TensorFlow Recommenders Addons v0.8.1

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.8.0

What's Changed

• [fix] only upload wheel with tf 2.16.2 due to pip limitation (#470) @jq

Thanks to our Contributors

@jq

Full Changelog: v0.8.0...v0.8.1

TensorFlow Recommenders Addons v0.8.0

Release Notes

This is the 8th release of TensorFlow Recommenders-Addons

Main Features

• [CI] switch the version to 0.8.0 and update the readme (#465) @jq
• use @platforms instead of @bazel_tools//platforms (#457) @LinGeLin
• support tf2.16.2 (#462) @jq
• [Fix] Some compatibility patches for Tensorflow versions and runtime environments (#461) @MoFHeka
• Bump actions/download-artifact from 1 to 4.1.7 in /.github/workflows (#459) @dependabot
• [Feat]combine cpu and gpu wheel (#456) @jq
• [FIX] upgrade to tf 2.16.2 (#454) @jq
• FIX should set synchronous=True for TFRA optimizer (#449) @jq
• [FIX]add scope_name_prefix option to control size of file name (#448) @jq
• [Fix] add HkvHashTableExportWithScores op (#447) @jq
• [fix]remove tracking of shadow var id (#446) @jq
• [New Feature]add normalized_emb (#443) @jq
• [FIX] horovod example and gpu unit test (#442) @jq
• [Fix] Untracked constant tensor cause fail to save Embedding and other bugs (#441) @MoFHeka

Thanks to our Contributors

@LinGeLin, @MoFHeka, @dependabot, @dependabot[bot] and @jq

TensorFlow Recommenders Addons v0.7.2

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.7.x

What's Changed

• [Bug Fix] add tf patch for TensorShape.as_list to allow dynamic shape by @jq in #438

Thanks to our Contributors

@jq

Full Changelog: v0.7.1...v0.7.2

TensorFlow Recommenders Addons v0.7.1

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.7.x

What's Changed

• [feature]support ragged tensor in all2all (#428) @jq
• [Feat] Add new setting num_of_buckets_per_alloc from HKV bata 12. (#433) @MoFHeka
• [doc] Added hkv evict strategy doc (#431) @LinGeLin
• [FIX] raggedtensor should pass name in the eager mode (#427) @jq
• [Doc] Update readme for bf16 (#424) @rhdong

Thanks to our Contributors

@LinGeLin, @MoFHeka, @jq and @rhdong

Full Changelog: v0.7.0...v0.7.1

TensorFlow Recommenders Addons v0.7.0

Release Notes

This is the 7th release of TensorFlow Recommenders-Addons

Main Features:

• [Feat] Support unique in FieldWiseEmbedding and SquashedEmbedding by @rhdong in #311
• [Feat] Add tf-serving config shell script by @rhdong in #339
• [feat] Added support for Tensorflow2 strategy distributed training and Horovod AllToAll synchronous distributed training. by @MoFHeka in #347
• Fix some typing errors by @fuhailin in #354
• [feat] Make saving model more easier when using HvdAllToAllEmbedding by adding save function overwriting patch in tf_save_restore_patch.py. by @MoFHeka in #362
• [Feat] Compatible with Tensorflow later than version 2.11.0. by @MoFHeka in #331
• [feat] Support for HKV by adding HKV source code by @LinGeLin in #377
• [Feature] Compatible with TensorFlow 2.13 and below. Also fix some HKV feature. by @MoFHeka in #378
• [Feat] Compatible with TensorFlow 2.15 by @MoFHeka in #386
• [Feat] Add bfloat16 value type support in Redis and CPU table backend. by @MoFHeka in #419
• [feat] Add macOS Apple Silicon pipeline by @poinwater in #422
• [Feat] Try to compatible with Keras 3 optimizer design and support CUDNN 9.0+. by @MoFHeka in #392
• [FEA]Support more hkv strategy by @LinGeLin in #389
• [FEA]add reserved_key_start_bit option to config hkv by @jq in #408
• [FEA]expose ragged_embedding_ops for safe_embedding_lookup_sparse by @jq in #410

Bug Fixs:

[fix] fix some core dump cases and make usage of CUDA streams more reasonable. Also fix ineffective checkpoint setting. by @MoFHeka in #328

