orgoca/CompileML

COBOL export: emit calibrated PD and reason codes about 3 hours ago

enhancement help wanted

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Document compatibility with WOE-transformed features about 3 hours ago

documentation good first issue credit-risk

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

FAQ: why not PMML, ONNX, or m2cgen? about 3 hours ago

documentation good first issue

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Champion/challenger: diff two artifacts at the band and decision level about 3 hours ago

enhancement help wanted

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Leaf-time exact attribution: remove the O(p²) explanation cost about 3 hours ago

enhancement help wanted

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Report distillation retention by segment, not just globally about 3 hours ago

enhancement help wanted credit-risk

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Fair lending: disparate impact testing and attribution-level disparity analysis about 3 hours ago

enhancement help wanted credit-risk

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Stability monitoring: PSI, CSI, and score-distribution drift about 3 hours ago

enhancement help wanted credit-risk

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Support monotone feature constraints in the distilled whitebox about 3 hours ago

enhancement help wanted credit-risk

orgoca/CompileML

0

Compile tree-ensemble models into deterministic, auditable decision artifacts — score, calibrated PD, risk bands, and adverse-action reason codes in one hashed JSON that produces identical integers in Python, SQL, and COBOL.

Python

#cobol#credit-risk#credit-scoring#explainable-ai#fintech#interpretability#lightgbm#machine-learning#mlops#model-deployment#model-governance#xgboost

Open Issues Need Help