The Premier Online Community for Real Estate Appraisers!
-
Welcome to AppraisersForum.com, the premier online
community for the discussion of real estate appraisal. Register a free account to be able to post and unlock additional forums and features.
A.I. Stuff
- Thread starter Digger88
- Start date
Elliott
Elite Member
- Joined
- Apr 23, 2002
- Professional Status
- Certified General Appraiser
- State
- Oregon
I asked Bard if Bitcoin was adopted by the US would that then end ransomware by hackers because it would be 'traceable' and it said, nope. So besides not having any 'backing' asset, an energy waster, having a $1 transaction charge (compared to $.35), and Charlie Unger saying it was 'rat poision', there is another reason it should not be adopted.
RCA
Elite Member
- Joined
- Jun 27, 2017
- Professional Status
- Certified General Appraiser
- State
- California
And now there is Apple's new "Ferret."
Using my RCA method, which ranks 100-200 market area comps based on residuals - which represent primarily quality/condition/aesthetic attributes, Ferret can be trained to objectively rank the subject based on photographs and give it a percentage CQA (Condition/Quality/Aesthetics) score indicating where it falls in the ranking against comps.
Caution: You will for a long time to come need an experienced appraiser who is very familiar with the subject market area to supervise Ferret and review the results. There is too much system entropy in many market areas to assume that photographs can properly model all price variables. Yet, the result should be useful support.
github.com
Notes:
1. Ferret was trained on 8 of these $17,209.99 graphics cards:
https://www.cdw.com/product/pny-nvidia-a100-80gb-pcie-gen-4-graphic-card/7065275#TS
2. You will need at least one of these to train Ferret on your own set of data.
-> Hopefully the price comes down. But more likely you are going to have to use a Cloud Ferret service, once one becomes available.
Using my RCA method, which ranks 100-200 market area comps based on residuals - which represent primarily quality/condition/aesthetic attributes, Ferret can be trained to objectively rank the subject based on photographs and give it a percentage CQA (Condition/Quality/Aesthetics) score indicating where it falls in the ranking against comps.
Caution: You will for a long time to come need an experienced appraiser who is very familiar with the subject market area to supervise Ferret and review the results. There is too much system entropy in many market areas to assume that photographs can properly model all price variables. Yet, the result should be useful support.
GitHub - apple/ml-ferret
Contribute to apple/ml-ferret development by creating an account on GitHub.
github.com
Notes:
1. Ferret was trained on 8 of these $17,209.99 graphics cards:
https://www.cdw.com/product/pny-nvidia-a100-80gb-pcie-gen-4-graphic-card/7065275#TS
2. You will need at least one of these to train Ferret on your own set of data.
-> Hopefully the price comes down. But more likely you are going to have to use a Cloud Ferret service, once one becomes available.