As we are completing version 3.2 of https://t.co/1pq3aM4VSP (now with a centralized customized meta data catalog) we are already hard at work on v4.0 with more ways 2 make sure data, transforms & models are fair, reducing risk & exposure to litigation. #fintech#risk#dataquality
Excel works, until it doesn't. With scheduled checks, @visu_ai (https://t.co/TNSt47Stdj) would have caught this issue (spelled out, in this case), and many more. Data quality matters. Get in touch to learn more.
#datamanagement#d…https://t.co/kVk5eDmpIt https://t.co/Q8MO3yOgwp
As we are completing version 3.2 of https://t.co/1pq3aM4VSP (now with a centralized customized meta data catalog) we are already hard at work on v4.0 with more ways 2 make sure data, transforms & models are fair, reducing risk & exposure to litigation. #fintech#risk#dataquality
@EricSchles@willkurt Encountered that problem some years back. stability is one of the metrics I track in EDA, in modeling, in pipelines and once deployed, on the incoming data, based on feature and granularity. Implemented in https://t.co/TNSt47Stdj
See other metrics:
https://t.co/5LTCDqu2OS
And that's not even covering the prediction part, @visu_ai build models for you and you can publish them at the click of a button...
#datascience simplified
@sarahcat21 Why not both? @visu_ai is a dashboard, but you can drag and drop notebooks and they become part of the visualizations, reports etc. You can also get clean data back in a @ProjectJupyter notebook or export to an #rstats session.
@f_dion@DionResearchLLC From spreadsheet to deployed model in production. Is this expressed in months, weeks, days, hours?
With @visu_ai it could even be expressed in minutes. Check out @f_dion tomorrow morning Feb 4th, in #wsnc for a demo.
The challenge is that there is an embarrassment of riches. For software dev, I know what I'll use for @visu_ai & for consulting, so I avoid those, as I'll have plenty of practice through the year. Hence I will look at what I know little of, & probably be useful in the future.
Let's start at the top: Data Ingest
Forget ETL. @visu_ai does automated ingestion, & allows the user to tweak the choices it made & learns from modifications the user makes. It knows about common things like addresses, and things like VINs, UPCs, & 100s more.
#etl#dataquality
"At one time we had to advocate for the use of Python in businesses and prompt them to engage with the the SciPy/PyData ecosystem. Now Python data science is one of the most in-demand IT skills across every industry."
And also, congratulations, @pwang
@amontalenti Funny you mention this. I think email is underrated. I've set up @visu_ai to email results. Ie. Instead of waiting in front of the computer, it'll email me concise explanatory reports. Spend a few minutes feeding it data, hit a button then walk away and work on other stuff.