keras API reference
400 public APIs from keras (keras-team/keras) — 183 classes, 199 functions, 18 methods. Signatures extracted by static analysis of the actual source.
Repository: keras-team/keras
| Kind | Count |
|---|---|
| Classes | 183 |
| Functions | 199 |
| Methods | 18 |
API list
class
guides.writing_your_own_callbacks.EarlyStoppingAtMinLossStop training when the loss is at its min, i.e.
func
keras.src.activations.activations.elu(x, alpha=1.0)Exponential Linear Unit.
func
keras.src.activations.activations.exponential(x)Exponential activation function.
func
keras.src.activations.activations.hard_shrink(x, threshold=0.5)Hard Shrink activation function.
func
keras.src.activations.activations.hard_sigmoid(x)Hard sigmoid activation function.
func
keras.src.activations.activations.hard_tanh(x)HardTanh activation function.
func
keras.src.activations.activations.leaky_relu(x, negative_slope=0.2)Leaky relu activation function.
func
keras.src.activations.activations.log_softmax(x, axis=-1)Log-Softmax activation function.
func
keras.src.activations.activations.mish(x)Mish activation function.
func
keras.src.activations.activations.relu6(x)Relu6 activation function.
func
keras.src.activations.activations.selu(x)Scaled Exponential Linear Unit (SELU).
func
keras.src.activations.activations.sigmoid(x)Sigmoid activation function.
func
keras.src.activations.activations.silu(x)Swish (or Silu) activation function.
func
keras.src.activations.activations.soft_shrink(x, threshold=0.5)Soft Shrink activation function.
func
keras.src.activations.activations.softplus(x)Softplus activation function.
func
keras.src.activations.activations.softsign(x)Softsign activation function.
func
keras.src.activations.activations.sparse_plus(x)SparsePlus activation function.
func
keras.src.activations.activations.sparse_sigmoid(x)Sparse sigmoid activation function.
func
keras.src.activations.activations.sparsemax(x, axis=-1)Sparsemax activation function.
func
keras.src.activations.activations.squareplus(x, b=4)Squareplus activation function.
func
keras.src.activations.activations.tanh(x)Hyperbolic tangent activation function.
func
keras.src.activations.activations.tanh_shrink(x)Tanh shrink activation function.
func
keras.src.activations.activations.threshold(x, threshold, default_value)Threshold activation function.
func
keras.src.activations.get(identifier)Retrieve a Keras activation function via an identifier.
func
keras.src.applications.convnext.ConvNeXtBlock(projection_dim, drop_path_rate=0.0, layer_scale_init_value=1e-06, name=None)ConvNeXt block.
class
keras.src.applications.convnext.LayerScaleLayer scale module.
class
keras.src.applications.convnext.StochasticDepthStochastic Depth module.
func
keras.src.applications.densenet.conv_block(x, growth_rate, name)A building block for a dense block.
func
keras.src.applications.densenet.dense_block(x, blocks, name)A dense block.
func
keras.src.applications.densenet.transition_block(x, reduction, name)A transition block.
class
keras.src.backend.common.keras_tensor.KerasTensorSymbolic tensor -- encapsulates a shape and a dtype.
func
keras.src.backend.common.keras_tensor.is_keras_tensor(x)Returns whether `x` is a Keras tensor.
func
keras.src.backend.common.masking.get_keras_mask(x)Gets the Keras mask attribute from the given tensor.
class
keras.src.backend.common.name_scope.name_scopeCreates a sub-namespace for variable paths.
class
keras.src.backend.common.remat.RematScopeA context manager for enabling rematerialization in Keras.
class
keras.src.backend.common.stateless_scope.StatelessScopeScope to prevent any update to Keras Variables.
class
keras.src.backend.common.symbolic_scope.SymbolicScopeScope to indicate the symbolic stage.
class
keras.src.backend.common.variables.VariableRepresents a backend-agnostic variable in Keras.
method
keras.src.backend.common.variables.Variable.dtype()The data type of the variable.
method
keras.src.backend.common.variables.Variable.name()The name of the variable.
method
keras.src.backend.common.variables.Variable.shape()The shape of the variable.
func
keras.src.backend.config.disable_flash_attention()Disable flash attention.
func
keras.src.backend.config.enable_flash_attention()Enable flash attention.
func
keras.src.backend.config.floatx()Return the default float type, as a string.
func
keras.src.backend.config.image_data_format()Return the default image data format convention.
func
keras.src.backend.config.set_floatx(value)Set the default float dtype.
func
keras.src.backend.jax.core.remat(f)Implementation of rematerialization.
class
keras.src.backend.jax.export.JaxExportArchiveJAX backend implementation of SavedModel export archive.
func
keras.src.backend.jax.nn.unfold(input, kernel_size, dilation=1, padding=0, stride=1)JAX implementation of Unfold.
class
keras.src.backend.numpy.core.custom_gradientDecorator for custom gradients.
class
keras.src.backend.openvino.core.custom_gradientDecorator for custom gradients.
