, 13 tweets, 4 min read Read on Twitter
Next talking about machine learning on serverless #ServerlessDaysBOS
ML is the practice of examining large amounts of data and finding patterns #ServerlessDaysBOS
ML starts with model training. More data = better model. Data is never in the right format so you need to format. Then you find the best statistical algorithm and get a model #ServerlessDaysBOS
One you have a trained model you deploy the model. Often training is the purvue of a data scientist and the application developer worries about deployment #ServerlessDaysBOS
With serverless you don't need the application developer in order to deploy the model #ServerlessDaysBOS
Using serverless patterns to deploy models is what serverless machine learning is all about #ServerlessDaysBOS
PaaS solution to deploy a model: using a container to wrap the scoring script and dependencies and deploy to manged service #ServerlessDaysBOS
Serverless solution to model deploy just put model and scoring script with proper configurations on the FaaS #ServerlessDaysBOS
Learnings for azure for ml: python is preferred, need it supported in azure functions; data scientists don't want containers; functions are great for event driven classification #ServerlessDaysBOS
For serverless inference and retraining - skeptical on if it works. Serverless is not great for compute intensive jobs but good for short lived low compute scenarios. #ServerlessDaysBOS
FaaS good for data processing. Don't need to worry about sdks and can use triggers and bindings #ServerlessDaysBOS
Serverless OCR? Yes! Tools beyond FaaS help here i.e. eventing infrastructure, workflow orchestration #ServerlessDaysBOS
Challenges of serverless ML: limits on time and memory; often need GPU or other special hardware; cold start times; sensitive data require more advanced security; cluster is better if incoming load is high #ServerlessDaysBOS
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Ann Guilinger
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!