์ฝ˜ํ…์ธ ๋กœ ๊ฑด๋„ˆ๋›ฐ๊ธฐ

Ultralytics ํ†ตํ•ฉ

Welcome to the Ultralytics Integrations page! This page provides an overview of our partnerships with various tools and platforms, designed to streamline your machine learning workflows, enhance dataset management, simplify model training, and facilitate efficient deployment.

Ultralytics YOLO ์—์ฝ”์‹œ์Šคํ…œ ๋ฐ ํ†ตํ•ฉ



Watch: Ultralytics YOLO11 Deployment and Integrations

๋ฐ์ดํ„ฐ ์„ธํŠธ ํ†ตํ•ฉ

  • Roboflow: ๊ฐ•๋ ฅํ•œ ์ฃผ์„, ์ „์ฒ˜๋ฆฌ ๋ฐ ์ฆ๊ฐ• ๊ธฐ๋Šฅ์„ ์ œ๊ณตํ•˜์—ฌ Ultralytics ๋ชจ๋ธ์— ๋Œ€ํ•œ ์›ํ™œํ•œ ๋ฐ์ดํ„ฐ ์„ธํŠธ ๊ด€๋ฆฌ๋ฅผ ์ง€์›ํ•ฉ๋‹ˆ๋‹ค.

๊ต์œก ํ†ตํ•ฉ

  • ์•„๋งˆ์กด ์„ธ์ด์ง€๋ฉ”์ด์ปค: Amazon SageMaker๋ฅผ ํ™œ์šฉํ•˜์—ฌ Ultralytics ๋ชจ๋ธ์„ ํšจ์œจ์ ์œผ๋กœ ๋นŒ๋“œ, ๊ต์œก ๋ฐ ๋ฐฐํฌํ•˜์—ฌ ML ์ˆ˜๋ช… ์ฃผ๊ธฐ๋ฅผ ์œ„ํ•œ ์˜ฌ์ธ์› ํ”Œ๋žซํผ์„ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค.

  • ClearML: Ultralytics ML ์›Œํฌํ”Œ๋กœ์šฐ ์ž๋™ํ™”, ์‹คํ—˜ ๋ชจ๋‹ˆํ„ฐ๋ง, ํŒ€ ํ˜‘์—… ์ด‰์ง„.

  • Comet ML: ๋จธ์‹ ๋Ÿฌ๋‹ ์‹คํ—˜์„ ์ถ”์ , ๋น„๊ต, ์ตœ์ ํ™”ํ•˜์—ฌ Ultralytics ์œผ๋กœ ๋ชจ๋ธ ๊ฐœ๋ฐœ์„ ๊ฐ•ํ™”ํ•˜์„ธ์š”.

  • DVC: Ultralytics ๋จธ์‹  ๋Ÿฌ๋‹ ํ”„๋กœ์ ํŠธ์˜ ๋ฒ„์ „ ๊ด€๋ฆฌ๋ฅผ ๊ตฌํ˜„ํ•˜์—ฌ ๋ฐ์ดํ„ฐ, ์ฝ”๋“œ, ๋ชจ๋ธ์„ ํšจ๊ณผ์ ์œผ๋กœ ๋™๊ธฐํ™”ํ•˜์„ธ์š”.

  • Google Colab: Google Colab์„ ์‚ฌ์šฉํ•˜์—ฌ ํ˜‘์—… ๋ฐ ๊ณต์œ ๋ฅผ ์ง€์›ํ•˜๋Š” ํด๋ผ์šฐ๋“œ ๊ธฐ๋ฐ˜ ํ™˜๊ฒฝ์—์„œ Ultralytics ๋ชจ๋ธ์„ ํ•™์Šตํ•˜๊ณ  ํ‰๊ฐ€ํ•˜์„ธ์š”.