func
keras.src.backend.tensorflow.core.remat(f)Implementation of rematerialization.
func
keras.src.backend.tensorflow.core.shape(x)Always return a tuple shape.
func
keras.src.backend.tensorflow.trainer.concat(tensors, axis=0)Concats `tensor`s along `axis`.
class
keras.src.backend.torch.core.custom_gradientDecorator for custom gradients.
func
keras.src.backend.torch.core.remat(f)Implementation of rematerialization.
class
keras.src.callbacks.backup_and_restore.BackupAndRestoreCallback to back up and restore the training state.
class
keras.src.callbacks.callback.CallbackBase class used to build new callbacks.
method
keras.src.callbacks.callback.Callback.on_epoch_end(epoch, logs=None)Called at the end of an epoch.
method
keras.src.callbacks.callback.Callback.on_predict_end(logs=None)Called at the end of prediction.
method
keras.src.callbacks.callback.Callback.on_train_end(logs=None)Called at the end of training.
class
keras.src.callbacks.callback_list.CallbackListContainer abstracting a list of callbacks.
class
keras.src.callbacks.csv_logger.CSVLoggerCallback that streams epoch results to a CSV file.
class
keras.src.callbacks.history.HistoryCallback that records events into a `History` object.
class
keras.src.callbacks.learning_rate_scheduler.LearningRateSchedulerLearning rate scheduler.
class
keras.src.callbacks.progbar_logger.ProgbarLoggerCallback that prints metrics to stdout.
class
keras.src.callbacks.remote_monitor.RemoteMonitorCallback used to stream events to a server.
class
keras.src.callbacks.tensorboard.TensorBoardEnable visualizations for TensorBoard.
class
keras.src.constraints.constraints.ConstraintBase class for weight constraints.
class
keras.src.constraints.constraints.MaxNormMaxNorm weight constraint.
class
keras.src.constraints.constraints.MinMaxNormMinMaxNorm weight constraint.
class
keras.src.constraints.constraints.NonNegConstrains the weights to be non-negative.
func
keras.src.datasets.cifar10.load_data()Loads the CIFAR10 dataset.
func
keras.src.datasets.cifar100.load_data(label_mode='fine')Loads the CIFAR100 dataset.
func
keras.src.datasets.fashion_mnist.load_data()Loads the Fashion-MNIST dataset.
func
keras.src.datasets.mnist.load_data(path='mnist.npz')Loads the MNIST dataset.
class
keras.src.distillation.distillation_loss.DistillationLossBase class for distillation loss computation.
class
keras.src.distillation.distillation_loss.FeatureDistillationFeature distillation loss.
class
keras.src.distribution.distribution_lib.DataParallelDistribution for data parallelism.
class
keras.src.distribution.distribution_lib.DistributionBase class for variable distribution strategies.
method
keras.src.distribution.distribution_lib.Distribution.num_data_shards()Total number of data shards.
method
keras.src.distribution.distribution_lib.Distribution.num_model_replicas()Number of model replicas.
class
keras.src.distribution.distribution_lib.ModelParallelDistribution that shards model variables.
class
keras.src.distribution.distribution_lib.TensorLayoutA layout to apply to a tensor.
class
keras.src.dtype_policies.dtype_policy.AWQDTypePolicyQuantized dtype policy for AWQ quantization.
class
keras.src.dtype_policies.dtype_policy.DTypePolicyA dtype policy for a Keras layer.
method
keras.src.dtype_policies.dtype_policy.DTypePolicy.name()Returns the name of this policy.
class
keras.src.dtype_policies.dtype_policy.GPTQDTypePolicyQuantized dtype policy for GPTQ quantization.
class
keras.src.dtype_policies.dtype_policy.Int4DTypePolicyQuantized dtype policy for int4 quantization.
class
keras.src.dtype_policies.dtype_policy.TernaryDTypePolicyQuantized dtype policy for ternary quantization.
func
keras.src.dtype_policies.serialize(dtype_policy)Serializes `DTypePolicy` instance.
class
keras.src.export.litert.LiteRTExporterExporter for the LiteRT (TFLite) format.
class
keras.src.export.saved_model_export_archive.SavedModelExportArchiveBase class for SavedModel export archive.
class
keras.src.initializers.constant_initializers.ConstantInitializer that generates tensors with constant values.
class
keras.src.initializers.constant_initializers.IdentityInitializer that generates the identity matrix.
class
keras.src.initializers.constant_initializers.OnesInitializer that generates tensors initialized to 1.
class
keras.src.initializers.constant_initializers.ZerosInitializer that generates tensors initialized to 0.
class
keras.src.initializers.initializer.InitializerInitializer base class: all Keras initializers inherit from this class.
class
keras.src.initializers.random_initializers.HeNormalHe normal initializer.
class
keras.src.initializers.random_initializers.HeUniformHe uniform variance scaling initializer.
class
keras.src.initializers.random_initializers.LecunNormalLecun normal initializer.
