scikit-learn

Machine learning in Python with scikit-learn. Use when working with supervised learning (classification, regression), unsupervised learning (clustering, dimensionality reduction), model evaluation, hyperparameter tuning, preprocessing, or building ML pipelines. Provides comprehensive reference documentation for algorithms, preprocessing techniques, pipelines, and best practices.

1.10x

Quality

75%

Does it follow best practices?

Impact

98%

1.10x

Average score across 6 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/scikit-learn/SKILL.md

Evaluation results

100%

Customer Churn Prediction

Pipeline construction with mixed data

Criteria

Without context

With context

Uses Pipeline

100%

ColumnTransformer for mixed data

100%

Numeric imputation strategy

100%

Categorical imputation strategy

100%

handle_unknown='ignore'

100%

Stratified split

100%

random_state set

100%

Fit on train only

100%

Meaningful step names

100%

Feature scaling present

100%

Uses sklearn Pipeline not manual

100%

Credit Card Fraud Detection

Hyperparameter tuning with imbalanced data

Criteria

Without context

With context

Uses Pipeline

100%

GridSearchCV or RandomizedSearchCV

100%

Double-underscore param notation

100%

StratifiedKFold or stratify in CV

100%

Imbalanced metric used

100%

Does NOT rely on accuracy alone

100%

class_weight or resampling

100%

n_jobs=-1 in search

100%

Stratified split

100%

random_state set

100%

best_estimator_ used

100%

10%

Customer Segmentation for Retail Analytics

Clustering analysis with optimal K selection

Criteria

Without context

With context

Scales before clustering

100%

Silhouette score for optimal K

100%

k-means++ init

100%

random_state set

100%

PCA for visualization

100%

Scatter plot saved

100%

Silhouette score reported

100%

Fit scaler on full data

100%

Cluster count reported

100%

Does NOT skip scaling

100%

88%

27%

House Price Estimation Model

Regression pipeline with persistence

Criteria

Without context

With context

Uses Pipeline

100%

TransformedTargetRegressor

100%

Target inverse transform

100%

RMSE reported

100%

MAE reported

100%

R2 reported

100%

joblib persistence

model.pkl exists

100%

Fit on train only

100%

random_state set

100%

results.json structure

75%

100%

10%

News Topic Classifier

Text classification with TF-IDF pipeline

Criteria

Without context

With context

TfidfVectorizer used

100%

stop_words parameter

100%

ngram_range parameter

100%

max_features parameter

100%

min_df or max_df parameter

100%

Uses Pipeline

100%

Stratified split

100%

classification_report used

100%

classifier.pkl saved

100%

Meaningful step names

100%

Retail Demand Forecasting Model

Temporal cross-validation for forecasting

Criteria

Without context

With context

TimeSeriesSplit used

100%

No random split for validation

100%

RMSE reported per model

100%

MAE reported per model

100%

Multiple models compared

100%

cross_validate or cross_val_score

62%

100%

cv_results.json structure

100%

Data is time-ordered

100%

random_state set

100%

Pipeline or fit-on-train

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: 086de41

Evaluated: about 1 month ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Customer Churn Prediction Credit Card Fraud Detection Customer Segmentation for Retail Analytics House Price Estimation Model News Topic Classifier Retail Demand Forecasting Model

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.