deepdetect: a deep learning runtime and REST server for unified training and inference across multiple modalities

What it solves

DeepDetect provides a unified runtime and interface for managing the full lifecycle of deep learning models. It simplifies the process of training, deploying, and running inference on a wide variety of model types—including images, text, and tabular data—without requiring the user to manage complex backend configurations manually.

How it works

DeepDetect operates as a deep learning runtime that can be used via a Python wheel (for in-process inference and a command-line tool), a REST server for long-running services, or Docker containers. It uses Torch as its primary backend for training and inference, with TensorRT support for optimized inference. The system organizes model repositories on the filesystem, avoiding the need for a database dependency.

Who it’s for

It is designed for developers and ML engineers who need a repeatable, automation-friendly way to train and serve models across different modalities (computer vision, NLP, and time series) using a single API surface.

Highlights

• Multi-modal support: Handles image classification, object detection, semantic segmentation, language models (BERT, GPT-2), tabular data, and time series.
• Unified Interface: Offers a CLI, REST API, and Python library for consistent model operations.
• Backend Optimization: Uses Torch for general operations and TensorRT for high-performance inference.
• Flexible Deployment: Supports local execution via Python wheels, remote serving via REST, and containerized deployment via Docker.
• Filesystem-based Storage: Manages model repositories directly on the disk for simplicity.