class
keras.src.initializers.random_initializers.LecunUniformLecun uniform initializer.
class
keras.src.initializers.random_initializers.OrthogonalInitializer that generates an orthogonal matrix.
class
keras.src.initializers.random_initializers.RandomNormalRandom normal initializer.
class
keras.src.initializers.random_initializers.RandomUniformRandom uniform initializer.
class
keras.src.layers.activations.activation.ActivationApplies an activation function to an output.
class
keras.src.layers.activations.elu.ELUApplies an Exponential Linear Unit function to an output.
class
keras.src.layers.activations.prelu.PReLUParametric Rectified Linear Unit activation layer.
class
keras.src.layers.activations.relu.ReLURectified Linear Unit activation function layer.
class
keras.src.layers.activations.softmax.SoftmaxSoftmax activation layer.
class
keras.src.layers.attention.additive_attention.AdditiveAttentionAdditive attention layer, a.k.a.
class
keras.src.layers.attention.attention.AttentionDot-product attention layer, a.k.a.
class
keras.src.layers.attention.grouped_query_attention.GroupedQueryAttentionGrouped Query Attention layer.
class
keras.src.layers.attention.multi_head_attention.MultiHeadAttentionMultiHeadAttention layer.
class
keras.src.layers.convolutional.base_conv_transpose.BaseConvTransposeAbstract N-D transposed convolution layer.
class
keras.src.layers.convolutional.base_depthwise_conv.BaseDepthwiseConvAbstract N-D depthwise convolution layer.
class
keras.src.layers.convolutional.base_separable_conv.BaseSeparableConvAbstract base layer for separable convolution.
class
keras.src.layers.convolutional.conv1d.Conv1D1D convolution layer (e.g.
class
keras.src.layers.convolutional.conv1d_transpose.Conv1DTranspose1D transposed convolution layer.
class
keras.src.layers.convolutional.conv2d.Conv2D2D convolution layer.
class
keras.src.layers.convolutional.conv2d_transpose.Conv2DTranspose2D transposed convolution layer.
class
keras.src.layers.convolutional.conv3d.Conv3D3D convolution layer.
class
keras.src.layers.convolutional.conv3d_transpose.Conv3DTranspose3D transposed convolution layer.
class
keras.src.layers.convolutional.depthwise_conv1d.DepthwiseConv1D1D depthwise convolution layer.
class
keras.src.layers.convolutional.depthwise_conv2d.DepthwiseConv2D2D depthwise convolution layer.
class
keras.src.layers.convolutional.separable_conv1d.SeparableConv1D1D separable convolution layer.
class
keras.src.layers.convolutional.separable_conv2d.SeparableConv2D2D separable convolution layer.
class
keras.src.layers.core.dense.DenseJust your regular densely-connected NN layer.
class
keras.src.layers.core.einsum_dense.EinsumDenseA layer that uses `einsum` as the backing computation.
class
keras.src.layers.core.identity.IdentityIdentity layer.
class
keras.src.layers.core.lambda_layer.LambdaWraps arbitrary expressions as a `Layer` object.
class
keras.src.layers.core.masking.MaskingMasks a sequence by using a mask value to skip timesteps.
class
keras.src.layers.core.wrapper.WrapperAbstract wrapper base class.
class
keras.src.layers.input_spec.InputSpecSpecifies the rank, dtype and shape of every input to a layer.
class
keras.src.layers.layer.LayerThis is the class from which all layers inherit.
method
keras.src.layers.layer.Layer.dtype()Alias of `layer.variable_dtype`.
method
keras.src.layers.layer.Layer.load_own_variables(store)Loads the state of the layer.
method
keras.src.layers.layer.Layer.metrics()List of all metrics.
method
keras.src.layers.layer.Layer.metrics_variables()List of all metric variables.
method
keras.src.layers.layer.Layer.path()The path of the layer.
method
keras.src.layers.layer.Layer.save_own_variables(store)Saves the state of the layer.
method
keras.src.layers.layer.Layer.trainable_variables()List of all trainable layer state.
class
keras.src.layers.merging.add.AddPerforms elementwise addition operation.
class
keras.src.layers.merging.average.AverageAverages a list of inputs element-wise..
class
keras.src.layers.merging.base_merge.MergeGeneric merge layer for elementwise merge functions.
class
keras.src.layers.merging.concatenate.ConcatenateConcatenates a list of inputs.
class
keras.src.layers.merging.dot.DotComputes element-wise dot product of two tensors.
func
keras.src.layers.merging.dot.batch_dot(x, y, axes=None)Batchwise dot product.
class
keras.src.layers.merging.maximum.MaximumComputes element-wise maximum on a list of inputs.
class
keras.src.layers.merging.minimum.MinimumComputes elementwise minimum on a list of inputs.
class
keras.src.layers.merging.multiply.MultiplyPerforms elementwise multiplication.
class
keras.src.layers.merging.subtract.SubtractPerforms elementwise subtraction.