  • IBM ์™“์Šจx: ์ตœ์ฒจ๋‹จ AI ๋„๊ตฌ, ์†์‰ฌ์šด ํ†ตํ•ฉ, ๊ณ ๊ธ‰ ๋ชจ๋ธ ๊ด€๋ฆฌ ์‹œ์Šคํ…œ์œผ๋กœ Ultralytics ๋ชจ๋ธ์˜ ํ•™์Šต ๋ฐ ํ‰๊ฐ€๋ฅผ ๊ฐ„์†Œํ™”ํ•˜๋Š” IBM Watsonx์˜ ๋ฐฉ๋ฒ•์„ ์•Œ์•„๋ณด์„ธ์š”.

  • ์ฃผํ”ผํ„ฐ๋žฉ: JupyterLab์˜ ๋Œ€ํ™”ํ˜• ๋ฐ ์‚ฌ์šฉ์ž ์ง€์ • ๊ฐ€๋Šฅํ•œ ํ™˜๊ฒฝ์„ ์‚ฌ์šฉํ•˜์—ฌ ์‰ฝ๊ณ  ํšจ์œจ์ ์œผ๋กœ Ultralytics ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๊ณ  ํ‰๊ฐ€ํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์•Œ์•„๋ณด์„ธ์š”.

  • Kaggle: ์‚ฌ์ „ ์„ค์น˜๋œ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ, GPU ์ง€์›, ํ˜‘์—… ๋ฐ ๊ณต์œ ๋ฅผ ์œ„ํ•œ ํ™œ๋ฐœํ•œ ์ปค๋ฎค๋‹ˆํ‹ฐ๊ฐ€ ์žˆ๋Š” ํด๋ผ์šฐ๋“œ ๊ธฐ๋ฐ˜ ํ™˜๊ฒฝ์—์„œ Kaggle์„ ์‚ฌ์šฉํ•˜์—ฌ Ultralytics ๋ชจ๋ธ์„ ๊ต์œกํ•˜๊ณ  ํ‰๊ฐ€ํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์‚ดํŽด๋ณด์„ธ์š”.

  • MLFlow: ์‹คํ—˜ ๋ฐ ์žฌํ˜„์„ฑ์—์„œ ๋ฐฐํฌ์— ์ด๋ฅด๊ธฐ๊นŒ์ง€ Ultralytics ๋ชจ๋ธ์˜ ์ „์ฒด ML ์ˆ˜๋ช… ์ฃผ๊ธฐ๋ฅผ ๊ฐ„์†Œํ™”ํ•ฉ๋‹ˆ๋‹ค.

  • Neptune: MLOps์šฉ์œผ๋กœ ์„ค๊ณ„๋œ ์ด ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ ์ €์žฅ์†Œ( Ultralytics )์—์„œ ML ์‹คํ—˜์— ๋Œ€ํ•œ ์ข…ํ•ฉ์ ์ธ ๋กœ๊ทธ๋ฅผ ์œ ์ง€ํ•˜์„ธ์š”.

  • Paperspace Gradient: Paperspace Gradient simplifies working on YOLO11 projects by providing easy-to-use cloud tools for training, testing, and deploying your models quickly.

  • ๋ ˆ์ด ํŠ : ๊ทœ๋ชจ์— ์ƒ๊ด€์—†์ด Ultralytics ๋ชจ๋ธ์˜ ํ•˜์ดํผํŒŒ๋ผ๋ฏธํ„ฐ๋ฅผ ์ตœ์ ํ™”ํ•˜์„ธ์š”.

  • ํ…์„œ๋ณด๋“œ: Ultralytics ML ์›Œํฌํ”Œ๋กœ์šฐ๋ฅผ ์‹œ๊ฐํ™”ํ•˜๊ณ , ๋ชจ๋ธ ๋ฉ”ํŠธ๋ฆญ์„ ๋ชจ๋‹ˆํ„ฐ๋งํ•˜๊ณ , ํŒ€ ํ˜‘์—…์„ ์ด‰์ง„ํ•˜์„ธ์š”.

  • Ultralytics HUB: ์‚ฌ์ „ ๊ต์œก์„ ๋ฐ›์€ Ultralytics ๋ชจ๋ธ ์ปค๋ฎค๋‹ˆํ‹ฐ์— ์•ก์„ธ์Šคํ•˜๊ณ  ๊ธฐ์—ฌํ•˜์„ธ์š”.

