Docker for Machine Learning!

preview_player
Показать описание
--

Docker containers are a popular way to deploy ML environments that run consistently across various platforms. Data scientists are interested in pre-configured platforms that support docker images pre-installed with deep learning frameworks such as TensorFlow, Apache MXNet etc. 
This presentation is about getting a quick high-level understanding of how to operationalize machine learning pipeline.  
In this session, we will cover all of the basic topics of containers in the context of machine learning by:
* Taking a simple machine learning problem, building a model and containerize the application.* Setting up the docker environment with GPU support and download and run docker images pre-installed with TensorFlow framework
* Cisco Converged Infrastructure Solutions for AI/ML

Speakers: Haseeb Niazi, Paniraja Koppa
Track: AI, Machine Learning, HPC
Рекомендации по теме
Комментарии
Автор

For those who are stuck because the build fails because of dependency install issue, try not useing the pip freeze command as it includes many dependencies that are not directly required. Either manually or using PyCharm's sync library feature generates a minimal set of dependencies required and this does not break the build process.

aashishadhikari
Автор

For those who do not see any response when they click the Execute button on the web page, you might be missing the following at the end of you docstring for the predict_svm() method after the parameters are defined. Also make sure to indent properly --> responses:
200:
description: Index of predicted class

aashishadhikari
Автор

I hate when tutorials don't release source code

---ktcs