Package index • tidylearn

Reading data

One interface over files, databases and cloud sources. Every reader returns a tibble.

tl_read(): Read data from diverse sources
tl_read_bigquery(): Read from Google BigQuery
tl_read_csv(): Read a CSV file
tl_read_db(): Read from a DBI database connection
tl_read_dir(): Read all matching files from a directory
tl_read_excel(): Read an Excel file
tl_read_github(): Read from GitHub
tl_read_json(): Read a JSON file
tl_read_kaggle(): Read from Kaggle
tl_read_mysql(): Read from a MySQL/MariaDB database
tl_read_parquet(): Read a Parquet file
tl_read_postgres(): Read from a PostgreSQL database
tl_read_rdata(): Read an RData file
tl_read_rds(): Read an RDS file
tl_read_s3(): Read from Amazon S3
tl_read_sqlite(): Read from a SQLite database
tl_read_tsv(): Read a TSV file
tl_read_zip(): Read data from a zip archive

Modelling

Fit, split, evaluate and cross-validate. Reach the underlying model object through the $fit slot.

tl_model(): Create a tidylearn model
tl_split(): Split data into train and test sets
tl_prepare_data(): Data Preprocessing for tidylearn
tl_evaluate(): Evaluate a tidylearn model
tl_cv(): Cross-validation for tidylearn models
tl_compare_cv(): Compare models using cross-validation
tl_calc_classification_metrics(): Calculate classification metrics
tl_get_best_model(): Get the best model from a pipeline
tl_stratified_models(): Stratified Features via Clustering
tl_semisupervised(): Semi-Supervised Learning via Clustering
tl_anomaly_aware(): Anomaly-Aware Supervised Learning
tl_step_selection(): Perform stepwise selection on a linear model
tl_reduce_dimensions(): Integration Functions: Combining Supervised and Unsupervised Learning
tl_add_cluster_features(): Cluster-Based Features

Pipelines

Compose the steps above into something you can save and re-run.

tl_pipeline(): Create a modeling pipeline
tl_run_pipeline(): Run a tidylearn pipeline
tl_save_pipeline(): Save a pipeline to disk
tl_load_pipeline(): Load a pipeline from disk
tl_predict_pipeline(): Make predictions using a pipeline
tl_compare_pipeline_models(): Compare models from a pipeline

Tuning and AutoML

tl_tune_deep(): Tune a deep learning model
tl_tune_grid(): Tune hyperparameters for a model using grid search
tl_tune_nn(): Tune a neural network model
tl_tune_random(): Tune hyperparameters using random search
tl_tune_xgboost(): Tune XGBoost hyperparameters
tl_auto_ml(): Auto ML: Automated Machine Learning Workflow
tl_auto_interactions(): Find important interactions automatically
tl_default_param_grid(): Create pre-defined parameter grids for common models

Diagnostics

What the model got wrong, and which observations drove it.

tl_check_assumptions(): Check model assumptions
tl_diagnostic_dashboard(): Create a comprehensive diagnostic dashboard
tl_influence_measures(): Calculate influence measures for a linear model
tl_detect_outliers(): Detect outliers in the data
tl_interaction_effects(): Calculate partial effects based on a model with interactions
tl_test_interactions(): Test for significant interactions between variables
tl_test_model_difference(): Perform statistical comparison of models using cross-validation
tl_explore(): Exploratory Data Analysis Workflow
tl_dashboard(): Create interactive visualization dashboard for a model

Unsupervised learning

Clustering, ordination and dimension reduction, each returning a tibble rather than a fitted object you have to take apart.

