Create a Keras Layer
Count the total number of scalars composing the weights.
Weight constraints
Callback used to stream events to a server.
Retrieve tensors for layers with multiple nodes
Initializer that generates tensors initialized to a constant value.
IMDB Movie reviews sentiment classification
Downloads a file from a URL if it not already in the cache.
callback_reduce_lr_on_plateau
Reduce learning rate when a metric has stopped improving.
Boston housing price regression dataset
Reuters newswire topics classification
MNIST database of handwritten digits
3D array representation of images
LeCun normal initializer.
Keras implementation
Fashion-MNIST database of fashion articles
Update tokenizer internal vocabulary based on a list of texts or list of
sequences.
Initializer that generates the identity matrix.
Loads an image into PIL format.
Generates batches of augmented/normalized data from image data and labels
export_savedmodel.keras.engine.training.Model
Export a Saved Model
TensorBoard basic visualizations
Clone a model instance.
callback_terminate_on_naan
Callback that terminates training when a NaN loss is encountered.
evaluate.keras.engine.training.Model
Evaluate a Keras model
Bidirectional wrapper for RNNs.
Initializer that generates a random orthogonal matrix.
CIFAR10 small image classification
CIFAR100 small image classification
Configure a Keras model for training
Element-wise absolute value.
Retrieves a layer based on either its name (unique) or index.
Evaluates the model on a data generator.
Retrieve the next item from a generator
Layer/Model configuration
flow_images_from_directory
Generates batches of data from images in a directory (with optional
augmented/normalized data)
Freeze and unfreeze weights
Representation of HDF5 dataset to be used instead of an R array
Layer/Model weights as R arrays
Generate batches of image data with real-time data augmentation. The data will be
looped over (in batches).
initializer_random_normal
Initializer that generates tensors with a normal distribution.
Bitwise reduction (logical AND).
Creates a 1D tensor containing a sequence of integers.
initializer_lecun_uniform
LeCun uniform initializer.
Bitwise reduction (logical OR).
Train a Keras model
Initializer that generates tensors initialized to 1.
Install Keras and the TensorFlow backend
Adds a bias vector to a tensor.
Binary crossentropy between an output tensor and a target tensor.
Creates a constant tensor.
Fits the model on data yielded batch-by-batch by a generator.
Check if Keras is Available
Returns the index of the maximum value along an axis.
Fit image data generator internal statistics to some sample data.
1D convolution.
Cumulative sum of the values in a tensor, alongside the specified axis.
imagenet_decode_predictions
Decodes the prediction of an ImageNet model.
Applies batch normalization on x given mean, var, beta and gamma.
imagenet_preprocess_input
Preprocesses a tensor or array encoding a batch of images.
He normal initializer.
Sets the values of many tensor variables at once.
Returns the index of the minimum value along an axis.
He uniform variance scaling initializer.
2D deconvolution (i.e. transposed convolution).
k_categorical_crossentropy
Categorical crossentropy between an output tensor and a target tensor.
2D convolution.
Destroys the current TF graph and creates a new one.
Depthwise 2D convolution with separable filters.
initializer_variance_scaling
Initializer capable of adapting its scale to the shape of weights.
Computes cos of x element-wise.
3D convolution.
Initializer that generates tensors initialized to 0.
3D deconvolution (i.e. transposed convolution).
Returns the static number of elements in a Keras variable or tensor.
Exponential linear unit.
Returns the dtype of a Keras tensor or variable, as a string.
Adds a 1-sized dimension at index axis
.
Turn a nD tensor into a 2D tensor with same 1st dimension.
Instantiate an identity matrix and returns it.
TF session to be used by the backend.
initializer_glorot_normal
Glorot normal initializer, also called Xavier normal initializer.
Get the uid for the default graph.
Casts a tensor to a different dtype and returns it.
Returns the value of more than one tensor variable.
Instantiates a Keras function
Returns the shape of tensor or variable as a list of int or NULL entries.
Selects x
in train phase, and alt
otherwise.
Multiplies 2 tensors (and/or variables) and returns a tensor .
Sets entries in x
to zero at random, while scaling the entire tensor.
Returns whether a tensor is a sparse tensor.
Retrieves the elements of indices indices
in the tensor reference
.
Cast an array to the default Keras float type.
Normalizes a tensor wrt the L2 norm alongside the specified axis.
Reduce elems using fn to combine them from left to right.
Returns the gradients of variables
w.r.t. loss
.
Reduce elems using fn to combine them from right to left.
Element-wise maximum of two tensors.
initializer_glorot_uniform
Glorot uniform initializer, also called Xavier uniform initializer.
Mean of a tensor, alongside the specified axis.
Returns a tensor with the same content as the input tensor.
k_ctc_label_dense_to_sparse
Converts CTC labels from dense to sparse.
Default image data format convention ('channels_first' or 'channels_last').
k_normalize_batch_in_training
Computes mean and std for batch then apply batch_normalization on batch.
Returns whether x
is a Keras tensor.
Element-wise inequality between two tensors.
Cumulative product of the values in a tensor, alongside the specified axis.
