github keras-team/keras v2.8.0-rc0
Keras Release 2.8.0 RC0
on GitHub

latest releases: v3.5.0, v3.4.1, v3.4.0...
pre-release2 years ago

Please see https://github.com/tensorflow/tensorflow/blob/r2.8/RELEASE.md for Keras release notes.

  • tf.keras:
    • Preprocessing Layers
      • Added a tf.keras.layers.experimental.preprocessing.HashedCrossing
        layer which applies the hashing trick to the concatenation of crossed
        scalar inputs. This provides a stateless way to try adding feature crosses
        of integer or string data to a model.
      • Removed keras.layers.experimental.preprocessing.CategoryCrossing. Users
        should migrate to the HashedCrossing layer or use
        tf.sparse.cross/tf.ragged.cross directly.
      • Added additional standardize and split modes to TextVectorization.
        • standardize="lower" will lowercase inputs.
        • standardize="string_punctuation" will remove all puncuation.
        • split="character" will split on every unicode character.
      • Added an output_mode argument to the Discretization and Hashing
        layers with the same semantics as other preprocessing layers. All
        categorical preprocessing layers now support output_mode.
      • All preprocessing layer output will follow the compute dtype of a
        tf.keras.mixed_precision.Policy, unless constructed with
        output_mode="int" in which case output will be tf.int64.
        The output type of any preprocessing layer can be controlled individually
        by passing a dtype argument to the layer.
    • tf.random.Generator for keras initializers and all RNG code.
      • Added 3 new APIs for enable/disable/check the usage of
        tf.random.Generator in keras backend, which will be the new backend for
        all the RNG in Keras. We plan to switch on the new code path by default in
        tf 2.8, and the behavior change will likely to cause some breakage on user
        side (eg if the test is checking against some golden nubmer). These 3 APIs
        will allow user to disable and switch back to legacy behavior if they
        prefer. In future (eg tf 2.10), we expect to totally remove the legacy
        code path (stateful random Ops), and these 3 APIs will be removed as well.
    • tf.keras.callbacks.experimental.BackupAndRestore is now available as
      tf.keras.callbacks.BackupAndRestore. The experimental endpoint is
      deprecated and will be removed in a future release.
    • tf.keras.experimental.SidecarEvaluator is now available as
      tf.keras.utils.SidecarEvaluator. The experimental endpoint is
      deprecated and will be removed in a future release.
    • Metrics update and collection logic in default Model.train_step() is now
      customizable via overriding Model.compute_metrics().
    • Losses computation logic in default Model.train_step() is now
      customizable via overriding Model.compute_loss().
    • jit_compile added to Model.compile() on an opt-in basis to compile the
      model's training step with XLA. Note that
      jit_compile=True may not necessarily work for all models.