• Fix: de.save_to_file_system not work in estimator by @PWZER in #329
• [fix]fix synchronous trainging error for redis backend by @fuhailin in #360
• [Fix] fix bugs when loading KV files and more. Also make horovod saving function is callable independent bug not patched origin save function. by @MoFHeka in #364
• [fix] DEHvdCheckpoint failed when importing Horovod unsuccessfully. by @MoFHeka in #372
• [Fix] Support TF CheckpointManager in 2.9. And support create Checkpoint with custom model key. by @MoFHeka in #373
• [Fix] CheckpointManager couldn't work properly and KV files couldn't be restored when user using a lazy build Keras model. by @MoFHeka in #374
• [Fix] Several bugs when using HvdAllToAllEmbedding to train model and save/restore KV files. by @MoFHeka in #376
• [fix] Change the parameter max_capacity sent by the cuckoo gpu api to… by @LinGeLin in
• [fix] Fixed setting FileSystemSaver buffer_size does not take effect by @LinGeLin in #384
• [Fix] Compiling error in TF2.13 by @MoFHeka in #396
• [fix] Supported hkv cpu operator base cuckoo cpu operator by @LinGeLin in #387
• [fix] forgot to pass hvd_rank parameter inner a check function of _de_keras_save_func. by @MoFHeka in #398
• [fix] fixed hkv cuda misalign in tf2.15 by @LinGeLin in #399
• [fix] Fixed buffer_size setting when hkv loads and saves files by @LinGeLin in #400
• [Fix] remove .numpy which failed in gpu/distributed env by @jq in #404
• [Fix] Compatible with TF2.15 register_checkpoint_saver when multiple TFRA Embedding in one trackables. by @MoFHeka in #403
• [Fix] Stack overflow in Redis Backend and RestrictPolicy couldn't be saved when using DE embedding layer. by @MoFHeka in #415
• fix version check by @jq in #416

What's Changed

• [CI] switch the version to 0.6.0 and update readme by @rhdong in #314
• Adding the notebook for dynamic_embedding demo by @thushv89 in #315
• [CI] clear CI cache folder and reduce size of CI docker img by @rhdong in #333
• [Doc] serving with triton by @lixiang-repo in #346
• [CI] Add .bazelversion file by @fuhailin in #355
• [build] Disable EV build by @poinwater in #357
• [Refactor]removed a build warning by @LinGeLin in #359
• [CICD]Enhance project's compatibility by @poinwater in #361
• [opt] Optimize memory usage when saving and loading hkv by @LinGeLin in #383
  #382
• [CI]Switch tf version to 2.15.0 by @jq in #388
• [CI] running GPU CI on self-hosted runnners. by @rhdong in #395
• enable ShadowVariable look up for safe_embedding_lookup_sparse by @jq in #397
• [CICD] update dev container by @rhdong in #402
• [opt] set retry_on_failure to false for tf allocator by @LinGeLin in #407
• [CI] Silence the warning during pytest by @jq in #421

Thanks to our Contributors

@fuhailin, @jq, @LinGeLin, @lixiang-repo, @MoFHeka, @poinwater, @PWZER, and @rhdong

New Contributors

• @thushv89 made their first contribution in #315
• @PWZER made their first contribution in #329
• @lixiang-repo made their first contribution in #346
• @fuhailin made their first contribution in #354
• @jq made their first contribution in #388

Full Changelog: v0.6.0...v0.7.0

TensorFlow Recommenders Addons v0.4.1

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.4.x

What's Changed

• Reverse update logic on disabling bp_v2 by @Lifann in #306
  • This optimization can improve the AUC. So we recommend you update it.
  • This optimization only works in session graph mode, and the Keras API would not be affected.

Thanks to @Lifann for finding and fixing it!

TensorFlow Recommenders Addons v0.3.2

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.3.x

What's Changed

• Reverse update logic on disabling bp_v2 by @Lifann in #306
  • This optimization can improve the AUC. So we recommend you update it.
  • This optimization only works in session graph mode, and the Keras API would not be affected.

Thanks to @Lifann for finding and fixing it!

TensorFlow Recommenders Addons v0.6.0

Release Notes

This is the 6th release of TensorFlow Recommenders-Addons

Main Features or Fixs:

• [Fix] Reverse update logic within bp_v2 by @Lifann in #306
  • This bug fix is critical and hurts the AUC. So we highly recommend you update it as soon as possible.
  • The related bug only happens in Graph Mode in synchronized training, and the Keras with eager mode would not be affected.
• [Feat] Support unique in FieldWiseEmbedding and SquashedEmbedding by @rhdong in #311
• [Feat] Add unique to keras.layers.BasicEmbedding by @rhdong in #309
• [Feat]Add the features for expanding and shrinking the number of tables in distributed training by independently saving files. by @MoFHeka in #305
• [Fix] fix error that self._saveable_fn would not exist when checkpoint param is False. by @MoFHeka in #310
• [Fix] Fixed Redis table unit test (#308) @LinGeLin
• [Fix]Fix the bug that rehash might not be triggered by @acmore in #299
• [CI] Removed build warning: 'type qualifier is meaningless on cast type' by @LinGeLin in #298
• [Fix] Remove redundant code by @WangShengguang in #294
• [Fix] no register table kernels for int32-float type. by @rhdong in #293

