Extract a "tar.gz" file into a local folder. @param inputPath Input file path. @param outputPath Output directory path. @throws IOException IO error.

A minimal character set, with a-z, A-Z, 0-9 and common punctuation etc

A minimal character set, with a-z, A-Z, 0-9 and common punctuation etc

Download the resource at getURL() to the local resource directory, and return the local copy as a File @return File of the local resource

Downloads Shakespeare training data and stores it locally (temp directory). Then set up and return a simple DataSetIterator that does vectorization ba

Downloads Shakespeare training data and stores it locally (temp directory). Then set up and return a simple DataSetIterator that does vectorization ba

Get the training data - a JavaRDD<DataSet> Note that this approach for getting training data is a special case for this example (modelling characters)

A minimal character set, with a-z, A-Z, 0-9 and common punctuation etc

Used post training to convert a String to a features INDArray that can be passed to the network output method @param reviewContents Contents of the r

Used post training to convert a String to a features INDArray that can be passed to the network output method @param reviewContents Contents of the r

- takes one record at a time from names list using iter iterator - stores it into Writable List and returns it @return

Normalize the data in a given column. Normally one would use datavec. @param c column to normalise. @param tmpData java float array.

Generate an xy plot of the datasets provided.

Generate an xy plot of the datasets provided.

Prepend the linear regression's constant term (ones column). This avoids the case in which all the features are zero thar produce a zero prediction, w

used for testing and training @param csvFileClasspath @param batchSize @param labelIndex @param numClasses @return @throws IOException @throws Interr

used for testing and training

@return Name of the resource (used for inferring local storage parent directory)

Generate a sample from the network, given an (optional, possibly null) initialization. Initialization can be used to 'prime' the RNN with a sequence y

Given a probability distribution over discrete classes, sample from the distribution and return the generated class index. @param distribution Probabi

A custom property used in this custom layer example. See the CustomLayerExampleReadme.md for details @param secondActivationFunction Second activatio

A custom property used in this custom layer example. See the README.md for details @param secondActivationFunction Second activation function for the

Element-wise differences are squared, and then summed. This is modelled after the content_loss method defined in https://harishnarayanan.org/writing/a