Summary
* 📢 KerasNLP and KerasCV are now becoming KerasHub 📢. KerasCV and KerasNLP have been consolidated into KerasHub package
* Models available now in KerasHub are albert, bart, bert, bloom, clip, csp_darknet, deberta_v3, deeplab_v3, densenet, distil_bert, efficientnet, electra, f_net, falcon, gemma, gpt2, gpt_neo_x, llama, llama3, mistral, mit, mobilenet, opt, pali_gemma, phi3, resnet, retinanet, roberta, sam, stable_diffusion_3, t5, vae, vgg, vit_det, whisper, xlm_roberta and xlnet.
* A new preprocessor flow has been added for vision and audio models
What's Changed
* Update python version in readme to 3.8 by haifeng-jin in https://github.com/keras-team/keras-hub/pull/618
* Modify our pip install line so we upgrade tf by mattdangerw in https://github.com/keras-team/keras-hub/pull/616
* Use Adam optimizer for quick start by mattdangerw in https://github.com/keras-team/keras-hub/pull/620
* Clean up class name and `self` in calls to `super()` by mbrukman in https://github.com/keras-team/keras-hub/pull/628
* Update word_piece_tokenizer.py by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/617
* Add DeBERTaV3 Conversion Script by abheesht17 in https://github.com/keras-team/keras-hub/pull/633
* Add AlbertTokenizer and AlbertPreprocessor by abheesht17 in https://github.com/keras-team/keras-hub/pull/627
* Create `Backbone` base class by jbischof in https://github.com/keras-team/keras-hub/pull/621
* Add TPU testing by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/591
* Add Base Preprocessor Class by abheesht17 in https://github.com/keras-team/keras-hub/pull/638
* Add keras_nlp.samplers by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/563
* Add ALBERT Backbone by abheesht17 in https://github.com/keras-team/keras-hub/pull/622
* Add a small script to count parameters in our presets by mattdangerw in https://github.com/keras-team/keras-hub/pull/610
* Clean up examples/ directory by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/637
* Fix Small BERT Typo by abheesht17 in https://github.com/keras-team/keras-hub/pull/651
* Rename examples/bert -> examples/bert_pretraining by mattdangerw in https://github.com/keras-team/keras-hub/pull/647
* Add FNet Preprocessor by abheesht17 in https://github.com/keras-team/keras-hub/pull/646
* Add FNet Backbone by abheesht17 in https://github.com/keras-team/keras-hub/pull/643
* Small DeBERTa Docstring Fixes by abheesht17 in https://github.com/keras-team/keras-hub/pull/666
* Add Fenced Docstring Testing by abheesht17 in https://github.com/keras-team/keras-hub/pull/640
* Corrected the epsilon value by soma2000-lang in https://github.com/keras-team/keras-hub/pull/665
* Consolidate docstring formatting weirdness in Backbone and Preprocessor base classes by mattdangerw in https://github.com/keras-team/keras-hub/pull/654
* Fix `value_dim` in `TransformerDecoder`'s cross-attn layer by abheesht17 in https://github.com/keras-team/keras-hub/pull/667
* Add ALBERT Presets by abheesht17 in https://github.com/keras-team/keras-hub/pull/655
* Add Base Task Class by abheesht17 in https://github.com/keras-team/keras-hub/pull/671
* Implement TopP, TopK and Beam samplers by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/652
* Add FNet Presets by abheesht17 in https://github.com/keras-team/keras-hub/pull/659
* Bump the year to 2023 by mattdangerw in https://github.com/keras-team/keras-hub/pull/679
* Add BART Backbone by abheesht17 in https://github.com/keras-team/keras-hub/pull/661
* Handle trainable and name in the backbone base class by mattdangerw in https://github.com/keras-team/keras-hub/pull/680
* Ignore Task Docstring for Testing by abheesht17 in https://github.com/keras-team/keras-hub/pull/683
* Light-weight benchmarking script by NusretOzates in https://github.com/keras-team/keras-hub/pull/664
* Conditionally import tf_text everywhere by mattdangerw in https://github.com/keras-team/keras-hub/pull/684
* Expose `token_embedding` as a Backbone Property by abheesht17 in https://github.com/keras-team/keras-hub/pull/676
* Move `from_preset` to base tokenizer classes by shivance in https://github.com/keras-team/keras-hub/pull/673
* add f_net_classifier and f_net_classifier_test by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/670
* import rouge_scorer directly from rouge_score package by sampathweb in https://github.com/keras-team/keras-hub/pull/691
* Fix typo in requirements file juypter -> jupyter by mattdangerw in https://github.com/keras-team/keras-hub/pull/693
* Temporary fix to get nightly green again by mattdangerw in https://github.com/keras-team/keras-hub/pull/696
* GPT2 Text Generation APIs by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/592
* Run keras saving tests on nightly and fix RobertaClassifier test by mattdangerw in https://github.com/keras-team/keras-hub/pull/692
* Speed up pip install keras-nlp; simplify deps by mattdangerw in https://github.com/keras-team/keras-hub/pull/697
* Add `AlbertClassifier` by shivance in https://github.com/keras-team/keras-hub/pull/668