class
keras.src.layers.normalization.batch_normalization.BatchNormalizationLayer that normalizes its inputs.
class
keras.src.layers.normalization.group_normalization.GroupNormalizationGroup normalization layer.
class
keras.src.layers.normalization.layer_normalization.LayerNormalizationLayer normalization layer (Ba et al., 2016).
class
keras.src.layers.normalization.rms_normalization.RMSNormalizationRoot Mean Square (RMS) Normalization layer.
class
keras.src.layers.normalization.unit_normalization.UnitNormalizationUnit normalization layer.
class
keras.src.layers.pooling.average_pooling1d.AveragePooling1DAverage pooling for temporal data.
class
keras.src.layers.pooling.average_pooling2d.AveragePooling2DAverage pooling operation for 2D spatial data.
class
keras.src.layers.pooling.base_adaptive_pooling.BaseAdaptivePoolingBase class shared by all adaptive pooling layers.
class
keras.src.layers.pooling.base_global_pooling.BaseGlobalPoolingBase global pooling layer.
class
keras.src.layers.pooling.base_pooling.BasePoolingBase pooling layer.
class
keras.src.layers.pooling.global_average_pooling2d.GlobalAveragePooling2DGlobal average pooling operation for 2D data.
class
keras.src.layers.pooling.global_average_pooling3d.GlobalAveragePooling3DGlobal average pooling operation for 3D data.
class
keras.src.layers.pooling.global_max_pooling1d.GlobalMaxPooling1DGlobal max pooling operation for temporal data.
class
keras.src.layers.pooling.global_max_pooling2d.GlobalMaxPooling2DGlobal max pooling operation for 2D data.
class
keras.src.layers.pooling.global_max_pooling3d.GlobalMaxPooling3DGlobal max pooling operation for 3D data.
class
keras.src.layers.pooling.max_pooling1d.MaxPooling1DMax pooling operation for 1D temporal data.
class
keras.src.layers.pooling.max_pooling2d.MaxPooling2DMax pooling operation for 2D spatial data.
class
keras.src.layers.preprocessing.image_preprocessing.aug_mix.AugMixPerforms the AugMix data augmentation technique.
class
keras.src.layers.preprocessing.image_preprocessing.center_crop.CenterCropA preprocessing layer which crops images.
class
keras.src.layers.preprocessing.image_preprocessing.cut_mix.CutMixCutMix data augmentation technique.
class
keras.src.layers.preprocessing.image_preprocessing.random_erasing.RandomErasingRandom Erasing data augmentation technique.
class
keras.src.layers.preprocessing.image_preprocessing.random_hue.RandomHueRandomly adjusts the hue on given images.
class
keras.src.layers.preprocessing.image_preprocessing.resizing.ResizingA preprocessing layer which resizes images.
class
keras.src.layers.preprocessing.index_lookup.IndexLookupMaps values from a vocabulary to integer indices.
class
keras.src.layers.preprocessing.pipeline.PipelineApplies a series of layers to an input.
class
keras.src.layers.regularization.alpha_dropout.AlphaDropoutApplies Alpha Dropout to the input.
class
keras.src.layers.regularization.dropout.DropoutApplies dropout to the input.
class
keras.src.layers.regularization.gaussian_dropout.GaussianDropoutApply multiplicative 1-centered Gaussian noise.
class
keras.src.layers.regularization.gaussian_noise.GaussianNoiseApply additive zero-centered Gaussian noise.
class
keras.src.layers.regularization.spatial_dropout.SpatialDropout1DSpatial 1D version of Dropout.
class
keras.src.layers.regularization.spatial_dropout.SpatialDropout2DSpatial 2D version of Dropout.
class
keras.src.layers.regularization.spatial_dropout.SpatialDropout3DSpatial 3D version of Dropout.
class
keras.src.layers.reshaping.cropping1d.Cropping1DCropping layer for 1D input (e.g.
class
keras.src.layers.reshaping.cropping2d.Cropping2DCropping layer for 2D input (e.g.
class
keras.src.layers.reshaping.cropping3d.Cropping3DCropping layer for 3D data (e.g.
class
keras.src.layers.reshaping.flatten.FlattenFlattens the input.
class
keras.src.layers.reshaping.repeat_vector.RepeatVectorRepeats the input n times.
class
keras.src.layers.reshaping.reshape.ReshapeLayer that reshapes inputs into the given shape.
class
keras.src.layers.reshaping.up_sampling1d.UpSampling1DUpsampling layer for 1D inputs.
class
keras.src.layers.reshaping.up_sampling2d.UpSampling2DUpsampling layer for 2D inputs.
class
keras.src.layers.reshaping.up_sampling3d.UpSampling3DUpsampling layer for 3D inputs.
class
keras.src.layers.reshaping.zero_padding1d.ZeroPadding1DZero-padding layer for 1D input (e.g.
class
keras.src.layers.reshaping.zero_padding2d.ZeroPadding2DZero-padding layer for 2D input (e.g.
