Confound-matched controls for classifier claims: 1:1 matching, bootstrap AUROC, anchored verdicts, and a refusal when the control is vacuous.
-
Updated
Oct 14, 2026 - Python
#
classifier-evaluation
Here are 6 public repositories matching this topic...
Built a university Pattern Recognition project that compares classical and machine-learning classifiers through feature selection, PCA/LDA transformations, hyperparameter tuning, and cross-validated evaluation on a real binary classification dataset.
pca lda hyperparameter-tuning svm-classifier knn-classification bayesian-classifier roc-auc kruskal-wallis lda-classification classifier-evaluation
-
Updated
Mar 5, 2026 - Python
The project focuses on analyzing neural activity data to classify neuron types (spiny and aspiny). It integrates unsupervised learning methods (PCA, Autoencoders) and supervised learning models (Logistic Regression, MLP) to build accurate classifiers that effectively analyze neurons' electrical responses.
pca-analysis supervised-learning logistic-regression data-compression mlp perceptron gradient-descent unsupervised-learning neuron autoencoders stochastic-gradient-descent neuronal-network roc-auc mlp-networks neural-classification high-dimensional-neural-datasets classifier-evaluation cortical-neurons 2d-and-3d-visualizations
-
Updated
Jan 9, 2025 - Jupyter Notebook
BigQuery SQL workflow for estimating classifier recall using stratified sampling and weighted evaluation.
-
Updated
May 26, 2026
A semantic cache is a classifier that is wrong silently. This is the safety gate (threshold AND deterministic vetoes) and the calibration harness that measures where its safety ends — which concluded the cache should not be turned on.
python caching prometheus calibration api-proxy cache-invalidation fastapi llm prompt-caching semantic-cache llm-inference classifier-evaluation openai-compatible
-
Updated
Aug 5, 2026 - Python
Agent-driven, independent reproduction of ICML 2026 paper "Instance-Level Costs for Nuanced Classifier Evaluation" (NEC metric) on public Jigsaw, Turkey and NHANES data. On Jigsaw (TF-IDF logistic regression), error rate is 2.92× NEC (paper ~3×).
machine-learning metrics scikit-learn reproducibility cost-sensitive-learning paper-reproduction classifier-evaluation icml-2026
-
Updated
Oct 4, 2026 - HTML
Add this topic to your repo
To associate your repository with the classifier-evaluation topic, visit your repo's landing page and select "manage topics."