Random forest implementation
Hi,
This question may be better suited to someone in technology, but I was wondering about the feasibility of doing the following. We are working on a project in Analytics that seems very computationally demanding. Is this something we could potentially do in production or is this just too much?
Here are the parameters:
- Each person will have about 100 data points that will need to be refreshed periodically, every month.
- We need to fit one random forest model per person.
- There are 50,000+ users and thus 50,000+ models.
- We need to score future data points based on the tree model for each user.
- We need to both train the model in production and also score the model in production.
Thank you.
This question may be better suited to someone in technology, but I was wondering about the feasibility of doing the following. We are working on a project in Analytics that seems very computationally demanding. Is this something we could potentially do in production or is this just too much?
Here are the parameters:
- Each person will have about 100 data points that will need to be refreshed periodically, every month.
- We need to fit one random forest model per person.
- There are 50,000+ users and thus 50,000+ models.
- We need to score future data points based on the tree model for each user.
- We need to both train the model in production and also score the model in production.
Thank you.
- jcma
- Posts: 5
- Joined: Fri Dec 11, 2015 3:36 pm
1 post
• Page 1 of 1
Who is online
Users browsing this forum: No registered users and 1 guest