Element-wise truth value of (x > y).
Repeats a 2D tensor.
Fuzz factor used in numeric expressions.
initializer_random_uniform
Initializer that generates tensors with a uniform distribution.
initializer_truncated_normal
Initializer that generates a truncated normal distribution.
Batchwise dot product.
Active Keras backend
Returns whether x
is a placeholder.
Element-wise equality between two tensors.
Compute the moving average of a variable.
Returns the value of a variable.
Element-wise value clipping.
Concatenates a list of tensors alongside the specified axis.
Returns the learning phase flag.
Runs CTC loss algorithm on each batch element.
Returns the shape of a variable.
Repeats the elements of a tensor along an axis.
Decodes the output of a softmax.
Evaluates the value of a variable.
Reverse a tensor along the specified axes.
Element-wise exponential.
Element-wise truth value of (x < y).
Flatten a tensor.
Selects x
in test phase, and alt
otherwise.
Returns the number of axes in a tensor, as an integer.
Instantiates an all-ones variable of the same shape as another tensor.
Default float type
Returns whether the targets
are in the top k
predictions
.
Apply 2D conv with un-shared weights.
Permutes axes in a tensor.
Minimum value in a tensor.
Element-wise log.
Iterates over the time dimension of a tensor
Computes log(sum(exp(elements across dimensions of a tensor))).
Returns a tensor with random binomial distribution of values.
Element-wise minimum of two tensors.
k_manual_variable_initialization
Sets the manual variable initialization flag.
Returns a tensor with normal distribution of values.
Returns the symbolic shape of a tensor or variable.
Element-wise sigmoid.
Element-wise square root.
Sets the learning phase to a fixed value.
Sets the value of a variable, from an R array.
3D Pooling.
Element-wise square.
Element-wise sign.
Transposes a tensor and returns it.
Computes sin of x element-wise.
Computes the one-hot representation of an integer tensor.
Instantiates an all-ones tensor variable and returns it.
Element-wise exponentiation.
Instantiates a variable with values drawn from a normal distribution.
Returns a tensor with uniform distribution of values.
Element-wise truth value of (x >= y).
Segment-wise linear approximation of sigmoid.
k_random_uniform_variable
Instantiates a variable with values drawn from a uniform distribution.
Standard deviation of a tensor, alongside the specified axis.
Element-wise truth value of (x <= y).
Apply 1D conv with un-shared weights.
Returns variables
but with zero gradient w.r.t. every other variable.
layer_activation_parametric_relu
Parametric Rectified Linear Unit.
Returns a tensor with truncated random normal distribution of values.
Map the function fn over the elements elems and return the outputs.
Creates a tensor by tiling x
by n
.
Reset graph identifiers.
Softmax activation function.
Instantiates a placeholder tensor and returns it.
Maximum value in a tensor.
Converts a sparse tensor into a dense tensor and returns it.
Keras Model composed of a linear stack of layers
Layer that adds a list of inputs.
Applies Alpha Dropout to the input.
Apply an activation function to an output.
Transposed 3D convolution layer (sometimes called Deconvolution).
layer_activation_thresholded_relu
Thresholded Rectified Linear Unit.
Cropping layer for 3D data (e.g. spatial or spatio-temporal).
Convolutional LSTM.
layer_activity_regularization
Layer that applies an update to the cost function based input activity.
2D Pooling.
Depthwise separable 2D convolution.
Resizes the images contained in a 4D tensor.
Prints message
and the tensor value when evaluated.
Resizes the volume contained in a 5D tensor.
Rectified linear unit.
Fast GRU implementation backed by CuDNN . Reshapes a tensor to the specified shape.
Applies Dropout to the input.
Turns positive integers (indexes) into dense vectors of fixed size.
Input layer
Softmax of a tensor.
Layer that computes a dot product between samples in two tensors.
Wraps arbitrary expression as a layer
Multiplies the values in a tensor, alongside the specified axis.
Pads the 2nd and 3rd dimensions of a 4D tensor.
Pads 5D tensor with zeros along the depth, height, width dimensions.
Permute the dimensions of an input according to a given pattern
layer_global_max_pooling_3d
Global Max pooling operation for 3D data.
Stacks a list of rank R
tensors into a rank R+1
tensor.
Upsampling layer for 1D inputs.
Removes a 1-dimension from the tensor at index axis
.
Instantiates a variable and returns it.
Repeats the input n times.
Element-wise rounding to the closest integer.
2D convolution with separable filters.
Gated Recurrent Unit - Cho et al.
Softplus of a tensor.
Sum of the values in a tensor, alongside the specified axis.
Softsign of a tensor.
Max pooling operation for temporal data.
Upsampling layer for 2D inputs.
Instantiates an all-zeros variable and returns it.
Switches between two operations depending on a scalar value.
Model performance metrics
Update the value of x
to new_x
.
Update the value of x
by adding increment
.
Layer that averages a list of inputs.
Instantiates an all-zeros variable of the same shape as another tensor.
R interface to Keras
k_sparse_categorical_crossentropy
Categorical crossentropy with integer targets.
