Add March Machine Learning Mania 2026 Kaggle submission notebook

[Copilot]

Adds a beginner-friendly, copy-paste-ready Kaggle notebook for generating a valid baseline submission to the March Machine Learning Mania 2026 competition, covering both Men's and Women's NCAA tournament predictions.

Changes

• New notebook: March_ML_Mania_2026_Kaggle_Submission_Marktechpost.ipynb
  • Loads SampleSubmissionStage2.csv (2026 season, all possible matchups) from the standard Kaggle input path
  • Fills Pred = 0.5 as a valid baseline (coin-flip; scores 0.0 pre-tournament per competition rules)
  • Runs sanity checks: ID format (SSSS_XXXX_YYYY), predictions in [0, 1], no duplicate IDs
  • Saves submission.csv to /kaggle/working/
  • Includes a runnable cell for the Kaggle CLI submit command

Submit command (included in notebook)

kaggle competitions submit \
    -c march-machine-learning-mania-2026 \
    -f /kaggle/working/submission.csv \
    -m "Baseline 0.5 submission"

Notes

• Pred = probability the lower TeamID team wins (per competition spec)
• Men's TeamIDs: 1000–1999; Women's TeamIDs: 3000–3999 — no overlap, single submission covers both
• Replace submission["Pred"] = 0.5 with model output to improve beyond baseline

---

🔒 GitHub Advanced Security automatically protects Copilot coding agent pull requests. You can protect all pull requests by enabling Advanced Security for your repositories. Learn more about Advanced Security.