Does your app use ML? Make it a product with TFX

on YouTube ↗14m

" summary

Announces TFX 1.0 release as an open source, enterprise-ready ML pipeline that runs on Colab, laptops, GKE, Kubeflow and in Vertex pipelines with IAM and VPC-SC support. Shows concrete tooling for NLP (BERT transfer learning, exact match/ROUGE/BLEU slices), mobile/web (TF Lite/TFJS quantization analysis), and a healthcare deployment that reports an estimated 25% increase in patient scores generated in the first year. Presents configs, runtimes, and measured outcomes.

AI-written summary · about

" chapters (5)

1. 0:00Intro
2. 3:22Introducing TFX
3. 5:32Whos already using TFX
4. 9:21TFX for NLP
5. 12:09Resources

" related talks

• 08mManage MLOps and deploy Machine Learning to production with the new and improved TFXRobert Crowe, Tris Warkentin · Google's Machine Learning Virtual Community Day 2021
• 37mEnd-to-End ML Pipelines with KubeFlow and TensorFlow Extended (TFX)Chris Fregly · Scale By The Bay 2019
• 39mDeploy your ML Workloads on Kubernetes with KubeflowKarl Weinmeister · Open Source 101 at Home 2020
• 22mKeynote: Kubeflow ML on KubernetesDavid Aronchick, Vishnu Kannan · KubeCon + CloudNativeCon Europe 2018
• 38mTensorFlow Extended (TFX): Machine Learning Pipelines and Model UnderstandingTulsee Doshi, Kevin Haas, Konstantinos Katsiapis · Google I/O 2019
• 13mAn introduction to MLOps with TensorFlow Extended (TFX)Robert Crowe · Google I/O 2022
• 32mKubeflow Deep DiveDavid Aronchick, Jeremy Lewi · KubeCon + CloudNativeCon Europe 2018
• 10mMachine Learning for mobileKemal El Moujahid · Google's Machine Learning Virtual Community Day 2021