reexports

compile

evaluate

predict

These objects are imported from other packages. Follow the links
below to see their documentation.
<dl>
 <dt>keras</dt>
<dd><code>compile</code>, <code>evaluate</code>, <code>fit</code></dd>
 <dt>stats</dt>
<dd><code><a href="/link/predict?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::predict">predict</a></code></dd>
</dl>

internal

Means to predict process flow, such as process outcome, next activity, next time, remaining time, and remaining trace. Off-the-shelf predictive models based on the concept of Transformers are provided, as well as multiple ways to customize the models. This package is partly based on work described in Zaharah A. Bukhsh, Aaqib Saeed, & Remco M. Dijkman. (2021). "ProcessTransformer: Predictive Business Process Monitoring with Transformer Network" <arXiv:2104.00721>.

Gert Janssenswillen

processpredictR

Process Prediction

Ivan Esin

Hasselt University 

reexports function

<dl><dt>optimizer</dt>
<dd>Default optimizer for ppred_model</dd>
<dt>loss</dt>
<dd>Default loss for ppred_model</dd>
<dt>metrics</dt>
<dd>Default metrics for ppred_model</dd>
<dt>train_data</dt>
<dd>A training dataset</dd>
<dt>batch_size</dt>
<dd>A batch size</dd>
<dt>num_epochs</dt>
<dd>A number of epochs</dd>
<dt>verbose</dt>
<dd>A verbose</dd>
<dt>callbacks</dt>
<dd><code><a href="/link/list?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::list">list</a></code>: A list of callbacks. <code>keras</code> default is <code>NULL</code>, but can be adjusted (ex. keras::callback_csv_logger(filename = paste("log_", object$task)), #or NULL
keras::callback_tensorboard())</dd>
<dt>shuffle</dt>
<dd><code><a href="/link/logical?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::logical">logical</a></code> (default <code><a href="/link/TRUE?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::TRUE">TRUE</a></code>): If <code><a href="/link/TRUE?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::TRUE">TRUE</a></code> shuffles the data</dd>
<dt>validation_split</dt>
<dd>A ratio to split on</dd>
<dt>object</dt>
<dd><code>ppred_model</code> (default <code><a href="/link/NULL?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::NULL">NULL</a></code>): ProcessTransformer model of class <code>ppred_model</code>.</dd>
<dt>test_data</dt>
<dd><code>ppred_examples_df</code> (default <code><a href="/link/NULL?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::NULL">NULL</a></code>): preprocessed test data.</dd>
<dt>append</dt>
<dd><code><a href="/link/logical?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::logical">logical</a></code> (default <code><a href="/link/FALSE?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::FALSE">FALSE</a></code>): if <code><a href="/link/TRUE?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::TRUE">TRUE</a></code>, returns a passed <code><a href="/link/data.frame?package=processpredictR&version=0.1.0" data-mini-rdoc="processpredictR::data.frame">data.frame</a></code> with predicted values.</dd>
<dt>...</dt>
<dd>Additional arguments</dd></dl>

Arguments

These objects are imported from other packages. Follow the links
below to see their documentation.
<dl>
 <dt>keras</dt>
<dd><code>compile</code>, <code>evaluate</code>, <code>fit</code></dd>


 <dt>stats</dt>
<dd><code><a href='https://rdrr.io/r/stats/predict.html'>predict</a></code></dd>


</dl>

Default compile function for ProcessTransformer model — reexports

<dl>

<dt>optimizer</dt>
<dd>Default optimizer for ppred_model</dd>


<dt>loss</dt>
<dd>Default loss for ppred_model</dd>


<dt>metrics</dt>
<dd>Default metrics for ppred_model</dd>


<dt>train_data</dt>
<dd>A training dataset</dd>


<dt>batch_size</dt>
<dd>A batch size</dd>


<dt>num_epochs</dt>
<dd>A number of epochs</dd>


<dt>verbose</dt>
<dd>A verbose</dd>


<dt>callbacks</dt>
<dd><code><a href='https://rdrr.io/r/base/list.html'>list</a></code>: A list of callbacks. <code>keras</code> default is <code>NULL</code>, but can be adjusted (ex. keras::callback_csv_logger(filename = paste("log_", object$task)), #or NULL
keras::callback_tensorboard())</dd>


<dt>shuffle</dt>
<dd><code><a href='https://rdrr.io/r/base/logical.html'>logical</a></code> (default <code><a href='https://rdrr.io/r/base/logical.html'>TRUE</a></code>): If <code><a href='https://rdrr.io/r/base/logical.html'>TRUE</a></code> shuffles the data</dd>


<dt>validation_split</dt>
<dd>A ratio to split on</dd>


<dt>object</dt>
<dd><code>ppred_model</code> (default <code><a href='https://rdrr.io/r/base/NULL.html'>NULL</a></code>): ProcessTransformer model of class <code>ppred_model</code>.</dd>


<dt>test_data</dt>
<dd><code>ppred_examples_df</code> (default <code><a href='https://rdrr.io/r/base/NULL.html'>NULL</a></code>): preprocessed test data.</dd>


<dt>append</dt>
<dd><code><a href='https://rdrr.io/r/base/logical.html'>logical</a></code> (default <code><a href='https://rdrr.io/r/base/logical.html'>FALSE</a></code>): if <code><a href='https://rdrr.io/r/base/logical.html'>TRUE</a></code>, returns a passed <code><a href='https://rdrr.io/r/base/data.frame.html'>data.frame</a></code> with predicted values.</dd>


<dt>...</dt>
<dd>Additional arguments</dd>

</dl>

reexports: Default compile function for ProcessTransformer model

Description

Arguments

See Also