tidy_apriori(): Tidy Apriori Algorithm
tidy_clara(): Tidy CLARA (Clustering Large Applications)
tidy_cutree(): Cut Hierarchical Clustering Tree
tidy_dbscan(): Tidy DBSCAN Clustering
tidy_dendrogram(): Plot Dendrogram
tidy_dist(): Tidy Distance Matrix Computation
tidy_gap_stat(): Tidy Gap Statistic
tidy_gower(): Gower Distance Calculation
tidy_hclust(): Tidy Hierarchical Clustering
tidy_kmeans(): Tidy K-Means Clustering
tidy_knn_dist(): Compute k-NN Distances
tidy_mds(): Tidy Multidimensional Scaling
tidy_mds_classical(): Classical (Metric) MDS
tidy_mds_kruskal(): Kruskal's Non-metric MDS
tidy_mds_sammon(): Sammon Mapping
tidy_mds_smacof(): SMACOF MDS (Metric or Non-metric)
tidy_pam(): Tidy PAM (Partitioning Around Medoids)
tidy_pca(): Tidy Principal Component Analysis
tidy_pca_biplot(): Create PCA Biplot
tidy_pca_screeplot(): Create PCA Scree Plot
tidy_rules(): Convert Association Rules to Tidy Tibble
tidy_silhouette(): Tidy Silhouette Analysis
tidy_silhouette_analysis(): Silhouette Analysis Across Multiple k Values
augment_dbscan(): Augment Data with DBSCAN Cluster Assignments
augment_hclust(): Augment Data with Hierarchical Cluster Assignments
augment_kmeans(): Augment Data with K-Means Cluster Assignments
augment_pam(): Augment Data with PAM Cluster Assignments
augment_pca(): Augment Original Data with PCA Scores
optimal_clusters(): Find Optimal Number of Clusters
optimal_hclust_k(): Determine Optimal Number of Clusters for Hierarchical Clustering
compare_clusterings(): Compare Multiple Clustering Results
compare_distances(): Compare Distance Methods
calc_validation_metrics(): Calculate Cluster Validation Metrics
calc_wss(): Calculate Within-Cluster Sum of Squares for Different k
explore_dbscan_params(): Explore DBSCAN Parameters
get_pca_loadings(): Get PCA Loadings in Wide Format
get_pca_variance(): Get Variance Explained Summary
suggest_eps(): Suggest eps Parameter for DBSCAN
standardize_data(): Standardize Data
create_cluster_dashboard(): Create Summary Dashboard

Association rules

inspect_rules(): Inspect Association Rules
filter_rules_by_item(): Filter Rules by Item
find_related_items(): Find Related Items
recommend_products(): Generate Product Recommendations
summarize_rules(): Summarize Association Rules
visualize_rules(): Visualize Association Rules

Plots

ggplot2 objects, returned rather than printed, so you can keep editing them.

tl_plot_cv_comparison(): Plot comparison of cross-validation results
tl_plot_cv_results(): Plot cross-validation results
tl_plot_deep_architecture(): Plot deep learning model architecture
tl_plot_deep_history(): Plot deep learning model training history
tl_plot_gain(): Plot gain chart for a classification model
tl_plot_importance_comparison(): Plot feature importance across multiple models
tl_plot_importance_regularized(): Plot variable importance for a regularized model
tl_plot_influence(): Plot influence diagnostics
tl_plot_interaction(): Plot interaction effects
tl_plot_intervals(): Create confidence and prediction interval plots
tl_plot_lift(): Plot lift chart for a classification model
tl_plot_model_comparison(): Plot model comparison
tl_plot_nn_architecture(): Plot neural network architecture
tl_plot_nn_tuning(): Plot neural network training history
tl_plot_partial_dependence(): Plot partial dependence for tree-based models
tl_plot_regularization_cv(): Plot cross-validation results for a regularized model
tl_plot_regularization_path(): Plot regularization path for a regularized model
tl_plot_svm_boundary(): Plot SVM decision boundary
tl_plot_svm_tuning(): Plot SVM tuning results
tl_plot_tree(): Plot a decision tree
tl_plot_tuning_results(): Plot hyperparameter tuning results
tl_plot_xgboost_importance(): Plot feature importance for an XGBoost model
tl_plot_xgboost_shap_dependence(): Plot SHAP dependence for a specific feature
tl_plot_xgboost_shap_summary(): Plot SHAP summary for XGBoost model
tl_plot_xgboost_tree(): Plot XGBoost tree visualization
plot_cluster_comparison(): Create Cluster Comparison Plot
plot_cluster_sizes(): Plot Cluster Size Distribution
plot_clusters(): Plot Clusters in 2D Space
plot_dendrogram(): Plot Dendrogram with Cluster Highlights
plot_distance_heatmap(): Create Distance Heatmap
plot_elbow(): Create Elbow Plot for K-Means
plot_gap_stat(): Plot Gap Statistic
plot_knn_dist(): Plot k-NN Distance Plot
plot_mds(): Plot MDS Configuration
plot_silhouette(): Plot Silhouette Analysis
plot_variance_explained(): Plot Variance Explained (PCA)
tl_xgboost_shap(): Generate SHAP values for XGBoost model interpretation

Tables

Formatted gt tables for reporting.