class
keras.src.layers.rnn.bidirectional.BidirectionalBidirectional wrapper for RNNs.
class
keras.src.layers.rnn.conv_lstm.ConvLSTMCellCell class for the ConvLSTM layer.
class
keras.src.layers.rnn.conv_lstm1d.ConvLSTM1D1D Convolutional LSTM.
class
keras.src.layers.rnn.conv_lstm2d.ConvLSTM2D2D Convolutional LSTM.
class
keras.src.layers.rnn.conv_lstm3d.ConvLSTM3D3D Convolutional LSTM.
class
keras.src.layers.rnn.gru.GRUGated Recurrent Unit - Cho et al.
class
keras.src.layers.rnn.gru.GRUCellCell class for the GRU layer.
class
keras.src.layers.rnn.lstm.LSTMLong Short-Term Memory layer - Hochreiter 1997.
class
keras.src.layers.rnn.lstm.LSTMCellCell class for the LSTM layer.
class
keras.src.layers.rnn.rnn.RNNBase class for recurrent layers.
class
keras.src.layers.rnn.simple_rnn.SimpleRNNCellCell class for SimpleRNN.
func
keras.src.legacy.backend.abs(x)DEPRECATED.
func
keras.src.legacy.backend.all(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.any(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.arange(start, stop=None, step=1, dtype='int32')DEPRECATED.
func
keras.src.legacy.backend.argmax(x, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.argmin(x, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.batch_dot(x, y, axes=None)DEPRECATED.
func
keras.src.legacy.backend.batch_flatten(x)DEPRECATED.
func
keras.src.legacy.backend.batch_get_value(tensors)DEPRECATED.
func
keras.src.legacy.backend.batch_normalization(x, mean, var, beta, gamma, axis=-1, epsilon=0.001)DEPRECATED.
func
keras.src.legacy.backend.batch_set_value(tuples)DEPRECATED.
func
keras.src.legacy.backend.bias_add(x, bias, data_format=None)DEPRECATED.
func
keras.src.legacy.backend.binary_crossentropy(target, output, from_logits=False)DEPRECATED.
func
keras.src.legacy.backend.binary_focal_crossentropy(target, output, apply_class_balancing=False, alpha=0.25, gamma=2.0, from_logits=False)DEPRECATED.
func
keras.src.legacy.backend.cast(x, dtype)DEPRECATED.
func
keras.src.legacy.backend.cast_to_floatx(x)DEPRECATED.
func
keras.src.legacy.backend.categorical_crossentropy(target, output, from_logits=False, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.categorical_focal_crossentropy(target, output, alpha=0.25, gamma=2.0, from_logits=False, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.clip(x, min_value, max_value)DEPRECATED.
func
keras.src.legacy.backend.concatenate(tensors, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.constant(value, dtype=None, shape=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.conv1d(x, kernel, strides=1, padding='valid', data_format=None, dilation_rate=1)DEPRECATED.
func
keras.src.legacy.backend.conv2d(x, kernel, strides=(1, 1), padding='valid', data_format=None, dilation_rate=(1, 1))DEPRECATED.
func
keras.src.legacy.backend.conv2d_transpose(x, kernel, output_shape, strides=(1, 1), padding='valid', data_format=None, dilation_rate=(1, 1))DEPRECATED.
func
keras.src.legacy.backend.conv3d(x, kernel, strides=(1, 1, 1), padding='valid', data_format=None, dilation_rate=(1, 1, 1))DEPRECATED.
func
keras.src.legacy.backend.cos(x)DEPRECATED.
func
keras.src.legacy.backend.count_params(x)DEPRECATED.
func
keras.src.legacy.backend.ctc_batch_cost(y_true, y_pred, input_length, label_length)DEPRECATED.
func
keras.src.legacy.backend.ctc_decode(y_pred, input_length, greedy=True, beam_width=100, top_paths=1)DEPRECATED.
func
keras.src.legacy.backend.ctc_label_dense_to_sparse(labels, label_lengths)DEPRECATED.
func
keras.src.legacy.backend.cumprod(x, axis=0)DEPRECATED.
func
keras.src.legacy.backend.cumsum(x, axis=0)DEPRECATED.
func
keras.src.legacy.backend.depthwise_conv2d(x, depthwise_kernel, strides=(1, 1), padding='valid', data_format=None, dilation_rate=(1, 1))DEPRECATED.
func
keras.src.legacy.backend.dot(x, y)DEPRECATED.
func
keras.src.legacy.backend.dropout(x, level, noise_shape=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.dtype(x)DEPRECATED.
func
keras.src.legacy.backend.elu(x, alpha=1.0)DEPRECATED.
func
keras.src.legacy.backend.equal(x, y)DEPRECATED.
func
keras.src.legacy.backend.eval(x)DEPRECATED.
func
keras.src.legacy.backend.exp(x)DEPRECATED.
func
keras.src.legacy.backend.expand_dims(x, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.eye(size, dtype=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.flatten(x)DEPRECATED.