* Make tokenizer, backbone, preprocessor properties settable on base class by mattdangerw in https://github.com/keras-team/keras-hub/pull/700
* Update to latest black by mattdangerw in https://github.com/keras-team/keras-hub/pull/708
* RobertaMaskedLM task and preprocessor by mattdangerw in https://github.com/keras-team/keras-hub/pull/653
* Default compilation for BERT/RoBERTa classifiers by jbischof in https://github.com/keras-team/keras-hub/pull/695
* Add start/end token padding to `GPT2Preprocessor` by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/704
* Don't install tf stable when building our nightly image by mattdangerw in https://github.com/keras-team/keras-hub/pull/711
* Add OPT Backbone and Tokenizer by mattdangerw in https://github.com/keras-team/keras-hub/pull/699
* Small OPT Doc-string Edits by abheesht17 in https://github.com/keras-team/keras-hub/pull/716
* Default compilation other classifiers by Plutone11011 in https://github.com/keras-team/keras-hub/pull/714
* Add BartTokenizer and BART Presets by abheesht17 in https://github.com/keras-team/keras-hub/pull/685
* Add an add_prefix_space Arg in BytePairTokenizer by shivance in https://github.com/keras-team/keras-hub/pull/715
* Opt presets by mattdangerw in https://github.com/keras-team/keras-hub/pull/707
* fix import of tensorflow_text in tf_utils by sampathweb in https://github.com/keras-team/keras-hub/pull/723
* Check for masked token in roberta tokenizer by mattdangerw in https://github.com/keras-team/keras-hub/pull/742
* Improve test coverage for special tokens in model tokenizers by mattdangerw in https://github.com/keras-team/keras-hub/pull/743
* Fix the sampler truncation strategy by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/713
* Add ALBERT Conversion Script by abheesht17 in https://github.com/keras-team/keras-hub/pull/736
* Add FNet Conversion Script by abheesht17 in https://github.com/keras-team/keras-hub/pull/737
* Add BART Conversion Script by abheesht17 in https://github.com/keras-team/keras-hub/pull/739
* Pass Correct LayerNorm Epsilon value to TransformerEncoder in Backbones by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/731
* Improving the layer Description. by Neeshamraghav012 in https://github.com/keras-team/keras-hub/pull/734
* Adding ragged support to SinePositionEncoding by apupneja in https://github.com/keras-team/keras-hub/pull/751
* Fix trailing space by mattdangerw in https://github.com/keras-team/keras-hub/pull/755
* Adding an AlbertMaskedLM task + Fix Projection layer dimension in MaskedLMHead by shivance in https://github.com/keras-team/keras-hub/pull/725
* New docstring example for TokenAndPosition Embedding layer. by Neeshamraghav012 in https://github.com/keras-team/keras-hub/pull/760
* Add a note for TPU issues for deberta_v3 by mattdangerw in https://github.com/keras-team/keras-hub/pull/758
* Add missing exports to models API by mattdangerw in https://github.com/keras-team/keras-hub/pull/763
* Autogenerate preset table by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/690
* Version bump to 0.5.0 by mattdangerw in https://github.com/keras-team/keras-hub/pull/767
* Adding a FNetMaskedLM task model and preprocessor by apupneja in https://github.com/keras-team/keras-hub/pull/740
* Add a DistilBertMaskedLM task model by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/724
* Add cache support to decoding journey by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/745
* Handle [MASK] token in DebertaV3Tokenizer by abheesht17 in https://github.com/keras-team/keras-hub/pull/759
* Update README for 2.4.1 release by mattdangerw in https://github.com/keras-team/keras-hub/pull/757
* Fix typo in test docstring by jbischof in https://github.com/keras-team/keras-hub/pull/791
* Fixed Incorrect Links for FNet and DeBERTaV3 models by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/793
* Patch 1 - doc-string spell fix by atharvapurdue in https://github.com/keras-team/keras-hub/pull/781
* Don't rely on core keras initializer config details by mattdangerw in https://github.com/keras-team/keras-hub/pull/802
* Simplify the cache decoding graph by mattdangerw in https://github.com/keras-team/keras-hub/pull/780
* Fix Fenced Doc-String 782 by atharvapurdue in https://github.com/keras-team/keras-hub/pull/785
* Solve 721 Deberta masklm model by Plutone11011 in https://github.com/keras-team/keras-hub/pull/732
* Add from_config to sampler by mattdangerw in https://github.com/keras-team/keras-hub/pull/803
* BertMaskedLM Task Model and Preprocessor by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/774
* Stop generation once end_token_id is seen by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/769
* Added model card links for all pretrained models. by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/795
* Initial PR demonstrating public API export logic. by fchollet in https://github.com/keras-team/keras-hub/pull/747