tl_table(): Create formatted tables for tidylearn models
tl_table_clusters(): Formatted cluster summary table
tl_table_coefficients(): Formatted model coefficients table
tl_table_comparison(): Compare multiple models in a formatted table
tl_table_confusion(): Formatted confusion matrix table
tl_table_importance(): Formatted feature importance table
tl_table_loadings(): Formatted PCA loadings table
tl_table_metrics(): Formatted evaluation metrics table
tl_table_variance(): Formatted PCA variance explained table

Compute and cloud

Where the work runs, and what it is allowed to reach.

tl_check_gpu(): Detect local GPU availability for tidylearn methods
tl_compute_advisor(): Advise on the best compute tier for a tidylearn fit
tl_transfer_learning(): Transfer Learning Workflow
tl_cloud_allow_host(): Allow an additional host for cloud uploads in this R session
tl_cloud_allowed_hosts(): Hosts tidylearn will currently upload to
tl_cloud_consent(): Grant or revoke cloud upload consent for this R session
tl_cloud_jobs(): Cloud jobs submitted in this R session

Utilities

tl_version(): Get tidylearn version information
`%>%`: Pipe operator

Methods

What print(), plot(), predict() and summary() do to each object the package returns.

plot(<tidylearn_eda>): Plot EDA results
plot(<tidylearn_model>): Plot method for tidylearn models
predict(<tidylearn_model>): Predict using a tidylearn model
predict(<tidylearn_stratified>): Predict from stratified models
predict(<tidylearn_transfer>): Predict with transfer learning model
print(<tidy_apriori>): Print Method for tidy_apriori
print(<tidy_dbscan>): Print Method for tidy_dbscan
print(<tidy_gap>): Print Method for tidy_gap
print(<tidy_hclust>): Print Method for tidy_hclust
print(<tidy_kmeans>): Print Method for tidy_kmeans
print(<tidy_mds>): Print Method for tidy_mds
print(<tidy_pam>): Print Method for tidy_pam
print(<tidy_pca>): Print Method for tidy_pca
print(<tidy_silhouette>): Print Method for tidy_silhouette
print(<tidylearn_automl>): Print auto ML results
print(<tidylearn_compute_advice>): Print method for tidylearn_compute_advice objects
print(<tidylearn_data>): Print a tidylearn_data object
print(<tidylearn_eda>): Print EDA results
print(<tidylearn_gpu_check>): Print method for tidylearn_gpu_check objects
print(<tidylearn_model>): Print method for tidylearn models
print(<tidylearn_pipeline>): Print a tidylearn pipeline
summary(<tidylearn_model>): Summary method for tidylearn models
summary(<tidylearn_pipeline>): Summarize a tidylearn pipeline

Concepts

Topic pages covering a whole area rather than a single function. Start here when you want the shape of something before the arguments.

tidylearn-classification: Classification Functions for tidylearn
tidylearn-cloud-consent: Cloud data-egress consent for tidylearn
tidylearn-cloud-cost: Cost controls for tidylearn cloud compute
tidylearn-cloud-endpoint: Cloud endpoint resolution for tidylearn
tidylearn-cloud-serialize: Model serialisation for tidylearn cloud compute
tidylearn-core: tidylearn: A Unified Tidy Interface to R's Machine Learning Ecosystem
tidylearn-deep-learning: Deep Learning for tidylearn
tidylearn-diagnostics: Advanced Diagnostics Functions for tidylearn
tidylearn-interactions: Interaction Analysis Functions for tidylearn
tidylearn-metrics: Metrics Functionality for tidylearn
tidylearn-model-selection: Model Selection Functions for tidylearn
tidylearn-neural-networks: Neural Networks for tidylearn
tidylearn-pipeline: Model Pipeline Functions for tidylearn
tidylearn-read-backends: Data Reading Backends for tidylearn
tidylearn-read: Data Reading Functions for tidylearn
tidylearn-regression: Regression Functions for tidylearn
tidylearn-regularization: Regularization Functions for tidylearn
tidylearn-svm: Support Vector Machines for tidylearn
tidylearn-tables: Table Functions for tidylearn
tidylearn-trees: Tree-based Methods for tidylearn
tidylearn-tuning: Hyperparameter Tuning Functions for tidylearn
tidylearn-visualization: Visualization Functions for tidylearn
tidylearn-workflows: High-Level Workflows for Common Machine Learning Patterns
tidylearn-xgboost: XGBoost Functions for tidylearn