Learn R Programming

⚠️There's a newer version (2.16.0) of this package.Take me there.

keras (version 2.6.0)

R Interface to 'Keras'

Description

Interface to 'Keras' , a high-level neural networks 'API'. 'Keras' was developed with a focus on enabling fast experimentation, supports both convolution based networks and recurrent networks (as well as combinations of the two), and runs seamlessly on both 'CPU' and 'GPU' devices.

Copy Link

Version

Install

install.packages('keras')

Monthly Downloads

17,691

Version

2.6.0

License

MIT + file LICENSE

Maintainer

Tomasz Kalinowski

Last Published

August 21st, 2021

Functions in keras (2.6.0)

(Deprecated) Base R6 class for Keras wrappers

(Deprecated) Base R6 class for Keras callbacks

activation_relu

Activation functions

Create a custom Layer

Fits the state of the preprocessing layer to the data being passed.

KerasConstraint

(Deprecated) Base R6 class for Keras constraints

(Deprecated) Base R6 class for Keras layers

application_densenet

Instantiates the DenseNet architecture.

application_inception_resnet_v2

Inception-ResNet v2 model, with weights trained on ImageNet

application_inception_v3

Inception V3 model, with weights pre-trained on ImageNet.

application_mobilenet

MobileNet model architecture.

application_nasnet

Instantiates a NASNet model.

application_mobilenet_v2

MobileNetV2 model architecture

callback_remote_monitor

Callback used to stream events to a server.

application_resnet50

ResNet50 model for Keras.

application_vgg

VGG16 and VGG19 models for Keras.

callback_tensorboard

TensorBoard basic visualizations

Clone a model instance.

callback_terminate_on_naan

Callback that terminates training when a NaN loss is encountered.

Bidirectional wrapper for RNNs.

callback_early_stopping

Stop training when a monitored quantity has stopped improving.

callback_lambda

Create a custom callback

callback_csv_logger

Callback that streams epoch results to a csv file

callback_learning_rate_scheduler

Learning rate scheduler.

dataset_cifar10

CIFAR10 small image classification

Custom metric function

callback_model_checkpoint

Save the model after every epoch.

dataset_boston_housing

Boston housing price regression dataset

fit.keras.engine.training.Model

Train a Keras model

export_savedmodel.keras.engine.training.Model

Export a Saved Model

Count the total number of scalars composing the weights.

Create a Keras Layer

dataset_cifar100

CIFAR100 small image classification

application_xception

Xception V1 model for Keras.

compile.keras.engine.training.Model

Configure a Keras model for training

Weight constraints

MNIST database of handwritten digits

Fits the model on data yielded batch-by-batch by a generator.

dataset_reuters

Reuters newswire topics classification

Keras backend tensor engine

create_layer_wrapper

Create a Keras Layer wrapper

(Deprecated) Create a Keras Wrapper

evaluate.keras.engine.training.Model

Evaluate a Keras model

fit_text_tokenizer

Update tokenizer internal vocabulary based on a list of texts or list of sequences.

Loads an image into PIL format.

evaluate_generator

Evaluates the model on a data generator.

initializer_he_uniform

He uniform variance scaling initializer.

dataset_fashion_mnist

Fashion-MNIST database of fashion articles

Layer/Model configuration

Representation of HDF5 dataset to be used instead of an R array

IMDB Movie reviews sentiment classification

Freeze and unfreeze weights

callback_progbar_logger

Callback that prints metrics to stdout.

callback_reduce_lr_on_plateau

Reduce learning rate when a metric has stopped improving.

Downloads a file from a URL if it not already in the cache.

Active Keras backend

flow_images_from_dataframe

Takes the dataframe and the path to a directory and generates batches of augmented/normalized data.

flow_images_from_directory

Generates batches of data from images in a directory (with optional augmented/normalized data)

Retrieves a layer based on either its name (unique) or index.

