This is a ShmuggingFace review mock. It is not Hugging Face, Kaggle, or a real dataset release.
😏
🚀LeadForge Lead Scoring v1 — Advanced
<> Code ⇩ Download

😏leadforge-dev · UPDATED 2026-05-24 AGO

LeadForge Lead Scoring v1 — Advanced

Advanced difficulty · 5,000 leads · ~8% conversion rate · LR AUC 0.886 (5-seed median)

⇩ Download
LeadForge Lead Scoring v1 — Advanced5,000 rows

Discussion

  • What is `snapshot_day = 30` and how does it affect which features are valid at inference time?

    #1 opened for review by leadforge-dev

    1 replies
  • Is `total_touches_all` a safe feature or a time-window leakage trap?

    #2 opened for review by leadforge-dev

    2 replies
  • LR and GBM AUCs are very close across tiers — does relational feature engineering help?

    #3 opened for review by leadforge-dev

    3 replies
  • How would you set a probability threshold for a team that can only work 50 leads per week?

    #4 opened for review by leadforge-dev

    4 replies
  • What happens to AUC when you evaluate on a chronological hold-out instead of a random split?

    #5 opened for review by leadforge-dev

    5 replies