* Add preset for finetuning GPT2 on CNN news by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/807
* Add API exports for metrics documented on keras.io by shivance in https://github.com/keras-team/keras-hub/pull/816
* Add API exports for samplers documented on keras.io by shivance in https://github.com/keras-team/keras-hub/pull/815
* Add API exports for models documented on keras.io by shivance in https://github.com/keras-team/keras-hub/pull/814
* Add API exports for tokenizers documented on keras.io by shivance in https://github.com/keras-team/keras-hub/pull/817
* Add API exports for layers documented on keras.io by fchollet in https://github.com/keras-team/keras-hub/pull/811
* Add keras_nlp.utils public API exports. by fchollet in https://github.com/keras-team/keras-hub/pull/819
* retrained bert_tiny_uncased_en_sst2_training.ipynb by susnato in https://github.com/keras-team/keras-hub/pull/771
* Temporary solution to avoid recompilation by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/808
* Call super.config() in BartBackbone's get_config() by shivance in https://github.com/keras-team/keras-hub/pull/818
* Update typo in README.md by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/821
* Add Whisper Backbone by abheesht17 in https://github.com/keras-team/keras-hub/pull/801
* Added note for tensorflow-text in the CONTRIBUTING guide by jaygala223 in https://github.com/keras-team/keras-hub/pull/805
* Roadmap update by jaygala223 in https://github.com/keras-team/keras-hub/pull/800
* Remove API export decorator from base classes by shivance in https://github.com/keras-team/keras-hub/pull/824
* Move integration tests out of repo sources. by fchollet in https://github.com/keras-team/keras-hub/pull/826
* Function merge_padding_and_attention_mask does not return an output with the desired shape when both padding and attention masks are given by abodinier in https://github.com/keras-team/keras-hub/pull/790
* Adding XXBackboneTPUTests by shivance in https://github.com/keras-team/keras-hub/pull/839
* Add a t5 tokenizer by mattdangerw in https://github.com/keras-team/keras-hub/pull/852
* Add compilation defaults for the BertMaskedLM task model by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/836
* added __init__ file for t5 by Akorex in https://github.com/keras-team/keras-hub/pull/853
* Modified Docstring for GPT2CasualLM by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/855
* Rework bert docstrings for progressive disclosure of complexity by mattdangerw in https://github.com/keras-team/keras-hub/pull/843
* Fix "causal" spelling in export decorator by abheesht17 in https://github.com/keras-team/keras-hub/pull/861
* Default compilation for Albert, Distilbert, Roberta MaskedLM by shivance in https://github.com/keras-team/keras-hub/pull/833
* Speed up default BERT testing roughly 3x by mattdangerw in https://github.com/keras-team/keras-hub/pull/859
* Add compilation defaults for the Fnet MaskedLM task model by soma2000-lang in https://github.com/keras-team/keras-hub/pull/834
* Default compilation for Debertav3MaskedLM model by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/835
* Remove from_preset from fnet tokenizer by mattdangerw in https://github.com/keras-team/keras-hub/pull/865
* Add T5 backbone by fchollet in https://github.com/keras-team/keras-hub/pull/828
* Speeding the tests for opt by susnato in https://github.com/keras-team/keras-hub/pull/886
* Move generate compilation to the task model by mattdangerw in https://github.com/keras-team/keras-hub/pull/804
* Speeding the tests for xlm_roberta by susnato in https://github.com/keras-team/keras-hub/pull/885
* Rework DistilBERT docstrings for progressive disclosure of complexity. by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/881
* Speeding the tests for T5 by susnato in https://github.com/keras-team/keras-hub/pull/888
* Rework OPT docstrings for progressive disclosure of complexity. by Warlord-K in https://github.com/keras-team/keras-hub/pull/893
* Get our fenced docstring tests working again by mattdangerw in https://github.com/keras-team/keras-hub/pull/895
* Speed up default RoBERTa testing roughly 3x by shivance in https://github.com/keras-team/keras-hub/pull/897
* Speeding the tests for whisper by susnato in https://github.com/keras-team/keras-hub/pull/887
* Update BytePairTokenizerCache to have similar dtypes for x and y in self.factors. by Sruinard in https://github.com/keras-team/keras-hub/pull/871
* Init `_backbone`, `_tokenizer` and `_preprocessor` in Task by jbischof in https://github.com/keras-team/keras-hub/pull/899
* Rework Whisper docstrings for progressive disclosure of complexity by susnato in https://github.com/keras-team/keras-hub/pull/903
* Speed up default DeBERTa_v3 testing roughly 3x by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/905
* Rework docstring of XLMRoberta by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/882
* Stripping the MASK token by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/876