Element-wise tanh.
Max pooling operation for spatial data.
Model loss functions
Separable 2D convolution.
Fully-connected RNN where the output is to be fed back to input.
Nesterov Adam optimizer
Generates a word rank-based probabilistic sampling table.
Model configuration as JSON
Average pooling for temporal data.
1D convolution layer (e.g. temporal convolution).
2D convolution layer (e.g. spatial convolution over images).
Fast LSTM implementation backed by CuDNN . layer_global_max_pooling_1d
Global max pooling operation for temporal data.
Add a densely-connected NN layer to an output
RMSProp optimizer
layer_global_max_pooling_2d
Global max pooling operation for spatial data.
Max pooling operation for 3D data (spatial or spatio-temporal).
Keras array object
Layer that computes the maximum (element-wise) a list of inputs.
Generates predictions for the input samples from a data generator.
Average pooling operation for 3D data (spatial or spatio-temporal).
Average pooling operation for spatial data.
Pads the middle dimension of a 3D tensor.
Update the value of x
by subtracting decrement
.
Keras Model
Remove the last layer in a model
Adadelta optimizer.
Returns predictions for a single batch of samples.
predict.keras.engine.training.Model
Generate predictions from a Keras model
Adagrad optimizer.
Variance of a tensor, alongside the specified axis.
Cropping layer for 1D input (e.g. temporal sequence).
Cropping layer for 2D input (e.g. picture).
summary.keras.engine.training.Model
Print a summary of a Keras model
One-hot encode a text into a list of word indexes in a vocabulary of size n.
Converts a text to a sequence of indexes in a fixed-size hashing space.
Spatial 3D version of Dropout.
Apply a layer to every temporal slice of an input.
Layer that subtracts two inputs.
Exponential Linear Unit.
Utility function for generating batches of temporal data.
Zero-padding layer for 2D input (e.g. picture).
Zero-padding layer for 3D data (spatial or spatio-temporal).
Convert text to a sequence of words (or tokens).
Replicates a model on different GPUs.
Normalize a matrix or nd-array
L1 and L2 regularization
Save/Load models using HDF5 files
Reset the states for a layer
Save/Load model weights using HDF5 files
Converts a class vector (integers) to binary class matrix.
Apply additive zero-centered Gaussian noise.
Single gradient update or model evaluation over one batch of samples.
layer_global_average_pooling_1d
Global average pooling operation for temporal data.
layer_activation_leaky_relu
Leaky version of a Rectified Linear Unit.
layer_global_average_pooling_2d
Global average pooling operation for spatial data.
layer_global_average_pooling_3d
Global Average pooling operation for 3D data.
Long Short-Term Memory unit - Hochreiter 1997.
layer_batch_normalization
Batch normalization layer (Ioffe and Szegedy, 2014).
Masks a sequence by using a mask value to skip timesteps.
Reshapes an output to a certain shape.
Depthwise separable 1D convolution.
Layer that concatenates a list of inputs.
3D convolution layer (e.g. spatial convolution over volumes).
Transposed 2D convolution layer (sometimes called Deconvolution).
Transform each text in texts in a sequence of integers.
Upsampling layer for 3D inputs.
texts_to_sequences_generator
Transforms each text in texts in a sequence of integers.
Flattens an input
Apply multiplicative 1-centered Gaussian noise.
Stochastic gradient descent optimizer
Zero-padding layer for 1D input (e.g. temporal sequence).
Pads sequences to the same length
Generates probability or class probability predictions for the input samples.
layer_locally_connected_1d
Locally-connected layer for 1D inputs.
layer_locally_connected_2d
Locally-connected layer for 2D inputs.
Objects exported from other packages
Serialize a model to an R object
Select a Keras implementation and backend
Generates skipgram word pairs.
Layer that computes the minimum (element-wise) a list of inputs.
Provide a scope with mappings of names to custom objects
Layer that multiplies (element-wise) a list of inputs.
Spatial 1D version of Dropout.
Spatial 2D version of Dropout.
Model configuration as YAML
Assign values to names
Adam optimizer
Adamax optimizer
plot.keras_training_history
Plot training history
Save a text tokenizer to an external file
Convert a list of sequences into a matrix.
Pipe operator
Text tokenization utility
Convert a list of texts to a matrix.
Instantiates the DenseNet architecture.
application_inception_resnet_v2
Inception-ResNet v2 model, with weights trained on ImageNet
Base R6 class for Keras constraints
Activation functions
ResNet50 model for Keras.
Instantiates a NASNet model.
Base R6 class for Keras callbacks
Inception V3 model, with weights pre-trained on ImageNet.
MobileNet model architecture.
callback_model_checkpoint
Save the model after every epoch.
Callback that streams epoch results to a csv file
Stop training when a monitored quantity has stopped improving.
Create a custom callback
Base R6 class for Keras layers
Keras backend tensor engine
VGG16 and VGG19 models for Keras.
Callback that prints metrics to stdout.
Xception V1 model for Keras.
callback_learning_rate_scheduler
Learning rate scheduler.