Hi folks! We’ll get started here in about 8 minutes

Thanks for joining early 🙂

You can see the live meeting notes here: https://otter.ai/u/deCTpIB5JJaLu8JAG4bIqhHqcGQ?utm_source=va_chat_linkScroll back to review anything you missed.

Here are the direct links to the resources and events mentioned!

Join the Data Science Chapter: https://usergroups.snowflake.com/data-science/Join the Data Science Discussion: https://community.snowflake.com/s/group/0F93r000000XcZ8CAK/data-scienceFind Your Local User Group Chapter: https://usergroups.snowflake.com/chapters/BLOG | Bringing Enterprise-Grade Python Innovation to the Data Cloud: https://www.snowflake.com/blog/snowpark-python-innovation-available-all-snowflake-customers/WEBINAR | Building Scalable Feature Engineering Pipelines: https://www.snowflake.com/webinar/thought-leadership/building-scalable-feature-engineering-pipelines-with-snowpark-python-and-scikit-learn/BUILD | The Data Cloud Dev Summit: https://www.snowflake.com/build/EVENT | Data Cloud World Tour: https://www.snowflake.com/data-cloud-world-tour/

How do we access the recording of this session in the future?

Hi Erika! The recording will be posted to the event page as well as the discussion group.

Thank you

Does anyone have any questions or action items?Let's capture them in the meeting notes: https://otter.ai/u/deCTpIB5JJaLu8JAG4bIqhHqcGQ?utm_source=va_chat_link

I have a question, how would describe integrating the statistical analysis, such as hypothesis testing into the machine learning model development? I currently use SAS Studio.

Can we use anomaly detection to find data quality issues ?

I'm working with a sequence of pressure data in a system that can't really be labeled -- trying to use dimensionality reduction then clustering to find anomalies

Great questions, we'll get to these in the next pause!

Thanks for sharing your use case, Brent! Would love to hear more.

How to find anomalies in grouped data -- for example let us say we have age groups of customers and their sales and we are interested to find anomalies in each group.

Does the data have to be loaded to h20 cloud? Can all the process be done on premise or on snowflake?

so h20 is basically like a python library and we just use the models in h20 and do the regular python coding to train,test,deploy and everything

This looks like a time series data - why to choose to not include the date field in the model?

can I use H2O as a repository and source of truth for my data and models?

Is there an approach to detecting anomalies in semi-structured and unstructured data?

Got it!

Thanks. I know this is a hard question.

Kurt dimensionality reduction techniques to reduce the search space can definitely help -- depends on how sparse the data set is

Akshay and Alejandro we will get to your questions about H2O momentarily!

How would you suggest to pick the hyper-parameters for Random Forest?

do you have any features in H2O to train these models

Are there any data wrangling features in this AI ecosystem?

Is this a paid service?

do you have the connector for the BI tools

the driverless AI?

like MicroStrategy

are you going to share the slice please?

Do you have features that lead from anomaly detection to active learning/data labeling as far as model retraining goes?

Slides

can you share the GitHub project you referenced pls. Thank you.

Does H20 has the connector for BI tools

Is driverless AI is like dataiku drag and drop method? No coding involved like the demo showed earlier?

where are the feature store been done ?

thanks for answering my question

Very insightful presentation!

Awesome thanks

https://github.com/h2oai/h2o-tutorials/tree/master/best-practices

Thank you!

Thank you!!

Very good presentation

Thank you !!!

thank you megan!

Thank you for great presentation!

Thank you so much!!!

https://github.com/h2oai/h2o-tutorials/tree/master/best-practices

Thank you very much

Thank you

Join the Data Science Chapter: https://usergroups.snowflake.com/data-science/Join the Data Science Discussion: https://community.snowflake.com/s/group/0F93r000000XcZ8CAK/data-scienceFind Your Local User Group Chapter: https://usergroups.snowflake.com/chapters/BLOG | Bringing Enterprise-Grade Python Innovation to the Data Cloud: https://www.snowflake.com/blog/snowpark-python-innovation-available-all-snowflake-customers/WEBINAR | Building Scalable Feature Engineering Pipelines: https://www.snowflake.com/webinar/thought-leadership/building-scalable-feature-engineering-pipelines-with-snowpark-python-and-scikit-learn/BUILD | The Data Cloud Dev Summit: https://www.snowflake.com/build/EVENT | Data Cloud World Tour: https://www.snowflake.com/data-cloud-world-tour/

Thank you Megan, Elsa.

thank you

Thank you!