  • Weights & Biases (W&B): ์‹คํ—˜์„ ๋ชจ๋‹ˆํ„ฐ๋งํ•˜๊ณ , ๋ฉ”ํŠธ๋ฆญ์„ ์‹œ๊ฐํ™”ํ•˜๊ณ , Ultralytics ํ”„๋กœ์ ํŠธ์—์„œ ์žฌํ˜„์„ฑ๊ณผ ํ˜‘์—…์„ ์ด‰์ง„ํ•˜์„ธ์š”.

๋ฐฐํฌ ํ†ตํ•ฉ

  • CoreML: CoreML, developed by Apple, is a framework designed for efficiently integrating machine learning models into applications across iOS, macOS, watchOS, and tvOS, using Apple's hardware for effective and secure model deployment.

  • Gradio ๐Ÿš€ ์‹ ๊ทœ: ์‹ค์‹œ๊ฐ„ ๋Œ€ํ™”ํ˜• ๊ฐ์ฒด ๊ฐ์ง€ ๋ฐ๋ชจ๋ฅผ ์œ„ํ•ด Gradio์™€ ํ•จ๊ป˜ Ultralytics ๋ชจ๋ธ์„ ๋ฐฐํฌํ•˜์„ธ์š”.

  • NCNN: Developed by Tencent, NCNN is an efficient neural network inference framework tailored for mobile devices. It enables direct deployment of AI models into apps, optimizing performance across various mobile platforms.

  • Neural Magic: ์ˆ˜๋Ÿ‰ํ™” ์ธ์‹ ํ›ˆ๋ จ(QAT) ๋ฐ ๊ฐ€์ง€์น˜๊ธฐ ๊ธฐ๋ฒ•์„ ํ™œ์šฉํ•˜์—ฌ Ultralytics ๋ชจ๋ธ์„ ์ตœ์ ํ™”ํ•˜์—ฌ ์šฐ์ˆ˜ํ•œ ์„ฑ๋Šฅ๊ณผ ๋” ๊ฐ„๊ฒฐํ•œ ํฌ๊ธฐ๋ฅผ ๊ตฌํ˜„ํ•ฉ๋‹ˆ๋‹ค.

  • ONNX: ์—์„œ ๋งŒ๋“  ์˜คํ”ˆ ์†Œ์Šค ํ˜•์‹ Microsoft ๋‹ค์–‘ํ•œ ํ”„๋ ˆ์ž„์›Œํฌ ๊ฐ„์— AI ๋ชจ๋ธ์„ ์‰ฝ๊ฒŒ ์ „์†กํ•˜๊ณ  Ultralytics ๋ชจ๋ธ์˜ ๋‹ค์–‘์„ฑ๊ณผ ๋ฐฐํฌ ์œ ์—ฐ์„ฑ์„ ํ–ฅ์ƒ์‹œํ‚ค๊ธฐ ์œ„ํ•ด ๋งŒ๋“  ์˜คํ”ˆ ์†Œ์Šค ํ˜•์‹์ž…๋‹ˆ๋‹ค.

  • OpenVINO: Intel's toolkit for optimizing and deploying computer vision models efficiently across various Intel CPU and GPU platforms.

  • PaddlePaddle: ๋ฐ”์ด๋‘์˜ ์˜คํ”ˆ์†Œ์Šค ๋”ฅ๋Ÿฌ๋‹ ํ”Œ๋žซํผ( PaddlePaddle )์€ AI ๋ชจ๋ธ์˜ ํšจ์œจ์ ์ธ ๋ฐฐํฌ๋ฅผ ์ง€์›ํ•˜๊ณ  ์‚ฐ์—… ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์˜ ํ™•์žฅ์„ฑ์— ์ค‘์ ์„ ๋‘ก๋‹ˆ๋‹ค.

