OneFlow | BestHub

DaTaobao Tech

Apr 26, 2024 · Artificial Intelligence

Accelerating Stable Diffusion Models: Evaluation of FlashAttention2, OneFlow, DeepCache, Stable-Fast, and LCM-LoRA

Our benchmark of FlashAttention2, OneFlow, DeepCache, Stable‑Fast, and LCM‑LoRA on Stable Diffusion models shows that DeepCache combined with PyTorch 2.2 consistently cuts inference time by 40‑50% with minimal code changes, while OneFlow offers 20‑40% speedups when compatible, making DeepCache the recommended default acceleration.

DeepCacheFlashAttention2LCM-LoRA

0 likes · 10 min read

Accelerating Stable Diffusion Models: Evaluation of FlashAttention2, OneFlow, DeepCache, Stable-Fast, and LCM-LoRA

DataFunSummit

Apr 11, 2023 · Artificial Intelligence

OneFlow Coop: Joint Optimization of Dynamic‑Graph Recomputation and Memory Allocation

This article introduces OneFlow Coop, a memory‑optimization technique that jointly optimizes dynamic‑graph recomputation strategies and GPU memory allocation by analyzing existing DTR limitations, proposing recomputable in‑place, op‑guided tensor allocation, and layout‑aware eviction modules, and demonstrating superior experimental results.

Deep LearningDynamic GraphGPU memory

0 likes · 18 min read

OneFlow Coop: Joint Optimization of Dynamic‑Graph Recomputation and Memory Allocation

Xiaohongshu Tech REDtech

Oct 14, 2022 · Artificial Intelligence

Machine Learning Frameworks in Industrial Applications: Challenges and Prospects

The talk highlighted how mainstream deep‑learning frameworks such as TensorFlow and PyTorch lower entry barriers, yet fast‑growing industrial platforms like Xiaohongshu face unique timeliness and sparse‑data challenges that often demand custom frameworks, prompting discussion of developer preparation, hardware optimization, and the distinct innovations of domestic versus foreign solutions.

Deep Learning FrameworksOneFlowTech Talk

0 likes · 6 min read

Machine Learning Frameworks in Industrial Applications: Challenges and Prospects

DataFunTalk

Aug 3, 2022 · Artificial Intelligence

Building a Complete Machine Learning Application with OpenMLDB and OneFlow: JD High‑Potential User Purchase Intent Prediction

This tutorial demonstrates how to use OpenMLDB together with OneFlow to build an end‑to‑end machine‑learning pipeline for predicting high‑potential JD users' purchase intent, covering environment setup, data loading, SQL table creation, offline feature extraction, DeepFM model training, model serving, online feature extraction, deployment, and real‑time inference.

DockerFeatureEngineeringModelServing

0 likes · 22 min read

Building a Complete Machine Learning Application with OpenMLDB and OneFlow: JD High‑Potential User Purchase Intent Prediction