Release 2.4.4

TensorFlow Serving using TensorFlow 2.4.4

Major Features and Improvements

• No Major Features and Improvements

Breaking Changes

• No Breaking changes

Bug Fixes and Other Changes

• Improve error message when version directory is not found (commit: d687d3e)
• Update TF Text to v2.3.0. (commit: 390732a)
• Make Internal Changes (commit: d599e95)
• Add release notes for TF Serving 2.3.0 (commit: 85a6a23)
• In shared batch scheduler, rename 'max_batch_size' to 'input_batch_size_limit'. (commit: ead408b)
• Implement unit test for external version of remote predict op. (commit: 9d4ae67)
• 1. Add an example which call the Remote Predict Op directly. (commit: d5b980f)
• Update default remote model name. (commit: a069a43)
• Adds user guide for Remote Predict Op (commit: fc82463)
• Add support for serving regress/classify (native) TF2 models. (commit: b724ced)
• Use LOG(WARNING) instead of LOG(INFO) for actual allocation problems. (commit: 4a826d0)
• Make Internal Change (commit: e24c50b)
• Move profiler services behind profiler namespace. (commit: d8faed1)
• Temporarily remove sm_80 to fix gpu image build failure. (commit: 6eab5ef)
• Upgrade to CUDA Version 11.0. Build docker images on a local machine successfully with command: docker build --pull -t $USER/tensorflow-serving-devel-gpu -f Dockerfile.devel-gpu . docker build --pull -t $USER/tensorflow-serving-gpu -f Dockerfile.gpu . (commit: b291c4d)
• Internal fix on TFLite support. (commit: 23a5e8b)
• Broaden net_http client visibility and use HTTPStatusCode in ClientResponse status (commit: 7292714)
• Fixit: fix order dependency in batching_session_test. (commit: 0eadb8f)
• Update CUDNN_VERSION to 8.0.4.30. (commit: 84e0189)
• Move some utilities from batching_session to batching_util for upcoming TFRT SavedModel batching supports. (commit: 43e7cfa)
• Temporarily disable failing test, until we root cause and fix it. (commit: aa6c5f5)
• Remove deprecated flag fail_if_no_model_versions_found from tensorflow serving binary (commit: 005f0f6)
• Internal Change. (commit: a05d524)
• Update version for 2.4.0-rc2 release. (commit: 2388d42)
• Adds LICENSE file to tensorflow-serving-api python package. (commit: 9366868)
• Update Dockerfile.devel (commit: e3afe78)
• Update Dockerfile.devel-gpu (commit: 5229c27)
• temporary fix for 2.4.0rc2 docker image (commit: af0ddc0)
• Update Dockerfile.devel (commit: c07d0cc)
• Update Dockerfile.devel-gpu (commit: 4dae597)
• Update setup.py (commit: 5b1ea00)
• Update setup.py (commit: b54e566)
• Update version for 2.4.0-rc3 release. (commit: abe4ffe)
• Update version for 2.4.0-rc4 release. (commit: 16e661a)
• Update version for 2.4.0 release. (commit: 48c4da3)
• Update version for 2.4.1 release. (commit: 03b2bed)
• Make Internal Change (commit: 644baeb)
• Enable download of TF Serving sources at arbitrary commit for GPU/MKL docker images. (commit: 4725d90)
• Fix TensorFlow serving build issue for MKL dockerfiles (commit: 4246080)
• Update TF Text to v2.4.3. (commit: 3bd8417)
• Update version for 2.4.2 release. (commit: a6d405e)
• Update version for 2.4.3 release. (commit: 9a377cd)
• Update version for 2.4.4 release. (#1931) (commit: d5e1076)

Thanks to our Contributors

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

Abolfazl Shahbazi, Adarshreddy Adelli