Retrieve tensors for layers with multiple nodes

Layer/Model weights as R arrays

image_data_generator

Generate batches of image data with real-time data augmentation. The data will be looped over (in batches).

initializer_ones

Initializer that generates tensors initialized to 1.

Get the vocabulary for text vectorization layers

initializer_lecun_uniform

LeCun uniform initializer.

initializer_variance_scaling

Initializer capable of adapting its scale to the shape of weights.

Bitwise reduction (logical OR).

2D convolution.

Element-wise value clipping.

initializer_glorot_normal

Glorot normal initializer, also called Xavier normal initializer.

Creates a 1D tensor containing a sequence of integers.

initializer_glorot_uniform

Glorot uniform initializer, also called Xavier uniform initializer.

flow_images_from_data

Generates batches of augmented/normalized data from image data and labels

Concatenates a list of tensors alongside the specified axis.

Element-wise exponential.

Retrieve the next item from a generator

Make a python class constructor

Batchwise dot product.

Element-wise absolute value.

Casts a tensor to a different dtype and returns it.

Bitwise reduction (logical AND).

initializer_he_normal

He normal initializer.

initializer_zeros

Initializer that generates tensors initialized to 0.

Fuzz factor used in numeric expressions.

3D array representation of images

initializer_orthogonal

Initializer that generates a random orthogonal matrix.

Returns the value of a variable.

fit_image_data_generator

Fit image data generator internal statistics to some sample data.

Returns the index of the minimum value along an axis.

Returns a tensor with the same content as the input tensor.

Returns the shape of tensor or variable as a list of int or NULL entries.

Element-wise equality between two tensors.

Returns the index of the maximum value along an axis.

initializer_random_normal

Initializer that generates tensors with a normal distribution.

k_ctc_label_dense_to_sparse

Converts CTC labels from dense to sparse.

Instantiates a Keras function

Returns the dtype of a Keras tensor or variable, as a string.

k_conv2d_transpose

2D deconvolution (i.e. transposed convolution).

Cumulative sum of the values in a tensor, alongside the specified axis.

Creates a constant tensor.

Flatten a tensor.

k_moving_average_update

Compute the moving average of a variable.

1D convolution.

3D convolution.

k_conv3d_transpose

3D deconvolution (i.e. transposed convolution).

k_image_data_format

Default image data format convention ('channels_first' or 'channels_last').

Apply 2D conv with un-shared weights.

k_is_keras_tensor

Returns whether x is a Keras tensor.

imagenet_decode_predictions

Decodes the prediction of an ImageNet model.

imagenet_preprocess_input

Preprocesses a tensor or array encoding a batch of images.

Keras implementation

image_dataset_from_directory

Create a dataset from a directory

Retrieves the elements of indices indices in the tensor reference.

Returns the gradients of variables w.r.t. loss.

Element-wise inequality between two tensors.

Computes sin of x element-wise.

Exponential linear unit.

Default float type

Minimum value in a tensor.

initializer_identity

Initializer that generates the identity matrix.

k_batch_normalization

Applies batch normalization on x given mean, var, beta and gamma.

k_cast_to_floatx

Cast an array to the default Keras float type.

initializer_constant

Initializer that generates tensors initialized to a constant value.

initializer_random_uniform

Initializer that generates tensors with a uniform distribution.

Decodes the output of a softmax.

Returns the number of axes in a tensor, as an integer.

Segment-wise linear approximation of sigmoid.

k_in_train_phase

Selects x in train phase, and alt otherwise.

k_get_variable_shape

Returns the shape of a variable.

Apply 1D conv with un-shared weights.

k_is_placeholder

Returns whether x is a placeholder.

Computes cos of x element-wise.

Returns whether the targets are in the top k predictions.

Returns the static number of elements in a Keras variable or tensor.

Creates a tensor by tiling x by n.

Normalizes a tensor wrt the L2 norm alongside the specified axis.

