THE SUMFIZZ LIBRARY

Data science and machine learning evaluation calculators

Explore 8 free tools for data science and machine learning evaluation. Find classification confusion matrix metrics, classification brier score, regression residual diagnostics and more. Each calculator explains its inputs and method, with examples you can try using your own numbers.

THE SUMFIZZ LIBRARY

What are we working out?

8 calculators

⌕Browse all calculators →

Classification confusion matrix metrics

Binary classification with the positive class chosen by you.

Data science and machine learning evaluation

Classification Brier score

Binary Brier score uses the mean squared difference between predicted event probability and observed zero/one outcome; range zero to one.

Data science and machine learning evaluation

Regression residual diagnostics

Unweighted paired residuals are prediction minus observation; all rows have equal importance.

Data science and machine learning evaluation

Regression percentage error eligibility

MAPE excludes zero observed outcomes and uses absolute observed values as denominators.

Data science and machine learning evaluation

Ranking reciprocal-rank summary

Each query contributes reciprocal rank of its first relevant result; zero denotes no relevant result.

Data science and machine learning evaluation

Dataset class imbalance audit

Classes are mutually exclusive and exhaustive; every entered class must have observations.

Data science and machine learning evaluation

Model inference batch throughput

One fixed hardware/payload configuration; each measurement is end-to-end batch latency.

Data science and machine learning evaluation

Model training token budget

Expected token arithmetic with constant mean length and utilization; each epoch rounds up its own step count.

Data science and machine learning evaluation