  • TF GraphDef: ๊ฐœ๋ฐœ์‚ฌ GoogleGraphDef ๋Š” TensorFlow ์˜ ๊ณ„์‚ฐ ๊ทธ๋ž˜ํ”„ ํ‘œํ˜„ ํ˜•์‹์œผ๋กœ, ๋‹ค์–‘ํ•œ ํ•˜๋“œ์›จ์–ด์—์„œ ๋จธ์‹  ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ์ตœ์ ํ™”ํ•˜์—ฌ ์‹คํ–‰ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

  • TF SavedModel: Developed by Google, TF SavedModel is a universal serialization format for TensorFlow models, enabling easy sharing and deployment across a wide range of platforms, from servers to edge devices.

  • TF.js: ๊ฐœ๋ฐœ์ž Google ์—์„œ ๊ฐœ๋ฐœํ•œ TF.js๋Š” ๋ธŒ๋ผ์šฐ์ €์™€ Node.js์—์„œ ๋จธ์‹  ๋Ÿฌ๋‹์„ ์šฉ์ดํ•˜๊ฒŒ ํ•˜๊ธฐ ์œ„ํ•ด ์ž๋ฐ”์Šคํฌ๋ฆฝํŠธ ๊ธฐ๋ฐ˜์˜ ML ๋ชจ๋ธ ๋ฐฐํฌ๋ฅผ ์ง€์›ํ•ฉ๋‹ˆ๋‹ค.

  • TFLite: ๊ฐœ๋ฐœ์‚ฌ GoogleTFLite๋Š” ๋ชจ๋ฐ”์ผ ๋ฐ ์—ฃ์ง€ ๋””๋ฐ”์ด์Šค์— ๋จธ์‹  ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ๋ฐฐํฌํ•˜๊ธฐ ์œ„ํ•œ ๊ฒฝ๋Ÿ‰ ํ”„๋ ˆ์ž„์›Œํฌ๋กœ, ์ตœ์†Œํ•œ์˜ ๋ฉ”๋ชจ๋ฆฌ ๊ณต๊ฐ„์œผ๋กœ ๋น ๋ฅด๊ณ  ํšจ์œจ์ ์ธ ์ถ”๋ก ์„ ๋ณด์žฅํ•ฉ๋‹ˆ๋‹ค.

  • TFLite Edge TPU: Developed by Google for optimizing TensorFlow Lite models on Edge TPUs, this model format ensures high-speed, efficient edge computing.

  • TensorRT: Developed by NVIDIA, this high-performance deep learning inference framework and model format optimizes AI models for accelerated speed and efficiency on NVIDIA GPUs, ensuring streamlined deployment.

  • TorchScript: ํ”„๋ ˆ์ž„์›Œํฌ์˜ ์ผ๋ถ€๋กœ ๊ฐœ๋ฐœ๋œ PyTorch ํ”„๋ ˆ์ž„์›Œํฌ์˜ ์ผ๋ถ€๋กœ ๊ฐœ๋ฐœ๋œ TorchScript ๋Š” Python ์ข…์†์„ฑ ์—†์ด ๋‹ค์–‘ํ•œ ํ”„๋กœ๋•์…˜ ํ™˜๊ฒฝ์—์„œ ๋จธ์‹  ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ํšจ์œจ์ ์œผ๋กœ ์‹คํ–‰ํ•˜๊ณ  ๋ฐฐํฌํ•  ์ˆ˜ ์žˆ๊ฒŒ ํ•ด์ค๋‹ˆ๋‹ค.

  • VS Code: An extension for VS Code that provides code snippets for accelerating development workflows with Ultralytics and also for anyone looking for examples to help learn or get started with Ultralytics.