Computes log(sum(exp(elements across dimensions of a tensor))).

k_depthwise_conv2d

Depthwise 2D convolution with separable filters.

TF session to be used by the backend.

Element-wise truth value of (x > y).

Softmax of a tensor.

Returns the symbolic shape of a tensor or variable.

Mean of a tensor, alongside the specified axis.

k_resize_images

Resizes the images contained in a 4D tensor.

Get the uid for the default graph.

initializer_lecun_normal

LeCun normal initializer.

Install TensorFlow and Keras, including all python dependencies

Instantiates a placeholder tensor and returns it.

k_permute_dimensions

Permutes axes in a tensor.

k_spatial_2d_padding

Pads the 2nd and 3rd dimensions of a 4D tensor.

Converts a sparse tensor into a dense tensor and returns it.

R interface to Keras

Element-wise log.

k_in_test_phase

Selects x in test phase, and alt otherwise.

Returns whether a tensor is a sparse tensor.

k_batch_set_value

Sets the values of many tensor variables at once.

Instantiates an all-zeros variable of the same shape as another tensor.

Sets the value of a variable, from an R array.

Reshapes a tensor to the specified shape.

Element-wise minimum of two tensors.

k_stop_gradient

Returns variables but with zero gradient w.r.t. every other variable.

Element-wise sign.

Multiplies the values in a tensor, alongside the specified axis.

k_learning_phase

Returns the learning phase flag.

Switches between two operations depending on a scalar value.

layer_cudnn_gru

Fast GRU implementation backed by CuDNN.

Element-wise rounding to the closest integer.

k_random_binomial

Returns a tensor with random binomial distribution of values.

Layer that adds a list of inputs.

layer_batch_normalization

Batch normalization layer (Ioffe and Szegedy, 2014).

Iterates over the time dimension of a tensor

Element-wise sigmoid.

layer_cudnn_lstm

Fast LSTM implementation backed by CuDNN.

Sum of the values in a tensor, alongside the specified axis.

Instantiates an all-zeros variable and returns it.

layer_activity_regularization

Layer that applies an update to the cost function based input activity.

initializer_truncated_normal

Initializer that generates a truncated normal distribution.

layer_concatenate

Layer that concatenates a list of inputs.

k_ctc_batch_cost

Runs CTC loss algorithm on each batch element.

k_batch_flatten

Turn a nD tensor into a 2D tensor with same 1st dimension.

k_batch_get_value

Returns the value of more than one tensor variable.

k_repeat_elements

Repeats the elements of a tensor along an axis.

Evaluates the value of a variable.

k_categorical_crossentropy

Categorical crossentropy between an output tensor and a target tensor.

k_clear_session

Destroys the current TF graph and creates a new one.

k_random_normal_variable

Instantiates a variable with values drawn from a normal distribution.

k_random_normal

Returns a tensor with normal distribution of values.

Element-wise exponentiation.

Softsign of a tensor.

k_truncated_normal

Returns a tensor with truncated random normal distribution of values.

Reduce elems using fn to combine them from left to right.

Keras array object

Computes the one-hot representation of an integer tensor.

Returns whether x is a symbolic tensor.

layer_conv_lstm_2d

Convolutional LSTM.

layer_activation_softmax

Softmax activation function.

k_resize_volumes

Resizes the volume contained in a 5D tensor.

k_manual_variable_initialization

Sets the manual variable initialization flag.

k_normalize_batch_in_training

Computes mean and std for batch then apply batch_normalization on batch.

k_greater_equal

Element-wise truth value of (x >= y).

Element-wise truth value of (x < y).

Reduce elems using fn to combine them from right to left.

layer_activation_thresholded_relu

Thresholded Rectified Linear Unit.

layer_gaussian_dropout

Apply multiplicative 1-centered Gaussian noise.

