> For the complete documentation index, see [llms.txt](https://deeplearning4j.konduit.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/master.md).

# Eclipse Deeplearning4j

Eclipse Deeplearning4j is an open-source, distributed deep learning framework for the JVM. It provides a complete ecosystem for building, training, and deploying neural networks in Java and Scala, with native GPU acceleration via CUDA and high-performance CPU computation via OpenBLAS and oneDNN.

## Getting Started

New to Deeplearning4j? Start here:

* [**Ecosystem Overview**](/en-1.0.0-rewrite/core-concepts/ecosystem-overview.md) — understand how all the libraries fit together
* [**Quickstart**](/en-1.0.0-rewrite/deeplearning4j/quickstart.md) — build and train your first neural network in 10 minutes
* [**Maven Setup**](/en-1.0.0-rewrite/configuration/maven.md) — add DL4J to your project
* [**Beginners Guide**](/en-1.0.0-rewrite/deeplearning4j/beginners.md) — learning resources and roadmap

## The Ecosystem

| Library            | Purpose                                        | Start Here                                                   |
| ------------------ | ---------------------------------------------- | ------------------------------------------------------------ |
| **ND4J**           | N-dimensional array library (NumPy for Java)   | [Overview](/en-1.0.0-rewrite/nd4j/overview.md)               |
| **Deeplearning4j** | Neural network training and inference          | [Quickstart](/en-1.0.0-rewrite/deeplearning4j/quickstart.md) |
| **SameDiff**       | Automatic differentiation and graph execution  | [Overview](/en-1.0.0-rewrite/nd4j/overview-2.md)             |
| **DataVec**        | Data loading, transformation, and ETL          | [Overview](/en-1.0.0-rewrite/datavec/overview.md)            |
| **Model Import**   | Import models from Keras, TensorFlow, ONNX     | [Overview](/en-1.0.0-rewrite/model-import/overview.md)       |
| **Python4J**       | Python interop from Java via CPython embedding | [Overview](/en-1.0.0-rewrite/python4j/overview.md)           |
| **OmniHub**        | Pretrained model registry                      | [Overview](/en-1.0.0-rewrite/omnihub/overview.md)            |
| **Arbiter**        | Hyperparameter optimization                    | [Overview](broken://pages/vyba7jIBoHmQwsE9K4r5)              |

## Core Concepts

Before diving into specific libraries, review the foundational concepts:

* [Tensors and NDArrays](/en-1.0.0-rewrite/core-concepts/tensors-and-ndarrays.md) — the fundamental data structure
* [Data Pipelines](/en-1.0.0-rewrite/core-concepts/data-pipelines.md) — loading and transforming data for training
* [Neural Network Fundamentals](/en-1.0.0-rewrite/core-concepts/neural-net-fundamentals.md) — layers, activations, loss functions
* [The Training Loop](/en-1.0.0-rewrite/core-concepts/training-loop.md) — how training works end-to-end
* [Evaluation](/en-1.0.0-rewrite/core-concepts/evaluation.md) — measuring model performance
* [Memory and Workspaces](/en-1.0.0-rewrite/core-concepts/memory-and-workspaces.md) — off-heap memory management

## Quick Links

| Task                                 | Page                                                                                        |
| ------------------------------------ | ------------------------------------------------------------------------------------------- |
| Configure Maven/Gradle dependencies  | [Maven Setup](/en-1.0.0-rewrite/configuration/maven.md)                                     |
| Set up GPU acceleration              | [GPU and CPU Setup](/en-1.0.0-rewrite/configuration/gpu-cpu.md)                             |
| Train a CNN for image classification | [Convolutional Layers](/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/convolutional.md) |
| Train an RNN/LSTM for sequences      | [Recurrent Layers](/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/recurrent.md)         |
| Import a Keras model                 | [Keras Import](/en-1.0.0-rewrite/model-import/overview-1.md)                                |
| Import a TensorFlow model            | [TensorFlow Import](/en-1.0.0-rewrite/model-import/overview-2/tensorflow.md)                |
| Import an ONNX model                 | [ONNX Import](/en-1.0.0-rewrite/model-import/overview-2/onnx.md)                            |
| Run distributed training on Spark    | [Spark How-To](/en-1.0.0-rewrite/deeplearning4j/overview-1/spark-howto.md)                  |
| Use pretrained models                | [Model Zoo](/en-1.0.0-rewrite/deeplearning4j/overview-2.md)                                 |
| Debug memory issues                  | [Memory Configuration](/en-1.0.0-rewrite/configuration/memory.md)                           |
| Call Python from Java                | [Python4J](/en-1.0.0-rewrite/python4j/overview.md)                                          |

## Versions

This documentation covers **Deeplearning4j 1.0.0-M2.1** (current). Use the version selector in the sidebar or the links below to access documentation for other releases.

| Version        | Status      | Docs                                                             |
| -------------- | ----------- | ---------------------------------------------------------------- |
| **1.0.0-M2.1** | **Current** | You are here                                                     |
| 1.0.0-M1.1     | Previous    | [View docs](https://deeplearning4j.konduit.ai/v/en-1.0.0-m1.1/)  |
| 1.0.0-M1       | Previous    | [View docs](https://deeplearning4j.konduit.ai/v/en-1.0.0-m1/)    |
| 1.0.0-beta7    | Legacy      | [View docs](https://deeplearning4j.konduit.ai/v/en-1.0.0-beta7/) |
| 1.0.0-beta6    | Legacy      | [View docs](https://deeplearning4j.konduit.ai/v/en-1.0.0-beta6/) |