RESTful Machine Learning with Flask and TensorFlow Serving
summary
A Python library that takes Keras models from notebook to production by scaffolding Docker and TensorFlow Serving, exposing a Flask + Swagger REST API that proxies predictions to TensorFlow Serving with Docker volumes, provides a CLI and automatic model versioning in SQLite via SQLAlchemy, and demos sub-second predictions. It works end-to-end for simple deployments but lacks extensive testing, batching and gRPC performance optimizations, and cloud/Kubernetes deployment examples; security and large-input throughput remain unresolved.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 6:03Deploying ML models challenges
- 12:25Motivation for new library
- 16:57Library features and goals
- 23:49Flask integration
- 26:38Live model deployment demo
- 31:05Open sourcing and future plans
- 32:15Q&A
related talks
01h 18mPutting Tensorflow Models into Production
01h 18mPutting Tensor Flow Models Into Production
46mAccelerating Machine Learning DevOps with Kubeflow
22mThe Containerization of Machine Learning: TensorFlow: Kubernetes, and Kubeflow
32mDeploying Python models to production
35mGetting Machine Learning Models Ready For Production Using Python
42mDeploying ML at Scale with Kubernetes and TFX
37mSimplifying Training Deep & Serving Learning Models...using Tensorflow