Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

news.ycombinator.com

Post reply on HN
Print view View original

41–50 of 62 posts

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#41

Post by streetcat1 »

george3d6 wrote: post #38
streetcat1 wrote: post #36

Hi So I assume that you are doing hyperparameter search? Can you share what optimization method you are using for search (e.g. random, gp )? Also, is the search can be distributed in parallel to multi node ? And, if mindsdb is not part of the db, what happen if minddb fail ? Also, do you support automatic retraining? If yes, can you elaborate more?

> So I assume that you are doing hyperparameter search? Can you share what optimization method you are using for search (e.g. random, gp )? Short answer is optuna and ax but only sometimes. Long answer lead me down a rabbit whole and it's 10k+ words and a few experiments deep. If you're interested in this are specifically ping me, but I've got nothing concrete, however I like discussing it. A recent paper I saw that…

Wow. Thanks for the answer and for the paper !. I myself implemented this: https://arxiv.org/pdf/1810.05934.pdf in go.

The issue with retraining is that you need new labels (assume supervised ML). so I wonder what process do you use to get those.

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#42

Post by torrmal »

streetcat1 wrote: post #40
torrmal wrote: post #39

Earlier quoted context omitted.

These are amazing questions Streetcat, We do some hyperparameter search using Optuna, we may be moving to Ray Tune because it can be highly parallelized. If MindsDB fails, it depends on how various DBs manage federated storage, but essentially you will get a query error. Funny that you mention automatic retraining, people have been asking for this recently, we will be supporting a retrain_frequency parameter in the c…

I am actually working on a product in the same area (auto ml/ mlops ) as a non YC startup... We might be able partner. I am not sure how to reach you?

absolutely lets connect!! jorge at mindsdb

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#43

Post by adam_carrigan »

streetcat1 wrote: post #40
torrmal wrote: post #39

Earlier quoted context omitted.

These are amazing questions Streetcat, We do some hyperparameter search using Optuna, we may be moving to Ray Tune because it can be highly parallelized. If MindsDB fails, it depends on how various DBs manage federated storage, but essentially you will get a query error. Funny that you mention automatic retraining, people have been asking for this recently, we will be supporting a retrain_frequency parameter in the c…

I am actually working on a product in the same area (auto ml/ mlops ) as a non YC startup... We might be able partner. I am not sure how to reach you?

Send us an email - Adam at MindsDB.com and Jorge at MindsDB.com

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#44

Post by hodgesrm »

adam_carrigan wrote: post #29
hodgesrm wrote: post #28

Go MindsDB!! We've enjoyed working with the MindsDB team at Altinity. The integration with ClickHouse makes clever use of the MySQL protocol to implement models as queryable tables. For anybody interested in the specifics check out the following article: https://altinity.com/blog/machine-learning-models-as-tables . We will watch your career with great interest.

Thanks, it has been amazing working with you also.

Based on the feedback here it may soon be time to do a follow-up talk on MindsDB at a future ClickHouse meetup. :)

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#45

Post by michaelpb »

george3d6 wrote: post #25
robertlagrant wrote: post #24

Is it inside the database? It looks as though it's actually in a separate server, that is called by the database.

From the user perspective it's inside the database, you can run mindsdb in the backgrond,connect it to the database once, and then do everything from within the database (i.e. connecting with a sql client to your database server and issuing commands the same way you would query "normal" tables). From a technical perspective it's a separate server that communicates with the database through various mechanisms (e.g. fe…

Gotcha, yeah not who were replying too, but I read/skimmed most of your post first thinking was a new DBMS specifically for machine learning (especially given the name), which is a lot less interesting to me and I was about to close the tab. I much prefer the phrasing you used on the GitHub, I think it's more clear: "Predictive AI layer for existing databases"

Anyway, I'm seriously going to look into using this on a product for work, thanks for sharing & your work on this! We are in the process of triaging a few different features for a product in education (can chat more about it if your curious!), including some fairly bog-standard predictive text features. We are also trying to avoid strictly depending on any one proprietary PaaS, so this might just fit the bill!

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#47

Post by torrmal »

omneity wrote: post #46

Congrats on your launch Adam! What deployment scenarios do you foresee? (given the license is GPL, which is infamous in the business world)

Hi Omneity, thanks for asking. You can deploy Mindsdb on a container, no application that you build querying the Mindsdb server has licensing dependencies, as its essentially no different from using most opensource databases out there.

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#49

Post by ghego1 »

Looks super interesting! I'm quite literally the ideal user you describe, and I can't wait to try this out. If I can get any meaningful outputs from my databases thank to mindsdb I'll be super happy.

Also, good idea the hosting, I am currently struggling in training my models as I don't have the needed processing power, instead of bulding a new PC for ML tasks I would very likely happily consider hosting the heavy lifting. I'll keep an eye on the release.

Re: Launch HN: MindsDB (YC W20) – Machine Learning Inside Your Database

#50

Post by polskibus »

torrmal wrote: post #31
polskibus wrote: post #30

How does your product differ from MS SQL's integrated R, for someone who only needs MS SQL Server support?

this is a great question, I actually think that if your language is also R, MSsql r integration is a great option, what we bring to the table for MSSQL users in particular is more options as well as better performance for some types of problems like high cardinality on time-series, for example: predicting inventory for all products in a database taking into account all previous inventory as well as say marketing data…

Would your offering be slower because the data needs to be transferred outside sql server and then back to publish model results?
Post reply on HN