trainModel: trainModel
In kailin-lu/recordlinkR: recordlinkR

Trains and saves a LSTM Variational Autoencoder

trainModel(namesA, namesB, dim.latent = 8, dim.encode = 64,
  dim.decode = 64, max.length = 12, num.encode.layers = 2,
  num.decode.layers = 2, batch.size = 32, epochs = 100, lr = 1e-04,
  validation.split = 0.2, save.dir = "~/blocking_models/",
  reconstruct = TRUE, reconstruct.n = 5, reconstruct.display = 20,
  earlystop = TRUE, earlystop.patience = 10, tensorboard = TRUE,
  tensorboard.runid = as.character(Sys.time()), verbose = 2)

`namesA`	Character vector of names
`namesB`	Character vector of names known to be matches of namesA. Length of namesA and namesB must be equal.
`dim.latent`	Number of dimensions to use in the latent vector
`dim.encode`	Number of hidden units to use in each of the encoding layers
`dim.decode`	Number of hidden units to use in each of the decoding layers
`max.length`	Maximum length of each name
`num.encode.layers`	Number of encoding layers to use
`num.decode.layers`	Number of decoding layers to use
`batch.size`	Batch size of model training
`epochs`	Number of epochs to train for
`lr`	Learning rate of model. This model is configured to use to the Adam optimizer.
`validation.split`	Percentage of data to save for validation
`save.dir`	Directory to save the trained encoder
`reconstruct`	Whether or not show reconstructions
`reconstruct.n`	How many reconstructions to show
`reconstruct.display`	After many epochs to show reconstructions
`earlystop`	TRUE if stopping early when validation loss is no longer decreasing, if FALSE then train for all epochs
`earlystop.patience`	Number of epochs to wait while validation loss does not decrease before stopping training early
`tensorboard`	TRUE if tensorboard metrics are to be recorded. Logs are recorded in the /tmp/ directory
`tensorboard.runid`	Unique identifier for the run to separate tensorboard logs
`verbose`	Verbosity level for training output, 0 = silence, 1 = minimal, 2 = verbose