Thanks to our Contributors

@Lifann, @LinGeLin, @MoFHeka and @rhdong

New Contributors

• @WangShengguang made their first contribution in #294
• @LinGeLin made their first contribution in #298

Full Changelog: https://github.com/tensorflow/recommenders-addons/commits/v0.6.0

TensorFlow Recommenders Addons v0.5.1

Release Notes

This is a patch version for TensorFlow Recommenders-Addons v0.5.0

What's Changed

• Reverse update logic on disabling bp_v2 by @Lifann in #306
  • This optimization can improve the AUC. So we recommend you update it.
  • This optimization only works in session graph mode, and the Keras API would not be affected.

Thanks to @Lifann for finding and fixing it!

TensorFlow Recommenders Addons v0.5.0

Release Notes

This is the 5th release of TensorFlow Recommenders-Addons

Main Features:

• tfra.dynamic_embedding

  • Add int64-int64 key-value types on GPU. by @Lifann in #240
  • Add HMACCUM Redis module and enable BPV2 in Redis implement by @thorneliu in #234, Readme
  • Add redis_hash_tags_hypodispersion and using_hash_storage_slice to Redis backend config. Change redis_connection_mode config, now redis_connection_mode = 2 will be standalone mode. by @MoFHeka in #241, Guide
  • Save and load cuckoo hashtable from HDFS by @luliyucoordinate in #257
  • Add embedding table compression script by @xiangzez in #261, Readme
  • [Opt] Avoid reallocating and copying device memory when the inputs of the Remove and ImportValues functions are already device variables. by @MoFHeka in #279
  • Make HDFS feature support S3 and compatible with tensorflow-io and tensorFlow<=2.6. by @MoFHeka in #281
  • Add bfloat16 value type support for CPU by @xiangzez in #268

What's Changed

• [fix] demo issue for 'horovod_synchronous' -> 'synchronous' by @rhdong in #233
• Add int64-int64 key-value in combination on GPU. by @Lifann in #240
• [Feat] add HMACCUM redis module and enable BPV2 in redis impl by @thorneliu in #234
• [feat] Add redis_hash_tags_hypodispersion and using_hash_storage_slice to redis backend config. Change redis_connection_mode config, now redis_connection_mode = 2 will be standalone mode. by @MoFHeka in #241
• Fix devel container by @rhdong in #248
• [fix] Fixed segment fault due to lambda may capture the thread_context reference with error address by @MoFHeka in #249
• [fix] stopping the Bazel in dev docker to update automatically . by @rhdong in #250
• [refactor] clean some warnings by @rhdong in #252
• fix(comment): Fix error in comment by @Lifann in #255
• [fix] Fixed bug which cause by hiredis updating failed to compile. by @MoFHeka in #260
• fix(optimizer): Bug: Variable set to trainable to false still updating embedding by @Lifann in #263
• [Feat] Save and load cuckoo hashtable from hdfs by @luliyucoordinate in #257
• Add embedding table compression script by @xiangzez in #261
• [Fix]debug CI fail by @rhdong in #267
• Add bfloat16 value type support for CPU by @xiangzez in #268
• [Fix] Avoid double lookup of tables when using ShadowVariable when the Optimizer is updating gradients by @MoFHeka in #262
• [CI] enable unit test on GPU in dev container. by @rhdong in #254
• [fix] Fix bug that user self-defined variable name would be modified to "TrainableWrapper" when Keras eager mode. by @MoFHeka in #275
• [fix] Compatible with compiling problem with TensorFlow version 2.4 and 2.8. by @MoFHeka in #276
• [Fix]Rehash gpu table before importing values and reduce size kernel call. by @acmore in #256
• [feat] Avoid reallocating and copying device memory when the inputs of the Remove and ImportValues functions are already device variables. by @MoFHeka in #279
• [fix] fix shadow variable lookup race condition by @alionkun in #280
• [feat] Make hdfs function compatible with tensorflow-io and TensorFlow version earlier than 2.6 by using FileSystem in TensorFlow Env. Also support S3 and so on now. by @MoFHeka in #281
• [fix] fix device of table shard was lost in savedmodel pb file when save model with keras api. by @MoFHeka in #283
• [release] change version to 0.5.0 without dev by @rhdong in #285

New Contributors

• @xiangzez made their first contribution in #261
• @alionkun made their first contribution in #280

Full Changelog: https://github.com/tensorflow/recommenders-addons/commits/v0.5.0