Patterson Consulting

[Patterson Consulting @PattersonCnsltg]

We also review some methods on how to do feature engineering in sequential data that can represent temporal features in a tabular structure.

[Patterson Consulting @PattersonCnsltg]

https://t.co/CySD6z8aMQ

In part 3 of 5 of our ongoing series on detecting credit card fraud, we introduce Amazon SageMaker Studio Lab and show how to connect it to data inside Snowflake.

1/2

[Patterson Consulting @PattersonCnsltg]

In this post we introduce @awscloud SageMaker Studio Lab, explain what it is and how it is different than other SageMaker versions, and then perform some further feature engineering with python on data inside @SnowflakeDB .

[Patterson Consulting @PattersonCnsltg]

/3 this series is a great example of an application being built out on the "modern data stack" where we leverage a native-SQL MPP cloud data warehouse to build our applications with #snowflake.

[Patterson Consulting @PattersonCnsltg]

https://t.co/wHYA004uyA

In this post we build on existing credit card detection research and datasets and build our scalable feature engineering pipeline on #snowflake and #snowpark .

[Patterson Consulting @PattersonCnsltg]

/2 We use the Scala dataframes API in Snowpark and then save the results back into Snowflake to be used in part 3, where we start trying model variables based on the feature engineering.

[Patterson Consulting @PattersonCnsltg]

https://t.co/8xLTPVG82A

Part 1 of 5 on our latest series on building real-world data applications. In this series we work with credit card transaction data to build and deploy a credit card analysis workbench for an end user.

[Patterson Consulting @PattersonCnsltg]

https://t.co/utU4DROPlu

In the final post in this series on Applied Predictive Maintenance, we use Snowpark to score data inside of @SnowflakeDB

[Patterson Consulting @PattersonCnsltg]

https://t.co/Pcx7Y06UHI

In part 5 of 6 of our series on Applied Predictive Maintenance we look at how to forecast the business value of machine learning models

[Patterson Consulting @PattersonCnsltg]

Part 4 of 6: Building Workflows with @scikit_learn and @GoogleColab for Predictive Maintenance Models https://t.co/aTmGOUG76s

[Josh Patterson - @[email protected] @jpatanooga]

https://t.co/4srA7Ng7xT

In part 3 of 6 of our ongoing series on Applied Predictive Maintenance, we take the data we loaded into #snowflake and do some exploratory data analysis #eda work to better understand the structure in the sensor dataset. #dataanalysis #machinelearning

[Patterson Consulting @PattersonCnsltg]

https://t.co/FYycNTrsIo

Check out Part 2 of our Predictive Maintenance series on "Loading sensor data into Snowflake"

[Patterson Consulting @PattersonCnsltg]

https://t.co/AmfCWQ6lcb

Check out our latest applied machine learning series on Applied Predictive Maintenance

[Patterson Consulting @PattersonCnsltg]

https://t.co/fWi5ksEDfD

Today we are sharing one of latest use case examples where we detect visual anomalies in manufacturing with Generative Adversarial Networks (GANs). GANs are unsupervised learning, so there are no labels on the dataset.

[Patterson Consulting @PattersonCnsltg]

@TheJeremyHaynes great question -- I imagine there are multiple projects experimenting with it in the same way we see projects built on GPT3

[Patterson Consulting @PattersonCnsltg]

https://t.co/qZ6VFEyVXE

A new issue of "Applied Machine Learning (Somewhat) Quarterly" is up on our blog. We lightly cover topics such as RAFT, DALL-E, Visual Transformers, and other language modeling topics.

[Patterson Consulting @PattersonCnsltg]

https://t.co/Bmn5zEf2C8

Video preview of the MIG device valuation

[Patterson Consulting @PattersonCnsltg]

https://t.co/pDCGDCw6uT

Curious how to value NVIDIA MIG device GPUs in terms of power vs performance? check out this article by @jpatanooga on understanding the value of @nvidia MIG devices

[Patterson Consulting @PattersonCnsltg]

https://t.co/lH4DxqgRss

Signup: https://t.co/spEZUrCP2a

The evolution of open source acceptance in enterprise is a key driver in how AI infrastructure is deployed today. This briefing takes your team through key areas of successful MLOps today based on lessons learned.

[Patterson Consulting @PattersonCnsltg]

- @kubeflow 1.3 is out the door, with significant improvements in Katib, TensorBoard integration, Kale, and the overall UX: https://t.co/dfM8vhP9xl