serving_basic.md file and making it up2date with TF2.x - including: (commit: cea306a)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)
• Fixing MKL builds due to missing 'build_with_openmp' option (commit: 0ed23df)
• Implement batch parallelism for tflite sessions (commit: fec1d5d)
• Use LOG_EVERY_N_SEC insted of LOG with local static time variable. (commit: c6b3936)
• Fix TensorFlow Serving build with MKL+OpenMP (commit: ddad074)
• add cors headers (commit: d65914b)
• Remove extra build options for TF Serving (commit: 63acc95)
• Remove hashtable custom op dependencies (commit: bb51722)
• allow http OPTIONS request and add default OPTIONS request handling (commit: 6287cb4)
• Enable aspired version which failed to load to attempt reload. (commit: 2530a33)
• Fixed a compilation error in aspired_versions_manager.cc (commit: 4ca9a4b)
• handle options requests and check if Origin header exists (commit: c512141)
• Add "_r" root event annotation to ProcessBatch events. (commit: e5c3aec)
• Bump minimum bazel version 3.7.2. (commit: 5edcd13)
• Update TensorFlow serving documentation with instructions to bind a host volume. The profiler in the docker will write to this volume, and on the host side TensorBoard will load the profile from the location. (commit: 7c21b22)
• Fix package build due to config move in: (commit: 18dd766)
• Fix memory leak from allocating input tensors (commit: 2f9b6a0)
• Allowing lossy floating point conversions for JSON inputs since JSON has only a single numeric type. This is inline with textproto conversion (and even C++ implicit double -> float conversion). (commit: 57dac6c)
• Adding enable_profiler command line flag. (commit: 7e8720d)
• Add logging in ServerCore. (commit: 623da67)
• Fix zlib. (commit: bc24390)
• Fix broken GPU build by add TF cuda options: (commit: 05377a9)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Abhinav Pundir, Abolfazl Shahbazi, Aurélien Geron, Bairen Yi, gbaned, handong, Hao Ziyu, Junqin Zhang, kiddos, Oliver Sampson, OniB, Runzhen Wang, skawasak, zou000

2.5.0-rc2

TensorFlow Serving using TensorFlow 2.5.0-rc2

Major Features and Improvements

• Upgrade to CUDA 11.2. (commit: 1975e3e)
• Update TF Text to v2.4.3. (commit: ccfb606)
• Experimental support for XLA/CPU models. (commit: 3c1b2b3)
• Add metrics to the Prometheus API (commit: dfb41f1)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)

Breaking Changes

• No Breaking changes

Bug Fixes and Other Changes

• Fix typo in REQUIRED_PACKAGES for grpcio (commit: b9ed0f8)
• update resnet_k8s.yaml file (commit: e7b7b33)
• Fix a compile warning thrown by gcc-9 (commit: 38a017d)
• Update json_tensor.cc (commit: a0a9d14)
• Add TfLiteInterpreterPool to make concurrent use of TfliteSession better (commit: d9efa43)
• Enable download of TF Serving sources at arbitrary commit for CPU docker image. (commit: de1ab9e)
• Updated tests to newer API (commit: 30dd2fe)
• When GRPC messages come to TF Serving GRPC server, server will create a new threads to handle each message. (commit: ac0eb73)
• Add dedicated aliases field to ModelServerConfig. (commit: 358f7d1)
• Update docker command line to work with GPUs (Fixes #1768). (commit: b41a28b)
• option to disable grpc over http (commit: f087290)
• use RecordRuntimeLatency (commit: d353931)
• Improve error message for file not exists. (commit: 78d47f7)
• Transition TensorFlow Serving to TensorFlow's new WORKSPACE protocol. (commit: 50a7ef3)
• fix http_rest_api_handler_test (commit: 53da4f8)
• add status label in request_latency (commit: 0483edb)
• fix timing of request_latency (commit: e20c91e)
• fix http_rest_api_handler_test (commit: 6254411)
• Clarifying object values in REST requests to include B64 encoding and similar key:value pair objects. (commit: 0536678)
• Register custom TfLite ParseExample and add benchmark (commit: 20fe3ca)
• Update docker.md (commit: 7ebcd15)
• Pre-allocate memory for certain vectors where the size is known. (commit: e208b6e)
• Updating serving_basic for adjusting serving_basic.md file and making it up2date with TF2.x - including: (commit: cea306a)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)
• Fixing MKL builds due to missing 'build_with_openmp' option (commit: 0ed23df)
• Implement batch parallelism for tflite sessions (commit: fec1d5d)
• Use LOG_EVERY_N_SEC insted of LOG with local static time variable. (commit: c6b3936)
• Fix TensorFlow Serving build with MKL+OpenMP (commit: ddad074)
• add cors headers (commit: d65914b)
• Remove extra build options for TF Serving (commit: 63acc95)
• Remove hashtable custom op dependencies (commit: bb51722)
• allow http OPTIONS request and add default OPTIONS request handling (commit: 6287cb4)
• Enable aspired version which failed to load to attempt reload. (commit: 2530a33)
• Fixed a compilation error in aspired_versions_manager.cc (commit: 4ca9a4b)
• handle options requests and check if Origin header exists (commit: c512141)
• Add "_r" root event annotation to ProcessBatch events. (commit: e5c3aec)
• Bump minimum bazel version 3.7.2. (commit: 5edcd13)
• Update TensorFlow serving documentation with instructions to bind a host volume. The profiler in the docker will write to this volume, and on the host side TensorBoard will load the profile from the location. (commit: 7c21b22)
• Fix package build due to config move in: (commit: 18dd766)
• Fix memory leak from allocating input tensors (commit: 2f9b6a0)
• Allowing lossy floating point conversions for JSON inputs since JSON has only a single numeric type. This is inline with textproto conversion (and even C++ implicit double -> float conversion). (commit: 57dac6c)
• Adding enable_profiler command line flag. (commit: 7e8720d)
• Add logging in ServerCore. (commit: 623da67)
• Fix zlib. (commit: bc24390)
• Fix broken GPU build by add TF cuda options: (commit: 05377a9)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Abhinav Pundir, Abolfazl Shahbazi, Aurélien Geron, Bairen Yi, gbaned, handong, Hao Ziyu, Junqin Zhang, kiddos, Oliver Sampson, OniB, Runzhen Wang, skawasak, zou000

