0.5.0

[Sxnan]

Highlight

• Provide a new set of API for Tensorflow and PyTorch. It is more intuitive and user-friendly.
• Support FlinkML estimator style API in Python, you can check out the quick start examples for Tensorflow 1.15 and PyTorch.
• Support iterative training with data from Flink table.
• Dependencies management of Python API is now seamlessly integrated with the dependencies management of PyFlink, detail can be found here.

What's Changed

• Bump dl-on-flink version by @Sxnan in Bump dl-on-flink version #700
• Update quick start doc by @Sxnan in Update quick start doc #695
• Bump hadoop-common from 2.8.0 to 2.10.1 by @dependabot in Bump hadoop-common from 2.8.0 to 2.10.1 #697
• Bump junit from 4.12 to 4.13.1 by @dependabot in Bump junit from 4.12 to 4.13.1 #688
• Bump httpclient from 4.5.6 to 4.5.13 by @dependabot in Bump httpclient from 4.5.6 to 4.5.13 #684
• Recommend using virtualenv for quick start by @Sxnan in Recommend using virtualenv for quick start #703
• Rename modules from flink-ml-* to dl-on-flink-* by @Sxnan in Rename modules from flink-ml-* to dl-on-flink-* #702
• Add inference example to quick start by @Sxnan in Add inference example to quick start #705
• Run example in github action nighlty by @Sxnan in Run example in github action nighlty #707
• Include pybind as submodule by @Sxnan in Include pybind as submodule #708
• Add github workflow to prepare release by @Sxnan in Add github workflow to prepare release #709
• Remove extra space of commands in quick start by @Sxnan in Remove extra space of commands in quick start #710
• Increase timeout waiting for AM in testing by @Sxnan in Increase timeout waiting for AM in testing #712
• Fix unstable unit test AppMasterServerTest#multiRegisterNode by @Sxnan in Fix unstable unit test AppMasterServerTest#multiRegisterNode #714
• Use the archive url to download flink binary by @Sxnan in Use the archive url to download flink binary #715
• Add checkstyle and spotless by @Sxnan in Add checkstyle and spotless #718
• Refactor RoleUtils and TFUtils API by @Sxnan in Refactor RoleUtils and TFUtils API #716
• NodeUtils set python path properties with entry file path at first by @Sxnan in NodeUtils set python path properties with entry file path at first #721
• Install dev wheels for nightly build by @Sxnan in Install dev wheels for nightly build #722
• AMInputFormat wait for am node finish before return from reachedEnd by @Sxnan in AMInputFormat wait for am node finish before return from reachedEnd #724
• Support Tensorflow on Flink iterative batch training by @Sxnan in Support Tensorflow on Flink iterative batch training #723
• Add Python TFUtils API by @Sxnan in Add Python TFUtils API #726
• Update TensorBoardPythonRunner by @Sxnan in Update TensorBoardPythonRunner #730
• Update DL on Flink Tensorflow Python API by @Sxnan in Update DL on Flink Tensorflow Python API #731
• Use RowCSVCoding as TF default encoding/decoding class by @Sxnan in Use RowCSVCoding as TF default encoding/decoding class #732
• Run python unit test in GitHub workflow by @Sxnan in Run python unit test in GitHub workflow #734
• Support pyFlink env configuration by @Sxnan in Support pyFlink env configuration #733
• Bump Supported Tensorflow version to 2.4.x by @Sxnan in Bump Supported Tensorflow version to 2.4.x #736
• Update Project Copyright Header by @Sxnan in Update Project Copyright Header #737
• Introduce the New PyTorch API by @Sxnan in Introduce the New PyTorch API #735
• Rename PytorchContext to PyTorchContext by @Sxnan in Rename PytorchContext to PyTorchContext #738
• Update quick start examples by @Sxnan in Update quick start examples #739
• Fix prepare release workflow by @Sxnan in Fix prepare release workflow #741
• Specify protobuf version for dl-on-flink-tensorflow wheel build by @Sxnan in Specify protobuf version for dl-on-flink-tensorflow wheel build #742
• DL on Flink support FlinkML API by @Sxnan in DL on Flink support FlinkML API #743
• Fix examples by @Sxnan in Fix examples #745
• Avoid tight loop in NodeOperator#onEpochWatermarkIncremented by @Sxnan in Avoid tight loop in NodeOperator#onEpochWatermarkIncremented #747
• Support adding extra cluster config properties to estimator by @Sxnan in Support adding extra cluster config properties to estimator #746
• Bump fastjson from 1.2.49 to 1.2.83 by @dependabot in Bump fastjson from 1.2.49 to 1.2.83 #748
• Add flink ml estimator examples by @Sxnan in Add flink ml estimator examples #750
• Update NodeOperator and NodeInputFormat close timeout by @Sxnan in Update NodeOperator and NodeInputFormat close timeout #749
• Fix unstable failover test by @Sxnan in Fix unstable failover test #751
• NodeServer#stopMLRunner should not swallow the InterruptedException by @Sxnan in NodeServer#stopMLRunner should not swallow the InterruptedException #752
• Add Java examples for Tensorflow and PyTorch by @Sxnan in Add Java examples for Tensorflow and PyTorch #753
• Add DL on Flink API docs by @Sxnan in Add DL on Flink API docs #754
• NodeServer is interrupted periodically if it is not terminated on time by @Sxnan in NodeServer is interrupted periodically if it is not terminated on time #755
• Specify grpcio version upper bound by @Sxnan in Specify grpcio version upper bound #756

Full Changelog: 0.4.0...0.5.0-rc0