* Possible fix for task.summary() by mattdangerw in https://github.com/keras-team/keras-hub/pull/901
* Speed up default FNet testing speedups. by Cyber-Machine in https://github.com/keras-team/keras-hub/pull/894
* Added TPU test for DebertaV3Backbone by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/924
* Fix failing TPU tests by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/931
* Add model contribution guide by abheesht17 in https://github.com/keras-team/keras-hub/pull/820
* Resolved roberta_checkpoint by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/874
* GLUE evaluation automation script by susnato in https://github.com/keras-team/keras-hub/pull/848
* Ensure shape in sample so that the shape is correct after TFLite conversion by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/902
* Returning all Beams and Probs and adding a Testing Unit by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/908
* Roberta docstring reworking by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/910
* Speeding the tests for Albert by soma2000-lang in https://github.com/keras-team/keras-hub/pull/873
* Mlm mask generator docstring adding example by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/916
* Don't save traces for saved model by mattdangerw in https://github.com/keras-team/keras-hub/pull/945
* Bump stable tf version to 2.12 by mattdangerw in https://github.com/keras-team/keras-hub/pull/944
* Speeding the tests for DistilBert by soma2000-lang in https://github.com/keras-team/keras-hub/pull/872
* Allow BPE to treat special tokens as one token by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/939
* Edit examples in samplers by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/957
* Add RandomSampler to Samplers by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/952
* Add BartPreprocessor by abheesht17 in https://github.com/keras-team/keras-hub/pull/856
* Remove the old sampler utilities by mattdangerw in https://github.com/keras-team/keras-hub/pull/948
* Use direct imports everywhere in library by mattdangerw in https://github.com/keras-team/keras-hub/pull/961
* Update docstrings for relocated `sampler` arg by jbischof in https://github.com/keras-team/keras-hub/pull/964
* Fix gpt2, t5 and fnet under mixed precision by mattdangerw in https://github.com/keras-team/keras-hub/pull/958
* Small fixes for special_tokens arg in BPE by abheesht17 in https://github.com/keras-team/keras-hub/pull/969
* Add contrastive sampler by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/896
* Mark num_classes as required in Classifier classes by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/971
* Rework model docstrings for progressive disclosure of complexity for f_net by ADITYADAS1999 in https://github.com/keras-team/keras-hub/pull/879
* Handle OOV token in XLMRoBERTaTokenizer's token_to_id function by abheesht17 in https://github.com/keras-team/keras-hub/pull/968
* Clean up the docker and lint setup by haifeng-jin in https://github.com/keras-team/keras-hub/pull/981
* Update generate() to work like fit() and predict() by mattdangerw in https://github.com/keras-team/keras-hub/pull/932
* Speed top-p sampler up by only sampling from top-k tokens by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/980
* Expose the generate_step compilable function by mattdangerw in https://github.com/keras-team/keras-hub/pull/982
* Fix decoder inputs in BART preprocessor by abheesht17 in https://github.com/keras-team/keras-hub/pull/984
* Convert string tensors to python strings in `generate()` by mattdangerw in https://github.com/keras-team/keras-hub/pull/983
* Adding a temperature argument to the base sampler class and related tests by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/951
* Track the task preprocessor layer as part of model by mattdangerw in https://github.com/keras-team/keras-hub/pull/985
* Add an XLMRobertaMaskedLM task model by shivance in https://github.com/keras-team/keras-hub/pull/950
* Add an activation argument to all classifiers by mattdangerw in https://github.com/keras-team/keras-hub/pull/991
* Remove activation from README quickstart by mattdangerw in https://github.com/keras-team/keras-hub/pull/992
* Rework albert docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/993
* Rework bart docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/994
* Rework deberta docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/995
* Misc fixes to docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/996
* Added temperature argument to the Contrastive Sampler by TheAthleticCoder in https://github.com/keras-team/keras-hub/pull/997
* Add `OPTCausalLM` and preprocessors by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/990
* Version bump to 0.5.0.dev0 by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1002
* Add a flag to restrict which docstring tests run by mattdangerw in https://github.com/keras-team/keras-hub/pull/999
* fix docstring for 0.5 release by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1005