func
keras.src.legacy.backend.foldl(fn, elems, initializer=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.foldr(fn, elems, initializer=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.gather(reference, indices)DEPRECATED.
func
keras.src.legacy.backend.get_value(x)DEPRECATED.
func
keras.src.legacy.backend.gradients(loss, variables)DEPRECATED.
func
keras.src.legacy.backend.greater(x, y)DEPRECATED.
func
keras.src.legacy.backend.greater_equal(x, y)DEPRECATED.
func
keras.src.legacy.backend.hard_sigmoid(x)DEPRECATED.
func
keras.src.legacy.backend.in_top_k(predictions, targets, k)DEPRECATED.
func
keras.src.legacy.backend.int_shape(x)DEPRECATED.
func
keras.src.legacy.backend.is_sparse(tensor)DEPRECATED.
func
keras.src.legacy.backend.l2_normalize(x, axis=None)DEPRECATED.
func
keras.src.legacy.backend.less(x, y)DEPRECATED.
func
keras.src.legacy.backend.less_equal(x, y)DEPRECATED.
func
keras.src.legacy.backend.log(x)DEPRECATED.
func
keras.src.legacy.backend.map_fn(fn, elems, name=None, dtype=None)DEPRECATED.
func
keras.src.legacy.backend.max(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.maximum(x, y)DEPRECATED.
func
keras.src.legacy.backend.mean(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.min(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.minimum(x, y)DEPRECATED.
func
keras.src.legacy.backend.moving_average_update(x, value, momentum)DEPRECATED.
func
keras.src.legacy.backend.name_scope(name)DEPRECATED.
func
keras.src.legacy.backend.ndim(x)DEPRECATED.
func
keras.src.legacy.backend.not_equal(x, y)DEPRECATED.
func
keras.src.legacy.backend.one_hot(indices, num_classes)DEPRECATED.
func
keras.src.legacy.backend.ones(shape, dtype=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.ones_like(x, dtype=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.permute_dimensions(x, pattern)DEPRECATED.
func
keras.src.legacy.backend.pool2d(x, pool_size, strides=(1, 1), padding='valid', data_format=None, pool_mode='max')DEPRECATED.
func
keras.src.legacy.backend.pool3d(x, pool_size, strides=(1, 1, 1), padding='valid', data_format=None, pool_mode='max')DEPRECATED.
func
keras.src.legacy.backend.pow(x, a)DEPRECATED.
func
keras.src.legacy.backend.prod(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.random_bernoulli(shape, p=0.0, dtype=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.random_normal(shape, mean=0.0, stddev=1.0, dtype=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.random_normal_variable(shape, mean, scale, dtype=None, name=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.random_uniform(shape, minval=0.0, maxval=1.0, dtype=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.random_uniform_variable(shape, low, high, dtype=None, name=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.relu(x, alpha=0.0, max_value=None, threshold=0.0)DEPRECATED.
func
keras.src.legacy.backend.repeat(x, n)DEPRECATED.
func
keras.src.legacy.backend.repeat_elements(x, rep, axis)DEPRECATED.
func
keras.src.legacy.backend.reshape(x, shape)DEPRECATED.
func
keras.src.legacy.backend.resize_images(x, height_factor, width_factor, data_format, interpolation='nearest')DEPRECATED.
func
keras.src.legacy.backend.resize_volumes(x, depth_factor, height_factor, width_factor, data_format)DEPRECATED.
func
keras.src.legacy.backend.reverse(x, axes)DEPRECATED.
func
keras.src.legacy.backend.round(x)DEPRECATED.
func
keras.src.legacy.backend.separable_conv2d(x, depthwise_kernel, pointwise_kernel, strides=(1, 1), padding='valid', data_format=None, dilation_rate=(1, 1))DEPRECATED.
func
keras.src.legacy.backend.set_value(x, value)DEPRECATED.
func
keras.src.legacy.backend.shape(x)DEPRECATED.
func
keras.src.legacy.backend.sigmoid(x)DEPRECATED.
func
keras.src.legacy.backend.sign(x)DEPRECATED.
func
keras.src.legacy.backend.sin(x)DEPRECATED.
func
keras.src.legacy.backend.softmax(x, axis=-1)DEPRECATED.
func
keras.src.legacy.backend.softplus(x)DEPRECATED.
func
keras.src.legacy.backend.softsign(x)DEPRECATED.
func
keras.src.legacy.backend.sparse_categorical_crossentropy(target, output, from_logits=False, axis=-1, ignore_class=None)DEPRECATED.
func
keras.src.legacy.backend.spatial_2d_padding(x, padding=((1, 1), (1, 1)), data_format=None)DEPRECATED.
func
keras.src.legacy.backend.spatial_3d_padding(x, padding=((1, 1), (1, 1), (1, 1)), data_format=None)DEPRECATED.
func
keras.src.legacy.backend.sqrt(x)DEPRECATED.
func
keras.src.legacy.backend.square(x)DEPRECATED.
