[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-mlflow":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"f562db8d-7e57-4a83-837e-602867b2f0d7","MLflow","mlflow",3,"MLflow 是用來管理機器學習實驗、模型版本與部署流程的開源平台，常和 MLOps、SageMaker、S3 一起出現。它讓訓練參數、指標、模型產物與追蹤紀錄可重現，也方便比較不同資料量與微調設定的效果。","MLflow is an open-source platform for tracking experiments, versioning models, and moving them through training and deployment workflows. It matters in MLOps because it keeps parameters, metrics, artifacts, and run history reproducible across fine-tuning setups, cloud pipelines, and model comparisons.",[12,21],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"8ebda40b-9172-4a86-bc27-52f4d301f210","mlops-explained-how-ml-teams-ship-models-zh","MLOps 是什麼？ML 團隊怎麼上線模型","MLOps 把模型訓練、測試、部署和監控變成可重複流程。這篇用 AWS 的視角，拆解它怎麼運作、為何重要，以及和 DevOps 的差別。","tools","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775143133896-37yf.png","zh","2026-04-02T15:18:31.788287+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":26,"image_url":27,"cover_image":27,"language":19,"created_at":28},"39da7a6f-9cdb-4df7-b624-4a6e65102f6f","aws-s3-sagemaker-unified-studio-fine-tuning-zh","AWS 用 S3 加速 LLM 微調","AWS 示範怎麼用 SageMaker Unified Studio、S3 和 MLflow，拿 DocVQA 資料微調 Llama 3.2 11B Vision Instruct，並比較 1,000、5,000、10,000 筆資料的訓練效果。","model-release","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775139361040-b83t.png","2026-04-02T14:15:37.601407+00:00"]