Dan Snow
Dan Snow
**What is the URL of the website or web application you want to add?** https://sno.ws **What is the total size of data transferred while loading your URL in a web...
The comps algorithm currently takes a _long_ time to run. It takes 12 hours even with a reduced search space. However, the algorithm is _highly_ parallelizable. It's possible that using...
Visual explainer of how multiple cards/PINs are aggregated into a single value in the res model.
Currently, in order to run reports you need both the `input/` and `output/` data to be aligned (matching the metadata of your target run). This can be cumbersome when testing...
They share _most_ of the same code but differ in critical ways. It may be worthwhile to merge them just to reduce duplication effort.
We should create a dedicated "error" model to predict either the % error or absolute error from the main model. This would be useful for a number of reasons; it...
While I think the model's overall time-trending is performing pretty well, it never hurts to test some tweaks/improvements. One thing we can do to check the model's performance/understanding of time...
Class 212 buildings are mixed residential and commercial space, prototypically a multi-family small apartment building with ground floor commercial. These buildings are incredibly tough to value and have historically been...
To improve model reproducibility, we should create Docker container images containing all the necessary dependencies to run all currently tagged versions of each model. This would involve building a Docker...
In order to make the models more reproducible, we should also export the compiler settings used for compiling LightGBM/XGBoost from source. This should be included in the export sent to...