Load TensorFlow Hub Model

Loads a Model from the TensorFlow Hub and wraps it into a Network with the defined inputs and the outputs of the Hub model.

See https://tfhub.dev/

* TensorFlow, the TensorFlow logo and any related marks are trademarks of Google Inc.

Options

Input definition

A list of input definitions. Each definition contain a shape which is a comma separated list of integers (or '?' for unknown dimensions). One line represents one input.%%00010The inputs can be named by starting a line with the input name followed by a colon.%%00010The type of the inputs can be defined by appending a semicolon followed by the name of the type as defined by TensorFlow.%%00010%%00010[: ], [, ...][; ]%%00010[: ], [, ...][; ]%%00010[...]%%00010%%00010Example for one input image of size 224, 224 with 3 channels:%%00010%%00010224, 224, 3%%00010%%00010Example for two inputs. A 2d image of unknown size and a vector of 100 integers:%%00010%%00010image: ?, ?, 3%%00010data: 100; int32

TensorFow Hub URL

The URL of the TensorFlow Hub model which should be used.

Nodes

• Component Input1 ×
• Component Output1 ×
• DL Python Network Creator1 ×
• List Box Configuration1 ×
• String Configuration1 ×
• Show all 6 nodes

Extensions

• FeatureKNIME Base nodes
• FeatureKNIME Deep Learning - Keras Integration
• FeatureKNIME Quick Forms

Links

• No links available