Multiplies 2 tensors (and/or variables) and returns a tensor.

is_keras_available

Check if Keras is Available

Reverse a tensor along the specified axes.

k_spatial_3d_padding

Pads 5D tensor with zeros along the depth, height, width dimensions.

Layer that averages a list of inputs.

Element-wise truth value of (x <= y).

k_sparse_categorical_crossentropy

Categorical crossentropy with integer targets.

layer_activation_selu

Scaled Exponential Linear Unit.

Instantiates a variable and returns it.

k_temporal_padding

Pads the middle dimension of a 3D tensor.

Element-wise square root.

Element-wise tanh.

layer_max_pooling_2d

Max pooling operation for spatial data.

layer_alpha_dropout

Applies Alpha Dropout to the input.

layer_repeat_vector

Repeats the input n times.

Main Keras module

layer_global_average_pooling_2d

Global average pooling operation for spatial data.

Applies Dropout to the input.

Variance of a tensor, alongside the specified axis.

layer_max_pooling_1d

Max pooling operation for temporal data.

Map the function fn over the elements elems and return the outputs.

Stacks a list of rank R tensors into a rank R+1 tensor.

layer_average_pooling_2d

Average pooling operation for spatial data.

Transposes a tensor and returns it.

layer_conv_1d_transpose

Transposed 1D convolution layer (sometimes called Deconvolution).

layer_activation

Apply an activation function to an output.

layer_conv_2d_transpose

Transposed 2D convolution layer (sometimes called Deconvolution).

layer_activation_parametric_relu

Parametric Rectified Linear Unit.

keras_model_sequential

Keras Model composed of a linear stack of layers

Sets entries in x to zero at random, while scaling the entire tensor.

k_binary_crossentropy

Binary crossentropy between an output tensor and a target tensor.

Adds a bias vector to a tensor.

layer_activation_leaky_relu

Leaky version of a Rectified Linear Unit.

Softplus of a tensor.

Standard deviation of a tensor, alongside the specified axis.

Prints message and the tensor value when evaluated.

Reset graph identifiers.

layer_average_pooling_3d

Average pooling operation for 3D data (spatial or spatio-temporal).

keras_model_custom

Create a Keras custom model

k_random_uniform

Returns a tensor with uniform distribution of values.

layer_depthwise_conv_2d

Depthwise separable 2D convolution.

2D convolution layer (e.g. spatial convolution over images).

Computes the hinge metric between y_true and y_pred

layer_average_pooling_1d

Average pooling for temporal data.

layer_spatial_dropout_3d

Spatial 3D version of Dropout.

Masks a sequence by using a mask value to skip timesteps.

layer_activation_relu

Rectified Linear Unit activation function

Layer that computes a dot product between samples in two tensors.

Layer that computes the maximum (element-wise) a list of inputs.

metric_categorical_crossentropy

Computes the crossentropy metric between the labels and predictions

layer_layer_normalization

Layer normalization layer (Ba et al., 2016).

layer_global_max_pooling_3d

Global Max pooling operation for 3D data.

Layer that computes the minimum (element-wise) a list of inputs.

metric_sensitivity_at_specificity

Computes best sensitivity where specificity is >= specified value

Permute the dimensions of an input according to a given pattern

metric_cosine_proximity

(Deprecated) metric_cosine_proximity

Gated Recurrent Unit - Cho et al.

1D convolution layer (e.g. temporal convolution).

layer_activation_elu

Exponential Linear Unit.

layer_locally_connected_1d

Locally-connected layer for 1D inputs.

layer_cropping_1d

Cropping layer for 1D input (e.g. temporal sequence).

layer_attention

Creates attention layer

Wraps arbitrary expression as a layer

metric_mean_relative_error

Computes the mean relative error by normalizing with the given values

metric_categorical_hinge

Computes the categorical hinge metric between y_true and y_pred

3D convolution layer (e.g. spatial convolution over volumes).

layer_gaussian_noise

Apply additive zero-centered Gaussian noise.

layer_spatial_dropout_1d

Spatial 1D version of Dropout.

layer_text_vectorization

Text vectorization layer

layer_global_average_pooling_1d

Global average pooling operation for temporal data.

layer_conv_3d_transpose

Transposed 3D convolution layer (sometimes called Deconvolution).

