Konduit Serving examples

This repository contains examples for using Konduit Serving.

Run examples

To run the examples in this repository, you will have to install a supported version of the Java Development Toolkit (JDK) and Python. JDK version 8 and Python 3.7+ is recommended.

Install Git and run the following command in your terminal:

git clone https://github.com/KonduitAI/konduit-serving-examples.git

Use the package manager pip to install the konduit package.

pip install konduit
# if you have git and python3 installed
# for Windows
konduit-python --os windows-x86_64

You may have to install PyTorch or TensorFlow to run some of the examples in this repository.

Detailed installation steps are available here.

Examples

Python

1. Serving a YAML configuration:
   • Command line: 00a_serve-from-yaml_cli.ipynb
   • Python: 00b_serve-from-yaml_py.ipynb
2. Serving a model built with TensorFlow 1.x
   • 01_tensorflow-basic-mnist.ipynb
   • 02_tensorflow-basic-bert.ipynb
3. Serving a model built with Deeplearning4j: 03_deeplearning4j-multilayernetwork.ipynb
4. Serving a DataVec TransformProcess: 04_datavec-etl.ipynb
5. Serving an ONNX model file with ONNX Runtime (PyTorch, CNTK, MXNet)
   • PyTorch: 05_onnx_pytorch.ipynb
6. Serving a Keras model:
   • LSTM: 06_keras.ipynb
   • MNIST with TensorFlow 2.0: 07_tensorflow-keras_basic-mnist.ipynb
7. Model monitoring
   1. Quickstart: monitoring/quickstart

Java

1. Configuring a DataVec transform process: BasicConfiguration.java
2. Loading and transforming an image file: BasicConfigurationImage.java
3. Serving a Python script: BasicConfigurationPython.java