-
docs: clarify references to upstream works and update official paper citations
-
Sanitize absolute server paths to relative paths
-
Sanitize absolute server paths to relative paths
-
Sanitize absolute server paths to relative paths
-
Sanitize absolute server paths to relative paths
-
Sanitize absolute server paths to relative paths
-
Sanitize absolute server paths to relative paths
-
Add Turkish Referanslar ve Kaynakça section (Sybil, NLST)
-
Clean up NLST reference text: remove duplicate prefix and remove Hugging Face repo mentions
-
Remove incorrect hypotheses (pixel compression, TorchIO, ETL) and establish Network Interruption / Missing Slices as root cause
-
Replace corporate phrasing: 1. Giriş ve Temel Bulgular
-
Style update: Left-align all paragraphs and bold all headings & bullet items
-
Style update: Left-align all paragraphs and bold all headings & bullet items
-
Update NLST_Sybil_Kesit_Silme_Hassasiyet_Analiz_Raporu.docx with SB 9-series comparison data in Section 7
-
Update Sybil_40_Hasta_Drift_Analizi_v2_22092026.docx with SB 9-series cropping analysis & network drop explanation
-
Upload Sybil_40_Hasta_Drift_Analizi_v2_22092026.docx
-
Add README for cropping_analysis
-
Upload slice_drop_worker.py to cropping_analysis
-
Upload launch_parallel_slice_drop.py to cropping_analysis
-
Upload generate_figures.py to cropping_analysis
-
Upload create_slice_drop_report.py to cropping_analysis
-
Upload fig5_negative_vs_positive_sensitivity.png to cropping_analysis
-
Upload fig4_patient_trajectory_spaghetti.png to cropping_analysis
-
Upload fig3_clinical_threshold_flips.png to cropping_analysis
-
Upload fig2_boxplot_score_shifts.png to cropping_analysis
-
Upload fig1_dose_response_curve.png to cropping_analysis
-
Upload slice_drop_all_2965_results.json to cropping_analysis
-
Upload slice_drop_analysis_ready.csv to cropping_analysis
-
Upload NLST_Sybil_Kesit_Silme_Hassasiyet_Analiz_Raporu.docx to cropping_analysis
-
Update drift report: record individual Sybil models comparison (ruled out), focus on pixel/ETL and D1/D2 expansion
-
Rewrite Conclusion & Analysis: rule out Z-axis inversion and multi-series, establish single-model vs ensemble root cause, and mandate D1/D2 sample expansion
-
Update drift report: add 1800s API reload architecture and expand D1/D2 sample re-test requirement
-
Add root-cause analysis (determinism, baseline hazard, Z-axis inversion, selection bias) to drift report
-
Upload Sybil_40_Hasta_Drift_Analizi_v1_22092026.docx
-
Upload NLST_Sybil_Ham_Skor_Boylamsal_Analiz_Raporu.docx with huggingface_hub
-
Upload NLST_Sybil_Boylamsal_Analiz_Raporu.docx with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload run_raw_sybil_survival_analysis.py with huggingface_hub
-
Upload fig6_raw_dynamic_auroc_horizon.png with huggingface_hub
-
Upload fig5_raw_kaplan_meier_survival.png with huggingface_hub
-
Upload fig4_raw_spaghetti_paired_cancer.png with huggingface_hub
-
Upload fig3_raw_cancer_waterfall_plot.png with huggingface_hub
-
Upload fig2_raw_delta_distribution_by_subgroup.png with huggingface_hub
-
Upload fig1_raw_longitudinal_trajectories.png with huggingface_hub
-
Upload raw_sybil_summary.json with huggingface_hub
-
Upload NLST_Sybil_Ham_Skor_Boylamsal_Analiz_Raporu.docx with huggingface_hub
-
Upload NLST_Sybil_Boylamsal_Analiz_Raporu.docx with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload run_full_sybil_survival_analysis.py with huggingface_hub
-
Upload fig6_dynamic_auroc_horizon.png with huggingface_hub