Skip to content

Tensorflow

What is Tensorflow?

TensorFlow is an open-source software library for dataflow programming across a range of tasks. It is a symbolic math library, and is also used for machine learning applications such as neural networks. In TensorFlow, mathematical operations are represented in the form of a graph. The graph represents the sequential flow of all operations to be performed by TensorFlow.

Use Tensorflow library

The Tensorflow library can be used in a chatbot to use an already generated model. Creating new models with Java is currently limited, so you'll need a Phython development environment. However, reading a generated model works fine using the Java-based chatbot on the Business Bot Platform.

Notice

Pay attention to the compatibility of the model. The model must have been created using Version 1.4.0, because some operating systems do not yet provide the required system libraries.

For model generation, you can use either the Inception V3-Model or a MobileNet for training. The principal difference is that Inception V3 is optimized for accuracy, while the MobileNets are optimized to be small and efficient, at the cost of some accuracy. At this point, we don't go deeper into model generation. For more information, please contact us.

How it works

First you have to copy the library libtensorflow-1.4.0.jar into the lib directory of the chatbot. Subsequently, copy the created model into the resource directory.

Now follow the tutorial with sample code at the TensorFlow website: Installing TensorFlow for Java.