Machine Learning with Apache Spark and H2O AutoML

#apachespark #machinelearning #h2O
In this video we are going to check Apache Spark ML pipeline using H2O AutoML estimators. We will be performing EDA, feature engineering and also run across various algos using H2O. We will pick the best algorithm and save the model pipeline
We will also how to setup and install spark on colab to get started..
For details End to End Spark ML video you can check - • End to End Machine Lea...

[@ijeffking]

I really love your pointers on H2OML. Thank you so much

[@abhi7975]

Hello sir,i am currently in 2nd year of emgineering,i know basic ml algorithms using scikit-learn,where should i start frm now

[@AIEngineeringLife]

You can check this playlist to understand how end to end ml works - kzbin.info/aero/PL3N9eeOlCrP6Y73-dOA5Meso7Dv7qYiUU Then play around with some datasets

[@nimjae3106]

hello. Thank you nice videos. I want to learn streaming framework. which framework do you recommend for real-time streaming data? I heard about apache kafka and apache pulsar.

[@AIEngineeringLife]

Apache Kafka if streaming platform but for processing framework it can be either of Spark Streaming, beam or flink.. Spark streaming has rich functions but might not be as real time as other 2

[@rameshthamizhselvan2458]

All your videos are crystal clear. please upload Reinforcement learning in the same way.

[@AIEngineeringLife]

Thanks Ramesh.. I am not expert in Reinforcement Learning. I typically try to make videos that I have closely worked on or have knowledge in similar domain. In case if I gain experience over time will do RL. Sorry at this time

[@sid321axn]

Very helpful thanks

[@lucasnildaimon7598]

New subscriber here, thank you for this video!

[@AIEngineeringLife]

Thanks Lucas for subscribing. I hope you find the channel useful :)

[@akashramkumar7473]

Thanks for sharing. can you share the notebook link, i dont see it in your github.. a few from the mastering apache spark course is also not in the git repo. Thanks

[@AIEngineeringLife]

Akash.. Did you check my Spark specific repo for it - github.com/srivatsan88/Mastering-Apache-Spark

[@akashramkumar7473]

@@AIEngineeringLife Thanks.

[@20shekhu]

Explained very well...Can you please share the notebook link

[@AIEngineeringLife]

Should be in my git repo. Have you check it - github.com/srivatsan88/

[@20shekhu]

@@AIEngineeringLife yes i checked...it is not thr

[@amrmakharita7185]

May i have access to this course notebooks?

[@AIEngineeringLife]

Sorry amr at this time the notebooks are not in my repo as I want those interested to work in parallel. I will upload it in near future

[@imransharief2891]

Boom finally spark with automl

[@AIEngineeringLife]

One major challenge with Spark is model and hyperparameter selection. Even though native methods are there they are slow, so thought why not cover it as well :)

[@hishailesh77]

Great demo. How to round the prediction values to Integer for multiclass classification problem ?

[@AIEngineeringLife]

If you set softmax if will give class probab and pick max of probability you get