goji_berries
Of Markets and Marriages: Review of Recent Work on the Causes of US Poverty
On-Writing – Writing Tips, Tricks, and Tools
Ethnicolr2 – Ethnicolr implementation with new models in PyTorch
Toss-up – Data and writeups from an analysis of winning the toss in cricket
Re:
on Saturday May 17, 2025 @05:27PM (#44017094)
Now, imagine I use GenAI to flood the Internet with 100,000 pages praising Mr. Fox. These aren't crude spam pages—they're well-written articles with proper grammar, coherent arguments, and seemingly legitimate citations. Each page offers minor variations on the same theme: "Mr. Fox is innovative," "Mr. Fox shows exceptional leadership," "Studies confirm Mr. Fox's approach is effective."
Pastations: Is It Time to Move Beyond Presentations?
New Python Package for Advanced Calibration Models (already has 3k downloads)
Market Failure When the Cost of Misclassification Is Higher for the Customer
Re:Market Failure When the Cost of Misclassification Is Higher for the...
on Tuesday May 13, 2025 @09:46PM (#43979863)
This leads to systematic under-investment in accuracy since decision-makers optimize for visible costs (bad hires, defaults) while externalizing invisible ones (missed talent, foregone profits).
Notary: PyPI Digital Attestation Shield Generation Service
Re:Notary: PyPI Digital Attestation Shield Generation Service
on Tuesday May 13, 2025 @02:32PM (#43976129)
Uses Sigstore/PyPI integrity API
When Cost of Misclassification Is Higher for the Customer Than for the Business
Re:When Cost of Misclassification Is Higher for the Customer Than for ...
on Monday May 12, 2025 @07:13PM (#43968323)
Consider a bank making decisions about loans. For the bank, making lending decisions optimally means reducing prediction errors*(cost of errors) minus the cost of making predictions (Keeping things simple here). The cost of any one particular error, especially the denial of a loan when eligible, is typically small for the bank, but very consequential for the applicant. So the applicant may be willing to pay the bank money to increase the accuracy of their decisions.
Ethnicolr_v2 – Replication materials and ms for name to race models
Clean-names – Deduplicate and parse list of `dirty names'
Not Recommended: Why Current Content Recommendation Systems Fail Us
Re:Not Recommended: Why Current Content Recommendation Systems Fail Us
on Saturday May 10, 2025 @01:16PM (#43947249)
Let's build better products.
Parse_searchable_rolls – Parse Searchable Electoral Rolls
A single page on DoorDash can make upward of 1000 gRPC calls
Re:Microservices Are a Tax Your Startup Probably Can't Afford
on Thursday May 08, 2025 @12:35PM (#43927930)
here's my take on this issue: https://www.gojiberries.io/building-together-separately-chal...
Parse_searchable_rolls – Parse Searchable Electoral Rolls
Extract Useful Information from Tables for LLMs
Cell Mates: Extracting Useful Information from Tables for LLMs
Add Related Repositories on GitHub
Adjacent: Add Related Repositories to Readme on GitHub
Re:Rmcp – R MCP Server
on Monday April 28, 2025 @03:44PM (#43825264)
I plan to expand it to support most of R functionality with the focus on its conventional use case, which is statistical analysis