From b6070c2074c1ae9f36bab40bdccd44bdebc8354e Mon Sep 17 00:00:00 2001 From: carlosuc3m <49989524+carlosuc3m@users.noreply.github.com> Date: Mon, 4 Sep 2023 13:26:16 +0200 Subject: [PATCH] Update README.md --- README.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 0e50ede6..8d2100f3 100644 --- a/README.md +++ b/README.md @@ -5,12 +5,14 @@ This project provides a Java library for running Deep Learning (DL) models agnostically, enabling communication between Java software and various Deep Learning frameworks (engines). It also allows the use of multiple DL frameworks in the same session, manages the different DL frameworks and brings the models from the [Bioimage.io](https://bioimage.io/#/) repository to Java. -This library is intended for developers and was originally built by the DeepIcy team as the backend of the DeepIcy plugin. +It is intended for developers and was originally built by the DeepIcy team as the backend of the DeepIcy plugin. [JDLL]() is able to load models and make inference, create tensors, download Bioiamge.io models and manage the supported DL frameworks. The library is designed in a modular way, allowing the main software to avoid dealing with the various objects and structures required by different DL frameworks. Instead the Java model runner provides interfaces for models and tensors that handle internally their creation and inference in the differnet Java engines. The main software only needs to interact with the Java model runner and does not need to worry whether the model is in PyTorch, Tensorflow or other framework. **FOR A MORE COMPREHENSIVE AND COMPLETE EXPLANATION OF JDLL, PLEASE VISIT THE [WIKI](https://github.com/bioimage-io/JDLL/wiki).** +# Quickstart + ## Setting Up the Model Runner 1. Download the dependency and include it in your project