2.5.0-rc1

TensorFlow Serving using TensorFlow 2.5.0-rc1

Major Features and Improvements

• Upgrade to CUDA 11.2 and and cuDNN 8.1.0 (commit: 1975e3e)
• Update TF Text to v2.4.3. (commit: ccfb606)
• Experimental support for serving JAX and XLA/CPU models. (commit: 3c1b2b3)
• Add latency and availability metrics to the Prometheus API (#1623) (commit: dfb41f1)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)
• Add Cross-Origin Resource Sharing (CORS) headers to REST API (#1817)

Breaking Changes

• No breaking changes

Bug Fixes and Other Changes

• Fix typo in REQUIRED_PACKAGES for grpcio (commit: b9ed0f8)
• Update resnet_k8s.yaml file (commit: e7b7b33)
• Fix a compile warning thrown by gcc-9 (commit: 38a017d)
• Fix typo (commit: dbcd54f)
• Update json_tensor.cc (commit: a0a9d14)
• Add TfLiteInterpreterPool to make concurrent use of TfliteSession better (commit: d9efa43)
• Enable download of TF Serving sources at arbitrary commit for CPU docker image. (commit: de1ab9e)
• Updated tests to newer API (commit: 30dd2fe)
• Control number of grpc threads for request handling to avoid OOM (Fixes #1785). (commit: ac0eb73)
• Add dedicated aliases field to ModelServerConfig. (commit: 358f7d1)
• Update docker command line to work with GPUs (Fixes #1768). (commit: b41a28b)
• Option to disable grpc over http (Fixes #1764) (commit: f087290)
• Remove an unused experimental config option "experimental_fixed_input_tensors_filepath". (commit: 3234fca)
• Removing CurriedSession, since it is no longer used. (commit: 87793ad)
• Improve error message for file not exists. (commit: 78d47f7)
• Fix inference request delay when model is switched (Fixes #1796). (commit: 803dd42)
• Transition TensorFlow Serving to TensorFlow's new WORKSPACE protocol. (commit: 50a7ef3)
• Clarifying object values in REST requests to include B64 encoding and similar key:value pair objects. (commit: 0536678)
• Remove experimental comment on TfLiteSession (commit: ab7f9a5)
• Register custom TfLite ParseExample and add benchmark (commit: 20fe3ca)
• Use respectful terms. (commit: b73bd7b)
• Pre-allocate memory for certain vectors where the size is known. (commit: e208b6e)
• Updating serving_basic for adjusting serving_basic.md file and making it up2date with TF2.x - including: (commit: cea306a)
• Use NullSafeStringView for potentially null pointer returned from libevent (commit: a46fdb2)
• Replace nullptr constructor for string_view with empty strings (commit: a98d164)
• Fixing MKL builds due to missing 'build_with_openmp' option (commit: 0ed23df)
• Implement batch parallelism for tflite sessions (commit: fec1d5d)
• Fix GPU docker image massive increase in size (#1813) (commit: 5a0dfd9)
• Fix TensorFlow Serving build with MKL+OpenMP (commit: ddad074)
• Remove hashtable custom op dependencies (commit: bb51722)
• Enable aspired version which failed to load to attempt reload. (commit: 2530a33)
• Fixed a compilation error in aspired_versions_manager.cc (commit: 4ca9a4b)
• Add "_r" root event annotation to ProcessBatch events. (commit: e5c3aec)
• Bump minimum bazel version 3.7.2. (commit: 5edcd13)
• Dont hardcode path to python3. (commit: 63b2d1c)
• Fix package build due to config move in: (commit: 18dd766)
• Add model_service_cc_grpc_proto (commit: a670ff5)
• Fix memory leak from allocating input tensors (commit: 2f9b6a0)
• Allowing lossy floating point conversions for JSON inputs (commit: 57dac6c)
• Adding enable_profiler command line flag. (commit: 7e8720d)
• Add logging in ServerCore. (commit: 623da67)
• Removes mention of ASCII (commit: 8e97b59)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Abhinav Pundir, Abolfazl Shahbazi, Aurélien Geron, Bairen Yi, gbaned, handong, Hao Ziyu, Junqin Zhang, kiddos, Oliver Sampson, OniB, Runzhen Wang, skawasak, zou000