๋‚ด๋ณด๋‚ด๊ธฐ ํ˜•์‹

๋˜ํ•œ ๋‹ค์–‘ํ•œ ํ™˜๊ฒฝ์— ๋ฐฐํฌํ•  ์ˆ˜ ์žˆ๋„๋ก ๋‹ค์–‘ํ•œ ๋ชจ๋ธ ๋‚ด๋ณด๋‚ด๊ธฐ ํ˜•์‹์„ ์ง€์›ํ•ฉ๋‹ˆ๋‹ค. ์‚ฌ์šฉ ๊ฐ€๋Šฅํ•œ ํ˜•์‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค:

ํ˜•์‹ format ์ธ์ˆ˜ ๋ชจ๋ธ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ ์ธ์ˆ˜
PyTorch - yolo11n.pt โœ… -
TorchScript torchscript yolo11n.torchscript โœ… imgsz, optimize, batch
ONNX onnx yolo11n.onnx โœ… imgsz, half, dynamic, simplify, opset, batch
OpenVINO openvino yolo11n_openvino_model/ โœ… imgsz, half, int8, batch
TensorRT engine yolo11n.engine โœ… imgsz, half, dynamic, simplify, workspace, int8, batch
CoreML coreml yolo11n.mlpackage โœ… imgsz, half, int8, nms, batch
TF SavedModel saved_model yolo11n_saved_model/ โœ… imgsz, keras, int8, batch
TF GraphDef pb yolo11n.pb โŒ imgsz, batch
TF Lite tflite yolo11n.tflite โœ… imgsz, half, int8, batch
TF Edge TPU edgetpu yolo11n_edgetpu.tflite โœ… imgsz
TF.js tfjs yolo11n_web_model/ โœ… imgsz, half, int8, batch
PaddlePaddle paddle yolo11n_paddle_model/ โœ… imgsz, batch
NCNN ncnn yolo11n_ncnn_model/ โœ… imgsz, half, batch

๊ฐ ํ†ตํ•ฉ์— ๋Œ€ํ•œ ์ž์„ธํ•œ ๋‚ด์šฉ๊ณผ ํ†ตํ•ฉ์„ ์ตœ๋Œ€ํ•œ ํ™œ์šฉํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์•Œ์•„๋ณด๋ ค๋ฉด ๋งํฌ๋ฅผ ์ฐธ์กฐํ•˜์„ธ์š”( Ultralytics). ์ „์ฒด ๋ณด๊ธฐ export ์„ธ๋ถ€ ์ •๋ณด์—์„œ ๋‚ด๋ณด๋‚ด๊ธฐ ํŽ˜์ด์ง€๋กœ ์ด๋™ํ•ฉ๋‹ˆ๋‹ค.

ํ†ตํ•ฉ์— ๊ธฐ์—ฌํ•˜๊ธฐ

์ปค๋ฎค๋‹ˆํ‹ฐ๊ฐ€ Ultralytics YOLO ์„ ๋‹ค๋ฅธ ๊ธฐ์ˆ , ๋„๊ตฌ, ํ”Œ๋žซํผ๊ณผ ์–ด๋–ป๊ฒŒ ํ†ตํ•ฉํ•˜๋Š”์ง€ ํ•ญ์ƒ ๊ธฐ๋Œ€๋ฉ๋‹ˆ๋‹ค! YOLO ์„ ์ƒˆ๋กœ์šด ์‹œ์Šคํ…œ๊ณผ ์„ฑ๊ณต์ ์œผ๋กœ ํ†ตํ•ฉํ–ˆ๊ฑฐ๋‚˜ ๊ณต์œ ํ•  ๊ท€์ค‘ํ•œ ์ธ์‚ฌ์ดํŠธ๊ฐ€ ์žˆ๋‹ค๋ฉด ํ†ตํ•ฉ ๋ฌธ์„œ์— ๊ธฐ์—ฌํ•ด ์ฃผ์„ธ์š”.

๊ฐ€์ด๋“œ๋‚˜ ํŠœํ† ๋ฆฌ์–ผ์„ ์ž‘์„ฑํ•˜๋ฉด ๋ฌธ์„œ๋ฅผ ํ™•์žฅํ•˜๊ณ  ์ปค๋ฎค๋‹ˆํ‹ฐ์— ๋„์›€์ด ๋˜๋Š” ์‹ค์ œ ์‚ฌ๋ก€๋ฅผ ์ œ๊ณตํ•˜๋Š” ๋ฐ ๋„์›€์„ ์ค„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. Ultralytics YOLO ์„ ์ค‘์‹ฌ์œผ๋กœ ์„ฑ์žฅํ•˜๋Š” ์ƒํƒœ๊ณ„์— ๊ธฐ์—ฌํ•  ์ˆ˜ ์žˆ๋Š” ํ›Œ๋ฅญํ•œ ๋ฐฉ๋ฒ•์ž…๋‹ˆ๋‹ค.

