Support Yolov5(4.0)/Yolov5(5.0)/YoloR/YoloX/Yolov4/Yolov3/CenterNet/CenterFace/RetinaFace/Classify/Unet. use darknet/libtorch/pytorch/mxnet to onnx to tensorrt

Torchserve server using a YoloV5 model running on docker with GPU and static batch inference to perform production ready and real time inference.

Batch LLM Inference with Ray Data LLM: From Simple to Advanced

PipelineScheduler optimizes workload distribution between servers and edge devices, setting optimal batch sizes to maximize throughput and minimize latency amid content dynamics and network instability. It also addresses resource contention with spatiotemporal inference scheduling to reduce co-location interference.

Analyze and generate unstructured data using LLMs, from quick experiments to billion token jobs.

Ray Saturday Dec 2022 edition

Torchfusion is a very opinionated torch inference on datafusion.

Serve pytorch inference requests using batching with redis for faster performance.

Support batch inference of Grounding DINO. "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"

LightGBM Inference on Datafusion

简单的 Ollama JSONL 批量推理工具 / Simple Ollama JSONL batch inference tool.

This repository provides sample codes, which enable you to learn how to use auto-ml image classification, or object detection under Azure ML(AML) environment.

Batch LLM Inference with Ray Data LLM: From Simple to Advanced

This repo simulates how an ML model moves to production in an industry setting. The goal is to build, deploy, monitor, and retrain a sentiment analysis model using Kubernetes (minikube) and FastAPI.

Production-grade customer segmentation pipeline built on Azure (Blob Storage, Data Factory, Azure ML, Batch Endpoint). Includes end-to-end data engineering, feature engineering, K-Means model training, and scalable batch inference.

We perform batch inference on lead scoring task using Pyspark.

MLOps project that recommends movies to watch implementing Data Engineering and MLOps best practices.

🚀 Process JSON data in batches with `llm-batch`, leveraging sequential or parallel modes for efficient interaction with LLMs.