Docker for Machine Learning!
summary
Builds and containerizes a complete ML pipeline: trains SVM and Random Forest on the Iris dataset, pickles the models, exposes prediction endpoints with Flask and Swagger via port 8888, and packages everything in a Dockerfile using an Anaconda base then pushes the image to Docker Hub. Demonstrates GPU-enabled deployment on Cisco UCS C480ML with 8 Nvidia V100 GPUs, pulls a TensorFlow NGC container, runs tf_cnn_benchmark on ImageNet, shows GPUs near 100% and server power ~3 kW, and the demos run successfully.
AI-written summary · about
chapters (6)
- 0:00Introduction to Docker for ML
- 1:26Why Docker for machine learning
- 8:57Building ML model with Iris dataset
- 15:29Exposing model as Flask API
- 20:32Dockerizing the ML model
- 23:24GPU support with Cisco UCS
related talks
54mRipped Machine Learning with Distributed Computing
38mAll You Need to Know to Build Your GPU Machine Learning Cloud [B]
35mHow does Docker run machine learning on specialized AI
33mDeploying and Scaling Spark ML and Tensorflow AI Models
25mDevcontainers for AI/ML Workflow Optimization
22mDock Your Jupyter Notebook
35mAccelerate deep learning in the cloud with custom ML environments
16mThe quandary of containerized machine learning