Update the value of x to new_x.

layer_embedding

Turns positive integers (indexes) into dense vectors of fixed size.

model_to_saved_model

Export to Saved Model format

layer_zero_padding_3d

Zero-padding layer for 3D data (spatial or spatio-temporal).

metric_mean_wrapper

Wraps a stateless metric function with the Mean metric

Assign values to names

layer_spatial_dropout_2d

Spatial 2D version of Dropout.

metric_binary_accuracy

Calculates how often predictions match binary labels

Adds a 1-sized dimension at index axis.

Cumulative product of the values in a tensor, alongside the specified axis.

Maximum value in a tensor.

Layer that subtracts two inputs.

Computes the recall of the predictions with respect to the labels

metric_categorical_accuracy

Calculates how often predictions match one-hot labels

Element-wise maximum of two tensors.

plot.keras_training_history

Plot training history

layer_upsampling_1d

Upsampling layer for 1D inputs.

Approximates the AUC (Area under the curve) of the ROC or PR curves

Long Short-Term Memory unit - Hochreiter 1997.

Add a densely-connected NN layer to an output

layer_cropping_2d

Cropping layer for 2D input (e.g. picture).

layer_upsampling_3d

Upsampling layer for 3D inputs.

layer_cropping_3d

Cropping layer for 3D data (e.g. spatial or spatio-temporal).

Instantiate an identity matrix and returns it.

Instantiates an all-ones variable of the same shape as another tensor.

metric-or-Metric

metric-or-Metric

optimizer_nadam

Nesterov Adam optimizer

layer_global_average_pooling_3d

Global Average pooling operation for 3D data.

layer_dense_features

Constructs a DenseFeatures.

metric_false_negatives

Calculates the number of false negatives

layer_locally_connected_2d

Locally-connected layer for 2D inputs.

metric_true_negatives

Calculates the number of true negatives

metric_true_positives

Calculates the number of true positives

metric_kullback_leibler_divergence

Computes Kullback-Leibler divergence

Instantiates an all-ones tensor variable and returns it.

Reshapes an output to a certain shape.

Computes the (weighted) mean of the given values

metric_recall_at_precision

Computes best recall where precision is >= specified value

Computes the (weighted) sum of the given values

layer_upsampling_2d

Upsampling layer for 2D inputs.

metric_false_positives

Calculates the number of false positives

Pads sequences to the same length

Rectified linear unit.

Repeats a 2D tensor.

k_random_uniform_variable

Instantiates a variable with values drawn from a uniform distribution.

metric_mean_squared_error

Computes the mean squared error between labels and predictions

metric_precision_at_recall

Computes best precision where recall is >= specified value

k_separable_conv2d

2D convolution with separable filters.

Flattens an input

Update the value of x by adding increment.

metric_binary_crossentropy

Computes the crossentropy metric between the labels and predictions

metric_mean_tensor

Computes the element-wise (weighted) mean of the given tensors

metric_sparse_categorical_accuracy

Calculates how often predictions match integer labels

Normalize a matrix or nd-array

timeseries_generator

Utility function for generating batches of temporal data.

metric_accuracy

Calculates how often predictions equal labels

metric_logcosh_error

Computes the logarithm of the hyperbolic cosine of the prediction error

Element-wise square.

layer_global_max_pooling_1d

Global max pooling operation for temporal data.

k_set_learning_phase

Sets the learning phase to a fixed value.

Layer that multiplies (element-wise) a list of inputs.

Removes a 1-dimension from the tensor at index axis.