What's Changed

  • Cleanup legacy Keras files by @qlzh727 in #14256
  • Sync OSS keras to head. by @qlzh727 in #14300
  • Update build script for GPU build. by @copybara-service in #14336
  • Move the LossReduction class from tf to Keras. by @copybara-service in #14362
  • Update keras API generate script. by @copybara-service in #14418
  • Adding extra target that are needed by PIP package dependency. by @copybara-service in #14421
  • Add test related targets to PIP package list. by @copybara-service in #14427
  • Sync OSS keras to head. by @copybara-service in #14428
  • Update visibility setting for keras/tests to enable PIP package testing. by @copybara-service in #14429
  • Remove items from PIP_EXCLUDED_FILES which is needed with testing PIP. by @copybara-service in #14431
  • Split bins into num_bins and bin_boundaries arguments for discretization by @copybara-service in #14507
  • Update pbtxt to use _PRFER_OSS_KERAS=1. by @copybara-service in #14519
  • Sync OSS keras to head. by @copybara-service in #14572
  • Sync OSS keras to head. by @copybara-service in #14614
  • Cleanup the bazelrc and remove unrelated items to keras. by @copybara-service in #14616
  • Sync OSS keras to head. by @copybara-service in #14624
  • Remove object metadata when saving SavedModel. by @copybara-service in #14697
  • Fix static shape inference for Resizing layer. by @copybara-service in #14712
  • Make TextVectorization work with list input. by @copybara-service in #14711
  • Remove deprecated methods of Sequential model. by @copybara-service in #14714
  • Improve Model docstrings by @copybara-service in #14726
  • Add migration doc for legacy_tf_layers/core.py. by @copybara-service in #14740
  • PR #43417: Fixes #42872: map_to_outputs_names always returns a copy by @copybara-service in #14755
  • Rename the keras.py to keras_lib.py to resolve the name conflict during OSS test. by @copybara-service in #14778
  • Switch to tf.io.gfile for validating vocabulary files. by @copybara-service in #14788
  • Avoid serializing generated thresholds for AUC metrics. by @copybara-service in #14789
  • Fix data_utils.py when name ends with .tar.gz by @copybara-service in #14777
  • Fix lookup layer oov token check when num_oov_indices > len(vocabulary tokens) by @copybara-service in #14793
  • Update callbacks.py by @jvishnuvardhan in #14760
  • Fix keras metric.result_state when the metric variables are sharded variable. by @copybara-service in #14790
  • Fix typos in CONTRIBUTING.md by @amogh7joshi in #14642
  • Fixed ragged sample weights by @DavideWalder in #14804
  • Pin the protobuf version to 3.9.2 which is same as the TF. by @copybara-service in #14835
  • Make variable scope shim regularizer adding check for attribute presence instead of instance class by @copybara-service in #14837
  • Add missing license header for leakr check. by @copybara-service in #14840
  • Fix TextVectorization with output_sequence_length on unknown input shapes by @copybara-service in #14832
  • Add more explicit error message for instance type checking of optimizer. by @copybara-service in #14846
  • Set aggregation for variable when using PS Strategy for aggregating variables when running multi-gpu tests. by @copybara-service in #14845
  • Remove unnecessary reshape layer in MobileNet architecture by @copybara-service in #14854
  • Removes caching of the convolution tf.nn.convolution op. While this provided some performance benefits, it also produced some surprising behavior for users in eager mode. by @copybara-service in #14855
  • Output int64 by default from Discretization by @copybara-service in #14841
  • add patterns to .gitignore by @haifeng-jin in #14861
  • Clarify documentation of DepthwiseConv2D by @vinhill in #14817
  • add DepthwiseConv1D layer by @fsx950223 in #14863
  • Make model summary wrap by @Llamrei in #14865
  • Update the link in Estimator by @hirobf10 in #14901
  • Fix int given for float args by @SamuelMarks in #14900
  • Fix RNN, StackedRNNCells with nested state_size, output_size TypeError issues by @Ending2015a in #14905
  • Fix the use of imagenet_utils.preprocess_input within a Lambda layer with mixed precision by @anth2o in #14917
  • Fix docstrings in MultiHeadAttention layer call argument return_attention_scores. by @guillesanbri in #14920
  • Check if layer has _metrics_lock attribute by @DanBmh in #14903
  • Make keras.Model picklable by @adriangb in #14748
  • Fix typo in docs by @seanmor5 in #14946
  • use getter setter by @fsx950223 in #14948
  • Close _SESSION.session in clear_session by @sfreilich in #14414
  • Fix keras nightly PIP package build. by @copybara-service in #14957
  • Fix EarlyStopping stop at fisrt epoch when patience=0 ; add auc to au… by @DachuanZhao in #14750
  • Add default value in Attention layer docs #50839 by @europeanplaice in #14952
  • Update 00-bug-template.md by @jvishnuvardhan in #14991
  • minor fixing by @slowy07 in #15008
  • Remove cast of y_true to y_pred data type in sparse categorical cross entropy loss by @old-school-kid in #15015
  • Improve a number of error messages in Keras layers. by @copybara-service in #15031
  • Fix typo in the test case name. by @copybara-service in #15050
  • BackupAndRestore callback: Allow the train_counter to be fault tolerant across training interruptions. by @copybara-service in #15018