* Serialize activation fn properly by mattdangerw in https://github.com/keras-team/keras-hub/pull/1007
* Try adding an error if activation and loss are mismatched by mattdangerw in https://github.com/keras-team/keras-hub/pull/1008
* Fix docstring for 0.5 release by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1009
* Switch to using pip_build for release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1011
* Make version number SSoT. by fchollet in https://github.com/keras-team/keras-hub/pull/827
* Add DTensor layout map class method for OPT by mattdangerw in https://github.com/keras-team/keras-hub/pull/1000
* Add DTensor layout map class method for GPT-2 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1014
* Standalone functions for generate pre/post processing for GPT-2 by mattdangerw in https://github.com/keras-team/keras-hub/pull/998
* install namex in the publish workflow by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1020
* Update publish-to-pypi.yml by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1021
* Standalone functions for generate pre/post processing for OPT by mattdangerw in https://github.com/keras-team/keras-hub/pull/1015
* Fix typos in export by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1024
* Fix unclosed fenced docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/1025
* Fix a bug with computing the output mask after generate by mattdangerw in https://github.com/keras-team/keras-hub/pull/1029
* small updates to the release doc by chenmoneygithub in https://github.com/keras-team/keras-hub/pull/1031
* Sampler docstring edit by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/1033
* Fix program crash for id_to_token() method in SentencePieceTokenizer by abuelnasr0 in https://github.com/keras-team/keras-hub/pull/1040
* Update our release process to preview docs before release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1043
* Add Whisper Tokenizer and Audio Feature Extractor by abheesht17 in https://github.com/keras-team/keras-hub/pull/847
* Also strip padding token for opt by mattdangerw in https://github.com/keras-team/keras-hub/pull/1028
* Add regex dep by mattdangerw in https://github.com/keras-team/keras-hub/pull/1044
* Add BartSeq2SeqLM and conditional text generation with BART by abheesht17 in https://github.com/keras-team/keras-hub/pull/974
* Support list/tuple inputs for special tokens in StartEndPacker layer by abheesht17 in https://github.com/keras-team/keras-hub/pull/1045
* Support list/tuple inputs for special tokens in MultiSegmentPacker layer by abheesht17 in https://github.com/keras-team/keras-hub/pull/1046
* Fix a misleading part of our cached MHA docs by mattdangerw in https://github.com/keras-team/keras-hub/pull/1048
* Always pass weight name by kwarg by mattdangerw in https://github.com/keras-team/keras-hub/pull/1053
* Always pass metrics in a list or dict by mattdangerw in https://github.com/keras-team/keras-hub/pull/1054
* Move `Defaults to` to end of arg docstring and standardise values by SamuelMarks in https://github.com/keras-team/keras-hub/pull/1057
* Fix beam search for BART by abheesht17 in https://github.com/keras-team/keras-hub/pull/1058
* Replace tf.dtype with "dtype" by mattdangerw in https://github.com/keras-team/keras-hub/pull/1059
* Test shapes directly by mattdangerw in https://github.com/keras-team/keras-hub/pull/1064
* Clean up metrics tests by mattdangerw in https://github.com/keras-team/keras-hub/pull/1063
* Remove metrics merge tests by mattdangerw in https://github.com/keras-team/keras-hub/pull/1065
* Fix whisper feature inputs by mattdangerw in https://github.com/keras-team/keras-hub/pull/1069
* Always specify shape when creating variables by mattdangerw in https://github.com/keras-team/keras-hub/pull/1067
* Remove ragged support from position embeddings by mattdangerw in https://github.com/keras-team/keras-hub/pull/1068
* Clean up dtype handling for preprocessing layers by mattdangerw in https://github.com/keras-team/keras-hub/pull/1066
* Add BART finetuned on CNN+DM for summarisation by abheesht17 in https://github.com/keras-team/keras-hub/pull/1060
* Fix saving bug by mattdangerw in https://github.com/keras-team/keras-hub/pull/1073
* Fix t5 forward pass by mattdangerw in https://github.com/keras-team/keras-hub/pull/1082
* Feat/make transformer decoder callable without causal mask by ferraric in https://github.com/keras-team/keras-hub/pull/1083
* Adding `GPTNeoXBackbone` by shivance in https://github.com/keras-team/keras-hub/pull/1056
* Add a common test case by mattdangerw in https://github.com/keras-team/keras-hub/pull/1095
* Update register_keras_serializable to use saving module by mattdangerw in https://github.com/keras-team/keras-hub/pull/1094
* Don't test tf format by mattdangerw in https://github.com/keras-team/keras-hub/pull/1104
* Add `GPTNeoXPreprocessor` by shivance in https://github.com/keras-team/keras-hub/pull/1093