func
keras.src.legacy.backend.squeeze(x, axis)DEPRECATED.
func
keras.src.legacy.backend.stack(x, axis=0)DEPRECATED.
func
keras.src.legacy.backend.std(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.stop_gradient(variables)DEPRECATED.
func
keras.src.legacy.backend.sum(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.switch(condition, then_expression, else_expression)DEPRECATED.
func
keras.src.legacy.backend.tanh(x)DEPRECATED.
func
keras.src.legacy.backend.temporal_padding(x, padding=(1, 1))DEPRECATED.
func
keras.src.legacy.backend.tile(x, n)DEPRECATED.
func
keras.src.legacy.backend.to_dense(tensor)DEPRECATED.
func
keras.src.legacy.backend.transpose(x)DEPRECATED.
func
keras.src.legacy.backend.truncated_normal(shape, mean=0.0, stddev=1.0, dtype=None, seed=None)DEPRECATED.
func
keras.src.legacy.backend.update(x, new_x)DEPRECATED.
func
keras.src.legacy.backend.update_add(x, increment)DEPRECATED.
func
keras.src.legacy.backend.update_sub(x, decrement)DEPRECATED.
func
keras.src.legacy.backend.var(x, axis=None, keepdims=False)DEPRECATED.
func
keras.src.legacy.backend.variable(value, dtype=None, name=None, constraint=None)DEPRECATED.
func
keras.src.legacy.backend.zeros(shape, dtype=None, name=None)DEPRECATED.
func
keras.src.legacy.backend.zeros_like(x, dtype=None, name=None)DEPRECATED.
class
keras.src.legacy.layers.AlphaDropoutDEPRECATED.
class
keras.src.legacy.layers.RandomHeightDEPRECATED.
class
keras.src.legacy.layers.RandomWidthDEPRECATED.
class
keras.src.legacy.layers.ThresholdedReLUDEPRECATED.
class
keras.src.legacy.preprocessing.image.ImageDataGeneratorDEPRECATED.
class
keras.src.legacy.preprocessing.image.IteratorBase class for image data iterators.
class
keras.src.legacy.preprocessing.image.NumpyArrayIteratorIterator yielding data from a Numpy array.
func
keras.src.legacy.preprocessing.image.apply_channel_shift(x, intensity, channel_axis=0)Performs a channel shift.
func
keras.src.legacy.preprocessing.image.random_rotation(x, rg, row_axis=1, col_axis=2, channel_axis=0, fill_mode='nearest', cval=0.0, interpolation_order=1)DEPRECATED.
func
keras.src.legacy.preprocessing.image.random_shear(x, intensity, row_axis=1, col_axis=2, channel_axis=0, fill_mode='nearest', cval=0.0, interpolation_order=1)DEPRECATED.
func
keras.src.legacy.preprocessing.image.random_shift(x, wrg, hrg, row_axis=1, col_axis=2, channel_axis=0, fill_mode='nearest', cval=0.0, interpolation_order=1)DEPRECATED.
func
keras.src.legacy.preprocessing.image.random_zoom(x, zoom_range, row_axis=1, col_axis=2, channel_axis=0, fill_mode='nearest', cval=0.0, interpolation_order=1)DEPRECATED.
class
keras.src.legacy.preprocessing.text.TokenizerDEPRECATED.
func
keras.src.legacy.preprocessing.text.hashing_trick(text, n, hash_function=None, filters='!"#$%&()*+,-./:;<=>?@[\\]^_`{|}~\t\n', lower=True, split=' ', analyzer=None)DEPRECATED.
func
keras.src.legacy.preprocessing.text.one_hot(input_text, n, filters='!"#$%&()*+,-./:;<=>?@[\\]^_`{|}~\t\n', lower=True, split=' ', analyzer=None)DEPRECATED.
func
keras.src.legacy.preprocessing.text.text_to_word_sequence(input_text, filters='!"#$%&()*+,-./:;<=>?@[\\]^_`{|}~\t\n', lower=True, split=' ')DEPRECATED.
func
keras.src.legacy.preprocessing.text.tokenizer_from_json(json_string)DEPRECATED.
class
keras.src.legacy.saving.serialization.NoopLoadingScopeThe default shared object loading scope.
class
keras.src.losses.loss.LossLoss base class.
class
keras.src.losses.losses.CTCCTC (Connectionist Temporal Classification) loss.
class
keras.src.losses.losses.CategoricalFocalCrossentropyComputes the alpha balanced focal crossentropy loss.
class
keras.src.losses.losses.DiceComputes the Dice loss value between `y_true` and `y_pred`.
class
keras.src.losses.losses.HingeComputes the hinge loss between `y_true` & `y_pred`.
class
keras.src.losses.losses.HuberComputes the Huber loss between `y_true` & `y_pred`.
class
keras.src.losses.losses.PoissonComputes the Poisson loss between `y_true` & `y_pred`.
class
keras.src.losses.losses.SquaredHingeComputes the squared hinge loss between `y_true` & `y_pred`.
