Andrew DalPino

Hi @yunatamos this is far too little information to make a recommendation based upon. Perhaps join our telegram channel and discuss in there. We have a few engineers working on...

Hello @zenichanin thank you for the great question! I did some experimenting on my own with Random Forest and the Divorce dataset and indeed RF performs poorly using continuous feature...

You're welcome @zenichanin, also Rubix ML 1.0.0 will not use Flysystem so you can try 1.0.0-beta when it's released as well.

Sure @zenichanin, I will do my best to help Can you post your code? That would be very helpful to me. Thank you!

> And my actual dataset file is a CSV file with the fields mentioned in the earlier post, with 1st row being the column names. Make sure to set the...

> Just found [rubixml.com](https://rubixml.com/), which might be interesting. We're currently working on GPU acceleration in Rubix ML with the [Tensor](https://github.com/RubixML/Tensor) extension so this feature might be practical to implement in...

Hahahhaa hilarious @Dibbyo456 ... ohhhhhh machine learning you! But yeah there can be many reasons for poor real-world performance even on seemingly obvious (for us) examples. What is the validation...

Sounds good @Dibbyo456, I reminded the Zephir team of the issue just now. Unfortunately, due to a lack of resources we haven't had a chance to improve the demo model...

Yes, there are. The first thing you can do is to use the TokenHashingVecotrizer instead of WordCountVectorizer. TokenHashingVecotrizer is a low-memory implementation of a bag-of-words vectorizer/embedder that uses a hashing...

Great question @samuraya! Servers are designed for inference only. Imagine having 10 servers running the same model behind a load balancer. It would be hard, resource consuming, and inconsistent to...