* Split layers into layers/modeling & layers/preprocessing by mattdangerw in https://github.com/keras-team/keras-hub/pull/1102
* Fix merge conflict from 1102 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1105
* Add a common base class for generative models by mattdangerw in https://github.com/keras-team/keras-hub/pull/1096
* Add `GPTNeoXCausalLMPreprocessor` by shivance in https://github.com/keras-team/keras-hub/pull/1106
* Add Whisper Presets by abheesht17 in https://github.com/keras-team/keras-hub/pull/1089
* Refactor `RotaryEmbedding` and `GPTNeoXAttention` by shivance in https://github.com/keras-team/keras-hub/pull/1101
* Remove all the secret keys for ci by mattdangerw in https://github.com/keras-team/keras-hub/pull/1126
* Fix publish to pypi action by mattdangerw in https://github.com/keras-team/keras-hub/pull/1127
* Update README for Keras Core by jbischof in https://github.com/keras-team/keras-hub/pull/1135
* Ignore errors in UTF-8 decoding by abheesht17 in https://github.com/keras-team/keras-hub/pull/1150
* Ports GPTNeoX to KerasCore by shivance in https://github.com/keras-team/keras-hub/pull/1137
* Small fix for mixed precision generation on tf by mattdangerw in https://github.com/keras-team/keras-hub/pull/1153
* Port DeBERTa to multi-backend by abheesht17 in https://github.com/keras-team/keras-hub/pull/1155
* Change all tensors passed to tf.data.Dataset to numpy by mattdangerw in https://github.com/keras-team/keras-hub/pull/1161
* Fix broken tests by mattdangerw in https://github.com/keras-team/keras-hub/pull/1163
* Pin keras-core to 0.1.0 while investigating failures by mattdangerw in https://github.com/keras-team/keras-hub/pull/1168
* Run GPU tests on Jax + Torch by ianstenbit in https://github.com/keras-team/keras-hub/pull/1160
* Fix flakes in masked lm testing by removing any indeterminism by mattdangerw in https://github.com/keras-team/keras-hub/pull/1171
* Always install the correct version with pip_build by mattdangerw in https://github.com/keras-team/keras-hub/pull/1174
* Remove tests for preprocessing inside a functional model by mattdangerw in https://github.com/keras-team/keras-hub/pull/1175
* Extend the timeout for large tests by mattdangerw in https://github.com/keras-team/keras-hub/pull/1103
* Add `GPTNeoXCausalLM` by shivance in https://github.com/keras-team/keras-hub/pull/1110
* Bump tensorflow to latest stable by mattdangerw in https://github.com/keras-team/keras-hub/pull/1170
* Add compute_output_shape to tokenizer by shivance in https://github.com/keras-team/keras-hub/pull/1166
* Stop pinning keras-core by mattdangerw in https://github.com/keras-team/keras-hub/pull/1178
* Port FNet by abheesht17 in https://github.com/keras-team/keras-hub/pull/1164
* Automate the update image flow by mattdangerw in https://github.com/keras-team/keras-hub/pull/1179
* Restore mask_position argument name by mattdangerw in https://github.com/keras-team/keras-hub/pull/1185
* Port contrastive sampler to multi-backend by mattdangerw in https://github.com/keras-team/keras-hub/pull/1187
* Port `BeamSampler` to core by shivance in https://github.com/keras-team/keras-hub/pull/1181
* Port metrics to multi-backend by mattdangerw in https://github.com/keras-team/keras-hub/pull/1186
* Generic `RotaryEmbedding` Layer by shivance in https://github.com/keras-team/keras-hub/pull/1180
* Raise ValueError when number of dims evaluate to zero by sampathweb in https://github.com/keras-team/keras-hub/pull/1198
* Add XLNetBackbone by susnato in https://github.com/keras-team/keras-hub/pull/1084
* Switch from tf.nest to dm-tree by mattdangerw in https://github.com/keras-team/keras-hub/pull/1199
* Fix CI for keras-core 0.1.4 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1202
* Fix ModuleNotFoundError `keras_nlp.models.xlnet` by shivance in https://github.com/keras-team/keras-hub/pull/1204
* Add support for "untied" embedding weights in language models by mattdangerw in https://github.com/keras-team/keras-hub/pull/1201
* Add start_index argument to all position embedding layers by mattdangerw in https://github.com/keras-team/keras-hub/pull/1209
* Remove windows line endings by mattdangerw in https://github.com/keras-team/keras-hub/pull/1210
* Fix Autograph error with perplexity metric by shivance in https://github.com/keras-team/keras-hub/pull/1211
* [JAX backend]: Fix errors with perplexity by shivance in https://github.com/keras-team/keras-hub/pull/1213
* Improve layer naming consistency by mattdangerw in https://github.com/keras-team/keras-hub/pull/1219
* Stop asserting key order in bart preprocessor by mattdangerw in https://github.com/keras-team/keras-hub/pull/1221
* Remove file level docstrings by mattdangerw in https://github.com/keras-team/keras-hub/pull/1222
* Fix typos by mattdangerw in https://github.com/keras-team/keras-hub/pull/1220
* Typo fix by mattdangerw in https://github.com/keras-team/keras-hub/pull/1223
* Fix RotaryEmbedding import by shivance in https://github.com/keras-team/keras-hub/pull/1217