class
keras.src.losses.losses.TverskyComputes the Tversky loss value between `y_true` and `y_pred`.
func
keras.src.losses.losses.huber(y_true, y_pred, delta=1.0)Computes Huber loss value.
class
keras.src.metrics.accuracy_metrics.AccuracyCalculates how often predictions equal labels.
class
keras.src.metrics.accuracy_metrics.BinaryAccuracyCalculates how often predictions match binary labels.
class
keras.src.metrics.confusion_metrics.FalseNegativesCalculates the number of false negatives.
class
keras.src.metrics.confusion_metrics.FalsePositivesCalculates the number of false positives.
class
keras.src.metrics.confusion_metrics.TrueNegativesCalculates the number of true negatives.
class
keras.src.metrics.confusion_metrics.TruePositivesCalculates the number of true positives.
class
keras.src.metrics.correlation_metrics.PearsonCorrelationCalculates the Pearson Correlation Coefficient (PCC).
class
keras.src.metrics.f_score_metrics.F1ScoreComputes F-1 Score.
class
keras.src.metrics.f_score_metrics.FBetaScoreComputes F-Beta score.
class
keras.src.metrics.hinge_metrics.HingeComputes the hinge metric between `y_true` and `y_pred`.
class
keras.src.metrics.hinge_metrics.SquaredHingeComputes the hinge metric between `y_true` and `y_pred`.
class
keras.src.metrics.iou_metrics.MeanIoUComputes the mean Intersection-Over-Union metric.
class
keras.src.metrics.metric.MetricEncapsulates metric logic and state.
method
keras.src.metrics.metric.Metric.result()Compute the current metric value.
class
keras.src.metrics.metrics_utils.AUCCurveType of AUC Curve (ROC or PR).
class
keras.src.metrics.metrics_utils.AUCSummationMethodType of AUC summation method.
class
keras.src.metrics.reduction_metrics.MeanCompute the (weighted) mean of the given values.
class
keras.src.metrics.reduction_metrics.MeanMetricWrapperWrap a stateless metric function with the `Mean` metric.
class
keras.src.metrics.reduction_metrics.SumCompute the (weighted) sum of the given values.
class
keras.src.metrics.regression_metrics.R2ScoreComputes R2 score.
method
keras.src.models.sequential.Sequential.pop(rebuild=True)Removes the last layer in the model.
func
keras.src.ops.core.cast(x, dtype)Cast a tensor to the desired dtype.
func
keras.src.ops.core.convert_to_numpy(x)Convert a tensor to a NumPy array.
func
keras.src.ops.core.fori_loop(lower, upper, body_fun, init_val)For loop implementation.
func
keras.src.ops.core.map(f, xs)Map a function over leading array axes.
func
keras.src.ops.core.shape(x)Gets the shape of the tensor input.
func
keras.src.ops.core.slice(inputs, start_indices, shape)Return a slice of an input tensor.
func
keras.src.ops.core.stop_gradient(variable)Stops gradient computation.
func
keras.src.ops.core.while_loop(cond, body, loop_vars, maximum_iterations=None)While loop implementation.
func
keras.src.ops.image.hsv_to_rgb(images, data_format=None)Convert HSV images to RGB.
func
keras.src.ops.image.rgb_to_grayscale(images, data_format=None)Convert RGB images to grayscale.
func
keras.src.ops.image.rgb_to_hsv(images, data_format=None)Convert RGB images to HSV.
func
keras.src.ops.linalg.inv(x)Computes the inverse of a square tensor.
func
keras.src.ops.linalg.norm(x, ord=None, axis=None, keepdims=False)Matrix or vector norm.
class
keras.src.ops.nn.AdaptiveAveragePoolAdaptive average pooling operation.
class
keras.src.ops.nn.AdaptiveMaxPoolAdaptive max pooling operation.
func
keras.src.ops.nn.adaptive_max_pool(inputs, output_size, data_format=None)Adaptive max pooling operation.
func
keras.src.ops.nn.average_pool(inputs, pool_size, strides=None, padding='valid', data_format=None)Average pooling operation.
func
keras.src.ops.nn.conv(inputs, kernel, strides=1, padding='valid', data_format=None, dilation_rate=1)General N-D convolution.
func
keras.src.ops.nn.hard_shrink(x, threshold=0.5)Hard Shrink activation function.
func
keras.src.ops.nn.hard_sigmoid(x)Hard sigmoid activation function.
func
keras.src.ops.nn.log_softmax(x, axis=-1)Log-softmax activation function.
func
keras.src.ops.nn.max_pool(inputs, pool_size, strides=None, padding='valid', data_format=None)Max pooling operation.
func
keras.src.ops.nn.relu(x)Rectified linear unit activation function.
About this data
These signatures were extracted from the public source of keras-team/keras
using Python's ast module. Argument names, default values,
type annotations and return types are taken verbatim from the code.
Implementation bodies are never stored. See
how it works for details.