Model: SKIL Model wrapper for DL4J, Keras, TensorFlow and other...

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

Description

SKIL Model wrapper for DL4J, Keras, TensorFlow and other models SKIL has a robust model storage, serving, and import system for supporting major deep learning libraries. SKIL can be used for end-to-end training, configuration, and deployment of models or alternatively you can import models into SKIL.

Usage

Model(model = NULL, model_id = NULL, name = NULL, version = NULL,
  experiment = NULL, labels = "", verbose = FALSE, create = TRUE)

Arguments

model:	Model file path or model instance
model_id:	integer. Unique id for model. If NULL, a unique id will be generated.
name:	string. Name for the model.
version:	integer. Version of the model. Defaults to 1.
experiment:	'Experiment' instance. If NULL, an 'Experiment' object will be created internally.
labels:	string. Labels for this model
verbose:	boolean. If TRUE, prints api response.
create:	boolean. Internal. Do not use.

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')
model <- Model(model="keras.h5", name="my_keras_model", experiment=exp)


## End(Not run)

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