๊ธฐ์—ฌํ•˜๋ ค๋ฉด ํ’€ ๋ฆฌํ€˜์ŠคํŠธ(PR) ์ œ์ถœ ๋ฐฉ๋ฒ•์— ๋Œ€ํ•œ ์•ˆ๋‚ด๋Š” ๊ธฐ์—ฌ ๊ฐ€์ด๋“œ๋ฅผ ์ฐธ์กฐํ•˜์„ธ์š”(๐Ÿ› ๏ธ). ์—ฌ๋Ÿฌ๋ถ„์˜ ๊ธฐ์—ฌ๋ฅผ ๊ฐ„์ ˆํžˆ ๊ธฐ๋‹ค๋ฆฌ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค!

Ultralytics YOLO ์ƒํƒœ๊ณ„๋ฅผ ๋”์šฑ ํ™•์žฅํ•˜๊ณ  ํ’๋ถ€ํ•œ ๊ธฐ๋Šฅ์œผ๋กœ ๋งŒ๋“ค๊ธฐ ์œ„ํ•ด ํ˜‘๋ ฅํ•ฉ์‹œ๋‹ค ๐Ÿ™!

์ž์ฃผ ๋ฌป๋Š” ์งˆ๋ฌธ

Ultralytics HUB๋ž€ ๋ฌด์—‡์ด๋ฉฐ, ์–ด๋–ป๊ฒŒ ML ์›Œํฌํ”Œ๋กœ์šฐ๋ฅผ ๊ฐ„์†Œํ™”ํ•˜๋‚˜์š”?

Ultralytics HUB is a cloud-based platform designed to make machine learning (ML) workflows for Ultralytics models seamless and efficient. By using this tool, you can easily upload datasets, train models, perform real-time tracking, and deploy YOLO11 models without needing extensive coding skills. You can explore the key features on the Ultralytics HUB page and get started quickly with our Quickstart guide.

๋ฐ์ดํ„ฐ ์ง‘ํ•ฉ ๊ด€๋ฆฌ๋ฅผ ์œ„ํ•ด Ultralytics YOLO ๋ชจ๋ธ์„ Roboflow ์™€ ์–ด๋–ป๊ฒŒ ํ†ตํ•ฉํ•˜๋‚˜์š”?

Ultralytics YOLO ๋ชจ๋ธ์„ Roboflow ์™€ ํ†ตํ•ฉํ•˜๋ฉด ์ฃผ์„, ์ „์ฒ˜๋ฆฌ ๋ฐ ๋ณด๊ฐ•์„ ์œ„ํ•œ ๊ฐ•๋ ฅํ•œ ๋„๊ตฌ๋ฅผ ์ œ๊ณตํ•˜์—ฌ ๋ฐ์ดํ„ฐ ์„ธํŠธ ๊ด€๋ฆฌ๊ฐ€ ํ–ฅ์ƒ๋ฉ๋‹ˆ๋‹ค. ์‹œ์ž‘ํ•˜๋ ค๋ฉด ํ†ตํ•ฉ ํŽ˜์ด์ง€์˜ Roboflow ํ†ตํ•ฉ ํŽ˜์ด์ง€์˜ ๋‹จ๊ณ„๋ฅผ ๋”ฐ๋ฅด์„ธ์š”. ์ด ํŒŒํŠธ๋„ˆ์‹ญ์€ ์ •ํ™•ํ•˜๊ณ  ๊ฐ•๋ ฅํ•œ YOLO ๋ชจ๋ธ์„ ๊ฐœ๋ฐœํ•˜๋Š” ๋ฐ ํ•„์ˆ˜์ ์ธ ํšจ์œจ์ ์ธ ๋ฐ์ดํ„ฐ ์„ธํŠธ ์ฒ˜๋ฆฌ๋ฅผ ๋ณด์žฅํ•ฉ๋‹ˆ๋‹ค.

