> 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/deeplearning4j.md).

# Deeplearning4j

- [Quickstart](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/quickstart.md): End-to-end quickstart guide — from Maven setup to training an MNIST classifier in Deeplearning4j
- [Core Concepts](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/concepts.md): Architecture overview of Deeplearning4j — MultiLayerNetwork, ComputationGraph, the training pipeline, and DL4J's relationship to ND4J
- [Beginners Guide](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/beginners.md): Getting started with deep learning and Deeplearning4j — prerequisites, learning path, and recommended resources
- [Cheat Sheet](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/cheat-sheet.md): Quick reference cheat sheet for Deeplearning4j — common configurations, layer types, and API patterns
- [Examples Tour](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/examples-tour.md): Tour of the dl4j-examples repository — classification, regression, CNN, RNN, and advanced examples
- [Neural Networks](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork.md): The MultiLayerNetwork API — building, configuring, training, evaluating, and using sequential neural networks
- [ComputationGraph](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/computationgraph.md): The ComputationGraph API — multi-input, multi-output, skip connections, and arbitrary DAG topologies
- [Layers Reference](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/layers.md): Complete reference for all layer types in Deeplearning4j — Dense, Activation, Dropout, Embedding, BatchNormalization, and more
- [Convolutional Layers](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/convolutional.md): CNN layers in Deeplearning4j — Conv1D/2D/3D, pooling, deconvolution, depthwise, separable, and upsampling layers
- [Recurrent Layers](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/recurrent.md): RNN layers in Deeplearning4j — LSTM, GRU, Bidirectional wrapper, masking, TBPTT, and sequence data handling
- [Autoencoders](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/autoencoders.md): Autoencoder and Variational Autoencoder layers in Deeplearning4j — architecture, configuration, and training
- [Graph Vertices](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/vertices.md): Vertex types for ComputationGraph — Merge, ElementWise, Subset, Stack, Reshape, and custom vertices
- [Data Iterators](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/iterators.md): DataSetIterator implementations — built-in iterators, custom iterators, async loading, and data splitting
- [Listeners](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/listeners.md): Training listeners — ScoreIterationListener, PerformanceListener, EvaluativeListener, CheckpointListener, and custom listeners
- [Evaluation](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/evaluation.md): Detailed evaluation guide — classification metrics, regression metrics, ROC curves, calibration, and multi-output evaluation
- [Model Persistence](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/model-persistence.md): Saving and loading neural networks — ModelSerializer, saving normalizers, and model format details
- [Transfer Learning](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/transfer-learning.md): Transfer learning in Deeplearning4j — TransferLearning.Builder, FineTuneConfiguration, freezing layers, and modifying pretrained networks
- [Early Stopping](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/early-stopping.md): Early stopping in Deeplearning4j — EarlyStoppingConfiguration, termination conditions, score calculators, and model savers
- [Custom Layers](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/custom-layer.md): Writing custom layers in Deeplearning4j — extending Layer, SameDiff-backed custom layers, and custom graph vertices
- [Visualization](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/multilayernetwork/visualization.md): Visualizing training progress — VertxUIServer, StatsListener, the training dashboard, and t-SNE visualization
- [NLP](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview.md): Natural language processing in Deeplearning4j — Word2Vec, Doc2Vec, tokenization, and text processing pipeline
- [Word2Vec](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview/word2vec.md): Word2Vec in Deeplearning4j — Skip-Gram, CBOW, training word vectors, and using pretrained embeddings
- [Doc2Vec](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview/doc2vec.md): Document vectors with ParagraphVectors (Doc2Vec) — training document embeddings and document similarity
- [Sentence Iterators](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview/sentence-iterator.md): Text input for NLP — SentenceIterator, BasicLineIterator, FileSentenceIterator, and custom iterators
- [Tokenization](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview/tokenization.md): Tokenizer factories in Deeplearning4j — DefaultTokenizerFactory, custom tokenizers, and preprocessors
- [Vocabulary Cache](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview/vocabulary-cache.md): VocabCache in Deeplearning4j — how vocabulary is stored, InMemoryLookupCache, and vocabulary management
- [Distributed Training](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1.md): When and how to use distributed training — architecture overview, Spark integration, and ParallelWrapper
- [Spark How-To](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/spark-howto.md): Step-by-step guide to distributed training with Apache Spark — setup, data loading, training, and evaluation
- [Spark Data Pipelines](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/spark-data-howto.md): Loading and preprocessing data on Apache Spark for distributed DL4J training
- [Spark API Reference](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/spark-api-reference.md): API reference for SparkDl4jMultiLayer, SparkComputationGraph, and TrainingMaster
- [Parameter Server](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/parameter-server.md): Gradient sharing via the Aeron-based parameter server for distributed training
- [Technical Reference](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/technical-reference.md): Technical details of distributed training — Strom ASGD algorithm, mesh networking, and fault tolerance
- [ParallelWrapper (Multi-GPU)](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-1/parallel-wrapper.md): Multi-GPU training with ParallelWrapper and high-throughput inference with ParallelInference in DL4J 1.0.0-rewrite.
- [Model Zoo](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-2.md): Pretrained models in Deeplearning4j — available architectures, using pretrained weights, and transfer learning from zoo models
- [Available Models](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-2/models.md): Complete list of pretrained models — VGG, ResNet, AlexNet, LeNet, YOLO, UNet, and other architectures
- [Training UI](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-3.md): Web-based training visualization — VertxUIServer, StatsListener, training dashboard, and t-SNE visualization
- [Troubleshooting](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/troubleshooting.md): Common training problems and solutions — NaN loss, slow convergence, overfitting, memory errors, and debugging tips
- [Benchmarking](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/benchmarking.md): How to benchmark DL4J and ND4J — OpProfiler, performance metrics, and comparing configurations
- [Building from Source](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/build-from-source.md): How to build Deeplearning4j from source — prerequisites, Maven build, libnd4j C++ compilation, CUDA and CPU builds, and common issues
- [Contributing](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/contributing.md): How to contribute to Deeplearning4j — Eclipse CLA, build process, project architecture, adding ops, writing examples, and pull request workflow
- [LLM & VLM Stack](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/overview-4.md): Complete guide to the samediff-llm generation pipeline, KV cache management, speculative decoding, continuous batching, tokenizers, evaluation framework, and model editing
- [PEFT & RL Alignment Training](https://deeplearning4j.konduit.ai/en-1.0.0-rewrite/deeplearning4j/peft-and-rl.md): Parameter-efficient fine-tuning (12 methods), reinforcement learning alignment (9 trainers), mixed-precision training, knowledge distillation, and dataset curation