  • Include Keras API design guidelines in the contribution docs. by @copybara-service in #15051
  • Fix clone_model to consider input_tensors by @LeonhardFeiner in #14982
  • Fix docs about GRU and cuDNN. by @bzamecnik in #15058
  • fix: typo spelling by @slowy07 in #14989
  • update contributing guide by @haifeng-jin in #15063
  • Make python code in Sequential docs consistent by @01-vyom in #15075
  • Create stale.yml by @rthadur in #15176
  • Update Contributing Guide by @haifeng-jin in #15133
  • fix "rbg" -> "rgb" in image_dataset_from_directory by @YoniChechik in #15177
  • Do not initialize tables automatically for lookup layers by @copybara-service in #15193
  • Log the best epoch when restoring the best weights in early stopping by @harupy in #15197
  • update contributing guide by @haifeng-jin in #15189
  • fix array for numpy support by @collinzrj in #15201
  • Add test case for batch normalization with renorm on TPUs. by @copybara-service in #15234
  • Update training.py by @jvishnuvardhan in #15195
  • [Follow-up for #15197] Fix a log message for weight restoration in early stopping by @harupy in #15222
  • Rename references to "K" as "backend" for consistency. by @copybara-service in #15242
  • Add Keras utility for making user programs deterministic. by @copybara-service in #15243
  • Update convolutional.py by @ymodak in #15158
  • Fix for bug causing failing test keras/utils/vis_utils_test.py test_layer_range_value_fail second value (empty list). by @ddrakard in #15215
  • Fix small mistake … optimizer to optimizers by @MohamedAliRashad in #15227
  • Added dense_shape property delegation by @jackd in #15199
  • handle when a List is used as validation_data instead of a Tuple, in model.fit() by @tarun-bisht in #15237
  • fix cropping layer return empty list if crop is higher than data shape by @arubiales in #14970
  • Make the import in integration_test consistent with other keras code. by @copybara-service in #15275
  • Specify stacklevel for warnings.warn to make it easier to identify which lines throw warnings by @harupy in #15209
  • Fixes Github tensorflow/issues/51710 by @copybara-service in #15280
  • Print expanded nested layers feature in models.summary() by @krishrustagi in #15251
  • adapt EarlyStopping auto mode to auc by @DachuanZhao in #15200
  • change misleading description of m_mul by @mikael-epigram in #15288
  • The latest update.Hope it can suit you by @DLPerf in #15295
  • Fixed and added ValueError for plot_model, model_to_dot and model.summary() by @krishrustagi in #15318
  • Fix a minor typo in CIFAR-10 and CIFAR-100 description by @Rishit-dagli in #15321
  • Fix cropping2D empty list #15325 by @arubiales in #15326
  • convert label_smoothing dtype to y dtype by @FancyXun in #15363
  • Exclude default bazel paths for VsCode by @bhack in #15343
  • fix: typo in SeparableConv1D by @carmineds in #15370
  • Add user bin path for pip installed packages by @bhack in #15385
  • Fix reset_metrics by @bhack in #15342
  • Kernel_size, pool_size should be positive integers by @WingsBrokenAngel in #15356
  • Reopening #48000, #48491, #48610 from tensorflow/tensorflow by @AdityaKane2001 in #15315
  • expand_nested bug fix and changing model.summary style by @krishrustagi in #15355
  • unnecessary casting and condition removed by @cemsina in #15399
  • Bug fix for issue #15211, PR #15233 ("Plot model show activations") by @ddrakard in #15286
  • fix a bug: when use tf.keras.layers.TextVectorization layer to load model by @mikuh in #15422
  • added correct initialisation for MHA by @FabianGroeger96 in #15423
  • Persist attribute "sparse" of IndexLookup layer by @diggerk in #15473
  • Updated Normalization import by @sachinprasadhs in #15476
  • Added layer trainable information in model summary by @mfidabel in #15459
  • Add efficientnet v2 to keras.applications by @sebastian-sz in #14935
  • Fix typos by @kianmeng in #15543
  • Update the contribution guide to include a applications section by @mattdangerw in #15447
  • chore: replace os.path.join with tf.io.gfile.join by @adriangb in #15551
  • Fixing typos. by @MohamadJaber1 in #15626
  • Updating the documentation of MAPE by @sanatmpa1 in #15604
  • Fix typo by @europeanplaice in #15639
  • [Docs] Changed typo ModelNetV3 to MobileNetV3. by @sebastian-sz in #15640
  • Adding a choice not to make batches in timeseries_dataset_from_array by @europeanplaice in #15646
  • Bypass the require a config warning for marge layers by @leondgarse in #15612
  • Allow keras.utils.Sequence sub-classes to use sparse/ragged tensors by @karlhigley in #15264
  • fix backend _GRAPH_LEARNING_PHASES graph circular reference. by @zhjunqin in #15520
  • doc: add link to SciKeras migration guide by @adriangb in #15723
  • Fix a minor typo in the docstring by @kykim0 in #15683
  • Fixed docstrings in keras/optimizer_v2/learning_rate_schedule.py by @AdityaKane2001 in #15718
  • Implement compute_output_shape for BaseDenseAttention by @mishc9 in #15720
  • SidecarEvaluator: Graduate from experimental endpoint. by @copybara-service in #15788
  • Update README.md by @aliencaocao in #15814

New Contributors

Full Changelog: v2.7.0-rc0...v2.8.0-rc0

Check out latest releases or
releases around keras-team/keras v2.8.0-rc0

Don't miss a new keras release

NewReleases is sending notifications on new releases.

Get notifications