import useBaseUrl from '@docusaurus/useBaseUrl';

import TilesGrid from "@site/src/components/TilesGrid"; import TileCard from "@site/src/components/TileCard"; import FeatureHighlights from "@site/src/components/FeatureHighlights"; import { Zap, GitBranch, Rocket, BookOpen, BarChart3 } from "lucide-react";

MLflow for Deep Learning

MLflow for ML Models provides comprehensive experiment tracking, model management, and deployment capabilities for deep learning workflows. From PyTorch training loops to TensorFlow models, MLflow streamlines your path from experimentation to production.

Get Started

<TilesGrid> <TileCard image="/images/logos/pytorch-logo.svg" title="PyTorch Guide" description="Track custom PyTorch training loops with automatic metric and artifact logging." href="/ml/deep-learning/pytorch" linkText="View guide →" containerHeight={64} /> <TileCard image="/images/logos/TensorFlow-logo.svg" title="TensorFlow Guide" description="Integrate TensorFlow models with autologging and TensorBoard visualization support." href="/ml/deep-learning/tensorflow" linkText="View guide →" containerHeight={64} /> <TileCard image="/images/logos/keras-logo.svg" title="Keras Guide" description="Use Keras 3.0 multi-backend capabilities with unified MLflow tracking." href="/ml/deep-learning/keras" linkText="View guide →" containerHeight={64} /> <TileCard image="/images/logos/spacy-logo.svg" title="spaCy Guide" description="Track spaCy NLP pipelines with automatic logging of model performance and artifacts." href="/ml/deep-learning/spacy" linkText="View guide →" containerHeight={64} /> <TileCard image="/images/logos/huggingface-logo.svg" imageDark="/images/logos/huggingface-logo-dark.svg" title="Transformers Guide" description="Integrate Hugging Face Transformers with MLflow for LLM and NLP model tracking." href="/ml/deep-learning/transformers" linkText="View guide →" containerHeight={64} /> <TileCard image="/images/logos/sentence-transformers-logo.png" title="Sentence Transformers" description="Track embedding models and similarity tasks with sentence-transformers integration." href="/ml/deep-learning/sentence-transformers" linkText="View guide →" containerHeight={64} /> </TilesGrid>

Why MLflow for Deep Learning?

<video src={useBaseUrl("/images/deep-learning/dl-run-comparison.mp4")} controls loop autoPlay muted aria-label="Deep Learning Training Comparison" />

<FeatureHighlights features={[ { icon: Zap, title: "One-Line Autologging", description: "Enable comprehensive tracking with a single line of code for PyTorch Lightning, TensorFlow, and Keras." }, { icon: BarChart3, title: "Real-Time Monitoring", description: "Track metrics, loss curves, and training progress live across epochs and batches." }, { icon: GitBranch, title: "Model Checkpoints", description: "Automatically save and version model checkpoints throughout training with complete lineage tracking." }, { icon: Rocket, title: "Production Deployment", description: "Deploy models with GPU acceleration, batch inference, and cloud platform integration." } ]} />

Learn More

<TilesGrid> <TileCard icon={BookOpen} iconSize={48} title="Model Registry" description="Manage model versions, aliases, and deployment lifecycle with centralized governance." href="/ml/model-registry" linkText="View registry docs →" containerHeight={64} /> <TileCard icon={BarChart3} iconSize={48} title="MLflow Tracking" description="Track experiments, parameters, metrics, and artifacts across all ML workflows." href="/ml/tracking" linkText="View tracking docs →" containerHeight={64} /> <TileCard icon={Rocket} iconSize={48} title="Model Deployment" description="Deploy deep learning models with GPU support, batch inference, and cloud integration." href="/ml/deployment" linkText="View deployment docs →" containerHeight={64} /> </TilesGrid>