* Update transformer_decoder for the proper naming of the sublayers. by qlzh727 in https://github.com/keras-team/keras-hub/pull/1230
* Replace tf with numpy by mattdangerw in https://github.com/keras-team/keras-hub/pull/1232
* Update to always using ops.shape by mattdangerw in https://github.com/keras-team/keras-hub/pull/1231
* Add a test harness based on keras-core's `run_layer_test` by mattdangerw in https://github.com/keras-team/keras-hub/pull/1238
* fixed token_to_id doc + error msg by jackd in https://github.com/keras-team/keras-hub/pull/1240
* Changed default TokenAndPositionEmbedding initializer to 'uniform' by jackd in https://github.com/keras-team/keras-hub/pull/1237
* Add compat shims for the upcoming keras-core release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1244
* Depend on latest keras-core by mattdangerw in https://github.com/keras-team/keras-hub/pull/1246
* Removed the undefined self.sequence_length by sahusiddharth in https://github.com/keras-team/keras-hub/pull/1245
* Bump devcontainer to 3.9 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1249
* Add a mixed precision test and fix mixed precision errors for layers by mattdangerw in https://github.com/keras-team/keras-hub/pull/1242
* Quick fix for 0.1.7 keras-core release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1251
* Small docstring fixes for the upcoming release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1253
* Don't export model internals publicly by mattdangerw in https://github.com/keras-team/keras-hub/pull/1255
* Bump master branch version number to 0.7.0.dev0 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1254
* Fix/allow different encoder and decoder feature dimensions in transformer decoder layer by ferraric in https://github.com/keras-team/keras-hub/pull/1260
* Doc updates to switch branding to Keras 3 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1259
* Remove unused TPU testing for backbones by mattdangerw in https://github.com/keras-team/keras-hub/pull/1266
* Make gelu a function, not a lambda so it can be loaded without safe_mode=False by calvingiles in https://github.com/keras-team/keras-hub/pull/1262
* Update requirements and install instructions for multi-backend keras by mattdangerw in https://github.com/keras-team/keras-hub/pull/1257
* Support Keras 3 installation by mattdangerw in https://github.com/keras-team/keras-hub/pull/1258
* Remove dtensor by mattdangerw in https://github.com/keras-team/keras-hub/pull/1268
* Add a lora dense layer by mattdangerw in https://github.com/keras-team/keras-hub/pull/1263
* Factor out testing routines for models by mattdangerw in https://github.com/keras-team/keras-hub/pull/1269
* Convert T5 to Keras 3 by nkovela1 in https://github.com/keras-team/keras-hub/pull/1274
* Fix missing backticks in DistilBertClassifier docstrings by Philmod in https://github.com/keras-team/keras-hub/pull/1278
* T5 checkpoint conversion with HF by nkovela1 in https://github.com/keras-team/keras-hub/pull/1277
* Use gelu_approximate directly in t5 presets by mattdangerw in https://github.com/keras-team/keras-hub/pull/1284
* Add preset tests and weights URLs by nkovela1 in https://github.com/keras-team/keras-hub/pull/1285
* Support loading keras 3 nightly by mattdangerw in https://github.com/keras-team/keras-hub/pull/1286
* Remove the use of `SentencePieceTrainer` from tests by tirthasheshpatel in https://github.com/keras-team/keras-hub/pull/1283
* Fix XLM-RoBERTa detokenize() by abheesht17 in https://github.com/keras-team/keras-hub/pull/1289
* Correct tie_embedding_weights and add logit checking by nkovela1 in https://github.com/keras-team/keras-hub/pull/1288
* Add detokenize testing for model tokenizers by mattdangerw in https://github.com/keras-team/keras-hub/pull/1290
* Fix Whisper by abheesht17 in https://github.com/keras-team/keras-hub/pull/1287
* Test against Keras 3 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1273
* Support TF_USE_LEGACY_KERAS by mattdangerw in https://github.com/keras-team/keras-hub/pull/1295
* Run workflows with read-only tokens by pnacht in https://github.com/keras-team/keras-hub/pull/1305
* Update CONTRIBUTING.md by mattdangerw in https://github.com/keras-team/keras-hub/pull/1310
* Add GitHub Action for Nightly by sampathweb in https://github.com/keras-team/keras-hub/pull/1309
* Fix the publish to pypi action by mattdangerw in https://github.com/keras-team/keras-hub/pull/1311
* Fix nightly tf failure by mattdangerw in https://github.com/keras-team/keras-hub/pull/1316
* Switch deberta to use the "int" dtype by mattdangerw in https://github.com/keras-team/keras-hub/pull/1315
* Add security policy by pnacht in https://github.com/keras-team/keras-hub/pull/1319
* Fix missing export for reversible embedding by mattdangerw in https://github.com/keras-team/keras-hub/pull/1327
* Add `version` API to keras_nlp by grasskin in https://github.com/keras-team/keras-hub/pull/1324
* Fix Keras 3 version check by sampathweb in https://github.com/keras-team/keras-hub/pull/1328
