Experiment: SKIL Experiment

In SkymindIO/skilr: Train, deploy and manage your R models with SKIL

Description

Experiments in SKIL are useful for defining different model configurations, encapsulating training of models, and carrying out different data cleaning tasks.

Usage

Experiment(work_space = NULL, experiment_id = NULL,
  name = "experiment", description = "experiment", verbose = FALSE,
  skil_server = NULL, create = TRUE)

Arguments

work_space:	'WorkSpace' instance. If NULL a workspace will be created.
experiment_id:	integer. Unique id for workspace. If 'None', a unique id will be generated.
name:	string. Name for the experiment.
description:	string. Description for the experiment.
verbose:	boolean. If FALSE, api response will be printed.
skil_server:	Optional 'Skil' instance, used when create is false.
create:	boolean. If TRUE a new experiment will be created.

Details

Experiments have a one-to-one relationship with Notebooks and have their own storage mechanism for saving different model configurations when seeking a best candidate.

Examples

## Not run: 
library(skilr)

# Creating a SKIL experiment
skil <- Skil(host='localhost', port=9008, user_id='admin', password='admin')
ws   <- Workspace(skil, name='test_workspace')
exp  <- Experiment(ws, name='test_experiment')


## End(Not run)

SkymindIO/skilr documentation built on May 26, 2019, 1:42 a.m.