Update the value of x by subtracting decrement.

metric_sparse_categorical_crossentropy

Computes the crossentropy metric between the labels and predictions

layer_separable_conv_2d

Separable 2D convolution.

metric_root_mean_squared_error

Computes root mean squared error metric between y_true and y_pred

(Deprecated) Generates probability or class probability predictions for the input samples.

optimizer_adadelta

Adadelta optimizer.

multi_gpu_model

Replicates a model on different GPUs.

Remove the last layer in a model

Model configuration as JSON

model_from_saved_model

Load a Keras model from the Saved Model format

text_to_word_sequence

Convert text to a sequence of words (or tokens).

metric_top_k_categorical_accuracy

Computes how often targets are in the top K predictions

L1 and L2 regularization

Text tokenization utility

optimizer_adamax

Adamax optimizer

Objects exported from other packages

save_model_weights_tf

Save model weights in the SavedModel format

sequences_to_matrix

Convert a list of sequences into a matrix.

sequential_model_input_layer

sequential_model_input_layer

save_text_tokenizer

Save a text tokenizer to an external file

text_hashing_trick

Converts a text to a sequence of indexes in a fixed-size hashing space.

One-hot encode a text into a list of word indexes in a vocabulary of size n.

layer_max_pooling_3d

Max pooling operation for 3D data (spatial or spatio-temporal).

metric_precision

Computes the precision of the predictions with respect to the labels

Computes the Poisson metric between y_true and y_pred

loss_cosine_proximity

(Deprecated) loss_cosine_proximity

metric_cosine_similarity

Computes the cosine similarity between the labels and predictions

layer_simple_rnn

Fully-connected RNN where the output is to be fed back to input.

Converts a class vector (integers) to binary class matrix.

make_sampling_table

Generates a word rank-based probabilistic sampling table.

metric_mean_squared_logarithmic_error

Computes the mean squared logarithmic error

with_custom_object_scope

Provide a scope with mappings of names to custom objects

texts_to_matrix

Convert a list of texts to a matrix.

predict_on_batch

Returns predictions for a single batch of samples.

optimizer_adagrad

Adagrad optimizer.

Reset the states for a layer

save_model_hdf5

Save/Load models using HDF5 files

metric_specificity_at_sensitivity

Computes best specificity where sensitivity is >= specified value

metric_mean_iou

Computes the mean Intersection-Over-Union metric

optimizer_rmsprop

RMSProp optimizer

metric_squared_hinge

Computes the squared hinge metric

Model configuration as YAML

texts_to_sequences

Transform each text in texts in a sequence of integers.

use_implementation

Select a Keras implementation and backend

Single gradient update or model evaluation over one batch of samples.

layer_zero_padding_1d

Zero-padding layer for 1D input (e.g. temporal sequence).

layer_global_max_pooling_2d

Global max pooling operation for spatial data.

Stochastic gradient descent optimizer

Generates skipgram word pairs.

summary.keras.engine.training.Model

Print a summary of a Keras model

layer_multi_head_attention

MultiHeadAttention layer

Save/Load models using SavedModel format

layer_separable_conv_1d

Depthwise separable 1D convolution.

predict_generator

Generates predictions for the input samples from a data generator.

metric_mean_absolute_error

Computes the mean absolute error between the labels and predictions

layer_zero_padding_2d

Zero-padding layer for 2D input (e.g. picture).

metric_mean_absolute_percentage_error

Computes the mean absolute percentage error between y_true and y_pred

Sets vocabulary (and optionally document frequency) data for the layer

metric_sparse_top_k_categorical_accuracy

Computes how often integer targets are in the top K predictions

predict.keras.engine.training.Model

Generate predictions from a Keras model

serialize_model

Serialize a model to an R object

texts_to_sequences_generator

Transforms each text in texts in a sequence of integers.

save_model_weights_hdf5

Save/Load model weights using HDF5 files

time_distributed

Apply a layer to every temporal slice of an input.