* Simplify running KerasNLP with Keras 3 by mattdangerw in https://github.com/keras-team/keras-hub/pull/1308
* Fix issues with version by mattdangerw in https://github.com/keras-team/keras-hub/pull/1332
* Fix typo in whisper presets files by mattdangerw in https://github.com/keras-team/keras-hub/pull/1337
* `ELECTRA` backbone implementation in keras by pranavvp16 in https://github.com/keras-team/keras-hub/pull/1291
* Fix t5 tokenizer expected output by mattdangerw in https://github.com/keras-team/keras-hub/pull/1348
* Add __init__.py for electra by mattdangerw in https://github.com/keras-team/keras-hub/pull/1352
* Remove lora dense for now by mattdangerw in https://github.com/keras-team/keras-hub/pull/1359
* Adds Kokoro Build script for Keras-NLP GPU tests by sampathweb in https://github.com/keras-team/keras-hub/pull/1355
* Fixes GPU Test failures for Keras 3 by sampathweb in https://github.com/keras-team/keras-hub/pull/1361
* Change Continuous config to also run only large tests by sampathweb in https://github.com/keras-team/keras-hub/pull/1362
* ElectraTokenizer by pranavvp16 in https://github.com/keras-team/keras-hub/pull/1357
* Add MistralAI's 7B Transformer as a backbone in KerasNLP Models by tirthasheshpatel in https://github.com/keras-team/keras-hub/pull/1314
* changing pooling output by mbrhd in https://github.com/keras-team/keras-hub/pull/1364
* Add `LlamaBackbone` by shivance in https://github.com/keras-team/keras-hub/pull/1203
* Align pip_build with keras by sampathweb in https://github.com/keras-team/keras-hub/pull/1374
* Remove cloudbuild config by mattdangerw in https://github.com/keras-team/keras-hub/pull/1375
* Fix one last bad preset hash by mattdangerw in https://github.com/keras-team/keras-hub/pull/1381
* Add a tokenizer for the Mistral backbone by tirthasheshpatel in https://github.com/keras-team/keras-hub/pull/1383
* Kaggle Presets by sampathweb in https://github.com/keras-team/keras-hub/pull/1365
* Fix mistral and electra tokenizer to match kaggle changes by mattdangerw in https://github.com/keras-team/keras-hub/pull/1387
* Align requirments with Keras by sampathweb in https://github.com/keras-team/keras-hub/pull/1386
* Add a preprocessor for the Mistral backbone by tirthasheshpatel in https://github.com/keras-team/keras-hub/pull/1385
* Switch to always expect full Kaggle preset handles by mattdangerw in https://github.com/keras-team/keras-hub/pull/1390
* Version bump for dev release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1391
* Version bump for final release by mattdangerw in https://github.com/keras-team/keras-hub/pull/1392
New Contributors
* haifeng-jin made their first contribution in https://github.com/keras-team/keras-hub/pull/618
* mbrukman made their first contribution in https://github.com/keras-team/keras-hub/pull/628
* soma2000-lang made their first contribution in https://github.com/keras-team/keras-hub/pull/665
* NusretOzates made their first contribution in https://github.com/keras-team/keras-hub/pull/664
* shivance made their first contribution in https://github.com/keras-team/keras-hub/pull/673
* Plutone11011 made their first contribution in https://github.com/keras-team/keras-hub/pull/714
* TheAthleticCoder made their first contribution in https://github.com/keras-team/keras-hub/pull/731
* Neeshamraghav012 made their first contribution in https://github.com/keras-team/keras-hub/pull/734
* apupneja made their first contribution in https://github.com/keras-team/keras-hub/pull/751
* Cyber-Machine made their first contribution in https://github.com/keras-team/keras-hub/pull/690
* atharvapurdue made their first contribution in https://github.com/keras-team/keras-hub/pull/781
* susnato made their first contribution in https://github.com/keras-team/keras-hub/pull/771
* jaygala223 made their first contribution in https://github.com/keras-team/keras-hub/pull/805
* abodinier made their first contribution in https://github.com/keras-team/keras-hub/pull/790
* Akorex made their first contribution in https://github.com/keras-team/keras-hub/pull/853
* Warlord-K made their first contribution in https://github.com/keras-team/keras-hub/pull/893
* Sruinard made their first contribution in https://github.com/keras-team/keras-hub/pull/871
* SamuelMarks made their first contribution in https://github.com/keras-team/keras-hub/pull/1057
* ferraric made their first contribution in https://github.com/keras-team/keras-hub/pull/1083
* ianstenbit made their first contribution in https://github.com/keras-team/keras-hub/pull/1160
* jackd made their first contribution in https://github.com/keras-team/keras-hub/pull/1240
* sahusiddharth made their first contribution in https://github.com/keras-team/keras-hub/pull/1245
* calvingiles made their first contribution in https://github.com/keras-team/keras-hub/pull/1262
* mbrhd made their first contribution in https://github.com/keras-team/keras-hub/pull/1364
**Full Changelog**: https://github.com/keras-team/keras-hub/compare/v0.4.0...v0.17.0.dev0