2.5.0-rc0

TensorFlow Serving using TensorFlow 2.5.0-rc0

Major Features and Improvements

• Upgrade to CUDA 11.2. (commit: 1975e3e)
• Experimental support for serving JAX and XLA/CPU models. (commit: 3c1b2b3)
• Add latency and availability metrics to the Prometheus API (#1623) (commit: dfb41f1)
• Update TF Text to v2.4.3. (commit: ccfb606)
• Support URL reserved characters for REST API (#1726) (commit: dd9c467)
• Add Cross-Origin Resource Sharing (CORS) headers to REST API (#1817)

Breaking Changes

• No breaking changes

Bug Fixes and Other Changes

• Fix typo in REQUIRED_PACKAGES for grpcio (commit: b9ed0f8)
• Update resnet_k8s.yaml file (commit: e7b7b33)
• Fix a compile warning thrown by gcc-9 (commit: 38a017d)
• Fix typo (commit: dbcd54f)
• Update json_tensor.cc (commit: a0a9d14)
• Add TfLiteInterpreterPool to make concurrent use of TfliteSession better (commit: d9efa43)
• Enable download of TF Serving sources at arbitrary commit for CPU docker image. (commit: de1ab9e)
• Updated tests to newer API (commit: 30dd2fe)
• Control number of grpc threads for request handling to avoid OOM (Fixes #1785). (commit: ac0eb73)
• Add dedicated aliases field to ModelServerConfig. (commit: 358f7d1)
• Update docker command line to work with GPUs (Fixes #1768). (commit: b41a28b)
• Option to disable grpc over http (Fixes #1764) (commit: f087290)
• Remove an unused experimental config option "experimental_fixed_input_tensors_filepath". (commit: 3234fca)
• Removing CurriedSession, since it is no longer used. (commit: 87793ad)
• Improve error message for file not exists. (commit: 78d47f7)
• Fix inference request delay when model is switched (Fixes #1796). (commit: 803dd42)
• Transition TensorFlow Serving to TensorFlow's new WORKSPACE protocol. (commit: 50a7ef3)
• Clarifying object values in REST requests to include B64 encoding and similar key:value pair objects. (commit: 0536678)
• Remove experimental comment on TfLiteSession (commit: ab7f9a5)
• Register custom TfLite ParseExample and add benchmark (commit: 20fe3ca)
• Use respectful terms. (commit: b73bd7b)
• Pre-allocate memory for certain vectors where the size is known. (commit: e208b6e)
• Updating serving_basic for adjusting serving_basic.md file and making it up2date with TF2.x - including: (commit: cea306a)
• Use NullSafeStringView for potentially null pointer returned from libevent (commit: a46fdb2)
• Replace nullptr constructor for string_view with empty strings (commit: a98d164)
• Fixing MKL builds due to missing 'build_with_openmp' option (commit: 0ed23df)
• Implement batch parallelism for tflite sessions (commit: fec1d5d)
• Fix GPU docker image massive increase in size (#1813) (commit: 5a0dfd9)
• Fix TensorFlow Serving build with MKL+OpenMP (commit: ddad074)
• Remove hashtable custom op dependencies (commit: bb51722)
• Enable aspired version which failed to load to attempt reload. (commit: 2530a33)
• Fixed a compilation error in aspired_versions_manager.cc (commit: 4ca9a4b)
• Add "_r" root event annotation to ProcessBatch events. (commit: e5c3aec)
• Bump minimum bazel version 3.7.2. (commit: 5edcd13)
• Dont hardcode path to python3. (commit: 63b2d1c)
• Fix package build due to config move in: (commit: 18dd766)
• Add model_service_cc_grpc_proto (commit: a670ff5)
• Fix memory leak from allocating input tensors (commit: 2f9b6a0)
• Allowing lossy floating point conversions for JSON inputs (commit: 57dac6c)
• Adding enable_profiler command line flag. (commit: 7e8720d)
• Add logging in ServerCore. (commit: 623da67)
• Removes mention of ASCII (commit: 8e97b59)

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Abhinav Pundir, Abolfazl Shahbazi, Aurélien Geron, Bairen Yi, gbaned, handong, Hao Ziyu, Junqin Zhang, kiddos, Oliver Sampson, OniB, Runzhen Wang, skawasak, zou000

2.2.2

TensorFlow Serving using TensorFlow 2.2.2.

Major Features and Improvements

• No major features or improvements.

Breaking Changes

• No breaking changes

Bug Fixes and Other Changes

• Reduces startup time on GPUs (commit: 54de106)

2.4.1

TensorFlow Serving using TensorFlow 2.4.1.

Major Features and Improvements

• No major features or improvements.

Breaking Changes

• No breaking changes

Bug Fixes and Other Changes

• This release is based on TF version 2.4.1