MLFlow๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ Ultralytics ๋ชจ๋ธ์˜ ์„ฑ๋Šฅ์„ ์ถ”์ ํ•  ์ˆ˜ ์žˆ๋‚˜์š”?

์˜ˆ, ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค. MLFlow๋ฅผ Ultralytics ๋ชจ๋ธ๊ณผ ํ†ตํ•ฉํ•˜๋ฉด ์‹คํ—˜์„ ์ถ”์ ํ•˜๊ณ  ์žฌํ˜„์„ฑ์„ ๊ฐœ์„ ํ•˜๋ฉฐ ์ „์ฒด ML ์ˆ˜๋ช… ์ฃผ๊ธฐ๋ฅผ ๊ฐ„์†Œํ™”ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ํ†ตํ•ฉ ์„ค์ •์— ๋Œ€ํ•œ ์ž์„ธํ•œ ์ง€์นจ์€ MLFlow ํ†ตํ•ฉ ํŽ˜์ด์ง€์—์„œ ํ™•์ธํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ํ†ตํ•ฉ์€ ๋ชจ๋ธ ๋ฉ”ํŠธ๋ฆญ์„ ๋ชจ๋‹ˆํ„ฐ๋งํ•˜๊ณ  ML ์›Œํฌํ”Œ๋กœ๋ฅผ ํšจ์œจ์ ์œผ๋กœ ๊ด€๋ฆฌํ•˜๋Š” ๋ฐ ํŠนํžˆ ์œ ์šฉํ•ฉ๋‹ˆ๋‹ค.

What are the benefits of using Neural Magic for YOLO11 model optimization?

Neural Magic optimizes YOLO11 models by leveraging techniques like Quantization Aware Training (QAT) and pruning, resulting in highly efficient, smaller models that perform better on resource-limited hardware. Check out the Neural Magic integration page to learn how to implement these optimizations for superior performance and leaner models. This is especially beneficial for deployment on edge devices.

๋Œ€ํ™”ํ˜• ๋ฐ๋ชจ๋ฅผ ์œ„ํ•ด Gradio์™€ ํ•จ๊ป˜ Ultralytics YOLO ๋ชจ๋ธ์„ ๋ฐฐํฌํ•˜๋ ค๋ฉด ์–ด๋–ป๊ฒŒ ํ•ด์•ผ ํ•˜๋‚˜์š”?

To deploy Ultralytics YOLO models with Gradio for interactive object detection demos, you can follow the steps outlined on the Gradio integration page. Gradio allows you to create easy-to-use web interfaces for real-time model inference, making it an excellent tool for showcasing your YOLO model's capabilities in a user-friendly format suitable for both developers and end-users.

์ด๋Ÿฌํ•œ ์ผ๋ฐ˜์ ์ธ ์งˆ๋ฌธ์„ ํ•ด๊ฒฐํ•จ์œผ๋กœ์จ ์‚ฌ์šฉ์ž ๊ฒฝํ—˜์„ ๊ฐœ์„ ํ•˜๊ณ  Ultralytics ์ œํ’ˆ์˜ ๊ฐ•๋ ฅํ•œ ๊ธฐ๋Šฅ์— ๋Œ€ํ•œ ๊ท€์ค‘ํ•œ ํ†ต์ฐฐ๋ ฅ์„ ์ œ๊ณตํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ FAQ๋ฅผ ํ†ตํ•ฉํ•˜๋ฉด ์„ค๋ช…์„œ๋ฅผ ํ–ฅ์ƒ์‹œํ‚ฌ ๋ฟ๋งŒ ์•„๋‹ˆ๋ผ Ultralytics ์›น์‚ฌ์ดํŠธ๋กœ์˜ ์œ ๊ธฐ์ ์ธ ํŠธ๋ž˜ํ”ฝ์„ ๋” ๋งŽ์ด ์œ ๋„ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.


๐Ÿ“… Created 11 months ago โœ๏ธ Updated 2 days ago

๋Œ“๊ธ€