Skip to content
#

treshold

Here are 6 public repositories matching this topic...

Banks lose customers silently. Without a model, the retention team has no idea who is about to leave until it is too late. Goal: flag at-risk customers early enough to act. Built a full ML pipeline on 10K customers. Result: 83% Recall, meaning 8 out of 10 churners identified before they leave, giving the retention team an actionable list.

  • Updated Aug 21, 2026
  • Jupyter Notebook

Proyek ini bertujuan untuk melakukan segmentasi area penyakit pada daun menggunakan pendekatan berbasis pengolahan citra digital (Image Processing). Fokus utama adalah untuk meningkatkan akurasi segmentasi dengan beberapa metode preprocessing, seperti CLAHE dan Gaussian Blur, serta evaluasi kinerja model menggunakan metrik IoU dan Dice Score

  • Updated Aug 6, 2025
  • Jupyter Notebook

Improve this page

Add a description, image, and links to the treshold topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the treshold topic, visit your repo's landing page and select "manage topics."

Learn more