Read and Analyze 'Anki' Flashcard Databases.

ankiR

Comprehensive R toolkit for reading, analyzing, and visualizing Anki flashcard collection databases. 135 functions for collection analysis. For FSRS algorithm implementation, see r-fsrs.

Installation

# From r-universe (recommended)
install.packages("ankiR", repos = "https://cran.r-universe.dev")

# From GitHub
remotes::install_github("chrislongros/ankiR")

# Arch Linux (AUR)
# yay -S r-ankir

Quick Start

library(ankiR)

# One-liner overview
anki_quick_summary()
#> Anki: 5847 cards (4521 mature, 892 young, 434 new) | Due: 127 | Streak: 47 days | Retention: 91.2%

# Detailed report
anki_report()

# Collection health check (0-100 score)
anki_health_check()

Core Functions (Start Here)

Advanced users: See full function list below for forecasting, burnout detection, A/B testing, gamification, and more.

What's New in 0.6.0

Learning Efficiency Analysis

anki_learning_efficiency()    # ROI: retention per time spent
anki_retention_by_type()      # Retention by card type (cloze, basic, media)
anki_roi_analysis()           # Knowledge half-life extension per study minute

Forgetting Curve Fitting

# Fit your personal forgetting curve and compare to FSRS defaults
curve <- anki_fit_forgetting_curve()
anki_plot_forgetting_curve(curve)

Optimal Review Times

anki_best_review_times()      # Find when you learn best
anki_session_analysis()       # Analyze study session patterns
anki_simulate_session(30)     # Simulate a 30-minute session

Sibling & Interference Analysis

anki_sibling_analysis()       # How do sibling cards affect each other?
anki_interference_analysis()  # Find cards you confuse with each other
anki_weak_areas()             # Tags/decks with lowest retention

Actionable Recommendations

anki_card_recommendations()   # Leeches to rewrite, cards to unsuspend
anki_health_check()           # Comprehensive health score (0-100)
anki_summary()                # One-liner stats
anki_today()                  # Today's activity breakdown

Academic/Exam Preparation

anki_exam_readiness("2024-06-15")  # Will you be ready for your exam?
anki_coverage_analysis()           # % complete by topic
anki_study_priorities()            # What to study first
anki_study_plan("2024-06-15", hours_per_day = 2)

Enhanced FSRS Analysis

fsrs_compare_parameters()     # Compare your FSRS params to defaults
fsrs_memory_states()          # Current memory state for all cards
fsrs_decay_distribution()     # FSRS-6 per-card decay analysis

New Export Formats

anki_to_obsidian_sr()         # Obsidian Spaced Repetition plugin
anki_to_mochi()               # Mochi Cards JSON
anki_to_json()                # Full collection as JSON
anki_progress_report("html")  # Shareable progress report

Enhanced Search

# Advanced search operators
anki_search_enhanced("added:7 rated:3:1")      # Added in 7 days, rated Again in 3 days
anki_search_enhanced("prop:lapses>5 is:leech") # High-lapse leeches
anki_search_enhanced("re:^The\\s+")            # Regex search
anki_search_enhanced("deck:German OR deck:Spanish")

# Find similar cards with TF-IDF
anki_find_similar(1234567890, method = "tfidf")

Forecasting

# Monte Carlo forecasting (recommended for irregular study habits)
mc <- anki_forecast_monte_carlo(days_ahead = 30, n_sim = 1000)
mc$summary                           # Daily forecasts with CIs
mc$prob_above(day = 7, threshold = 100)  # P(>100 reviews on day 7)
anki_plot_monte_carlo(mc)            # Visualize with confidence bands

# Statistical forecasting (ARIMA, Holt-Winters, Seasonal)
anki_forecast_enhanced(method = "holt", days_ahead = 30)

# Compare methods
anki_compare_forecasts(days_ahead = 14)

# Scenario-based projections
anki_workload_projection(days = 30)

Visualizations

anki_plot_heatmap()           # Calendar heatmap
anki_plot_retention()         # Retention over time
anki_plot_forecast()          # Upcoming workload
anki_plot_difficulty()        # FSRS difficulty distribution
anki_plot_intervals()         # Interval distribution
anki_plot_hours()             # Reviews by hour
anki_plot_weekdays()          # Reviews by weekday
anki_plot_forgetting_curve()  # Personal forgetting curve

Time Series Analysis

anki_ts_retention(by = "week")
anki_ts_intervals(by = "week")
anki_ts_decompose()           # Trend + seasonal + residual
anki_ts_anomalies()           # Unusual study days
anki_ts_forecast()            # Forecast future reviews

Core Features

Search (Anki-like syntax)

anki_search("deck:Medical tag:cardiology")
anki_search("is:due -is:suspended prop:ivl>30")

anki_leeches()       # Problem cards
anki_mature()        # Cards with ivl >= 21
anki_due()           # Due for review

FSRS Data Reading

ankiR reads FSRS data from your Anki collection but does not implement the FSRS algorithm. For algorithm implementation, see fsrs-r-pure.

anki_cards_fsrs()                    # Get cards with FSRS parameters
fsrs_current_retrievability()        # Current memory state from DB
fsrs_forgetting_index()              # % below target retention
fsrs_get_parameters()                # Read FSRS parameters from collection
fsrs_compare_parameters()            # Compare your params to defaults
fsrs_memory_states()                 # Get memory states for all cards
fsrs_export_reviews()                # Export reviews for external optimizer
fsrs_prepare_for_optimizer()         # Prepare data for fsrs-r-pure

Comparative Analysis

anki_compare_periods()    # This month vs last month
anki_compare_decks()      # Side-by-side deck stats
anki_benchmark()          # Compare to FSRS averages

Export

anki_to_csv("Medical", "medical.csv")
anki_to_org("Medical", "medical.org")
anki_to_markdown("Medical", "medical.md", format = "obsidian")
anki_to_obsidian_sr("Medical", "medical_sr.md")
anki_to_mochi("Medical", "medical.json")
anki_to_json(output = "collection.json")
anki_progress_report(format = "html")

Interactive Dashboard

anki_dashboard()  # Launches Shiny app

Function Reference

Requirements

• R >= 4.1
• Anki 2.1+ (collection.anki2 format)
• Optional: ggplot2 (plots), shiny (dashboard)

Related Projects

• r-fsrs - FSRS algorithm implementation (Rust bindings)
• FSRS4Anki - FSRS for Anki

License

MIT.