Ultralytics (@ultralytics)

Ultralytics가 비전 AI 모델의 실전 성능 검증 중요성을 강조하며, 학습 성능이 좋아도 실제 환경에서 실패할 수 있다고 설명한다. 검증 기능을 통해 버전 비교, 핵심 지표 추적, 배포 전 오류 분석을 지원하며 모델 준비 상태를 확인할 수 있다고 안내한다.

https://x.com/ultralytics/status/2036467607189020934

#ultralytics #visionai #validation #yolo #modeldeployment

Perplexity (@perplexity_ai)

NVIDIA의 Nemotron 3 Super 모델이 이제 Perplexity 플랫폼과 Agent API, 그리고 Perplexity의 Computer 제품에서 이용 가능해졌다는 발표로, 모델의 접근성과 배포 채널 확대를 알리는 중요한 배포 업데이트입니다.

https://x.com/perplexity_ai/status/2032521063918420286

#nvidia #nemotron3 #perplexity #agentapi #modeldeployment

Sudo su (@sudoingX)

짧은 코멘트: 'pocket'에서부터 3090까지 전체 제품군이 출시(shipping)되고 있다는 언급으로, 소형(경량) 모델들이 휴대폰 등 포켓 디바이스에서 구동될 기회를 얻고 있음을 환영하는 내용. 모바일·엣지 배포 확대를 시사.

https://x.com/sudoingX/status/2029252426310795652

#edgeai #mobile #llm #gpu #modeldeployment

Building a machine learning model is only half the journey — deploying it brings your work to life.
From dataset selection and model training to deployment using Streamlit, Gradio, or cloud platforms like AWS and GCP — this roadmap helps you go from idea to interactive app fast.

Don’t just train models. Deploy them.

📕 https://ebokify.com/machine-learning

#MachineLearning #DataScience #MLOps #AI #ModelDeployment #Python #DeepLearning #Streamlit #Gradio #AWS #GCP

🤖 MLOps: The Missing Link in Your Machine Learning Strategy 🔗

MLOps bridges the gap between data science and engineering, creating sustainable ML systems that actually work in the real world.

A proper MLOps workflow includes:
🔄 Automated data ingestion
🧪 Continuous model training
📊 Performance monitoring
🚨 Drift detection
🚀 Seamless redeployment

👀 https://link.illustris.org/mlopscode2prod

#MachineLearning #MLOps #DataScience #AIEngineering #ModelDeployment #DataDrift #AIPipelines

Learn how to build and deploy a Hugging Face NLP model using Docker.

https://linuxexpert.org/building-and-deploying-a-hugging-face-model-with-docker/

#HuggingFace #Docker #NLP #MachineLearning #AI #Python #ModelDeployment #FastAPI #Transformers #DataScience #APIDevelopment #AIModels

Question about R, mlflow and models...

I am trying to register a R model using the crate flavor in mlflow, and I have some doubts.

I have been able to log and register the model. I have also tested that I can load the model again and use it for prediction (inputs/outputs are data.frames).

I was thinking... that would mean I should write the inference part in R, wouldn't it?

How could I deploy the model so it can be served as a general web service (REST API), not actually relying on final users to use R?

I'm now quite tired, but the only solution I have found is to maybe use plumbr to expose an API receiving a JSON with all the inputs as simple types, and generating the data.frame inside, as I have always done.

Do you think this can be done directly using a crated function? Has anybody done something similar?

Thanks in advance. I think this is a discussion worth having, as there is a lack of documentation on this topic for us R users. :(

#rstats #ml #machinelearning #models #mlflow #ai #datascience #data #prediction #mlops #modeldeployment