Skip to content

List of 100 Top MicroSims

This is a list of the 100 most important MicroSims for a Data Science course according to GPT-5. Each MicroSim is described in a level-2 header with the description, learning goals and input controls in level 3 headers. We will have agents generate these MicroSims one at a time.

Note that if students don't have a strong statistics background, they should review them.

1. Exploring Data Points

Description

Students click to add or remove points on a 2D scatter plot, instantly seeing the effect on the overall distribution.

Learning Goals

  • Recognize individual observations in a dataset

  • Understand x--y coordinate representation

  • See how adding/removing points changes data shape

Input Controls

  1. Add Point (click on canvas)

  2. Remove Point (click existing point)

  3. Clear All Points (button)

2. Histogram Builder

Description

Students adjust bin sizes to see how histograms change, revealing over-smoothing and under-smoothing effects.

Learning Goals

  • Understand bins and frequencies

  • Relate bin size to data detail retention

Input Controls

  1. Bin Size Slider

  2. Dataset Selector (dropdown: normal, uniform, skewed)

  3. Toggle Grid Lines (checkbox)

3. Mean and Median Explorer

Description

Drag points along a number line to see how mean and median shift differently.

Learning Goals

  • Differentiate mean vs. median

  • Observe robustness of median to outliers

Input Controls

  1. Drag Points (mouse)

  2. Add Outlier (button)

  3. Reset Points (button)

4. Correlation Playground

Description

Students drag clusters of points to adjust correlation, watching the correlation coefficient update in real-time.

Learning Goals

  • Visualize correlation strength and direction

  • Understand positive, negative, and zero correlation

Input Controls

  1. Drag Cluster (mouse)

  2. Add Noise (slider)

  3. Show Best Fit Line (toggle)

5. Least Squares Line Fitter

Description

Adjust slope and intercept manually to minimize sum of squared errors, with real-time residual visualization.

Learning Goals

  • Understand slope, intercept, and residuals

  • Experience trial-and-error fitting

Input Controls

  1. Slope Slider

  2. Intercept Slider

  3. Toggle Residual Squares (checkbox)

6. R² Intuition Builder

Description

Manipulate data spread around a fitted line to see how R² changes.

Learning Goals

  • Understand coefficient of determination

  • Relate R² to model fit quality

Input Controls

  1. Noise Level Slider

  2. Number of Points Slider

  3. Reset Dataset (button)

7. Train-Test Split Visualizer

Description

Randomly split a dataset and see how train/test points differ in model performance.

Learning Goals

  • Understand importance of splitting data

  • See overfitting risk when test set is too small

Input Controls

  1. Train/Test Ratio Slider

  2. Resample Dataset (button)

  3. Model Complexity Slider

8. Cross-Validation Simulator

Description

Animate k-fold cross-validation, showing shifting train/test subsets and aggregated scores.

Learning Goals

  • Understand cross-validation mechanics

  • See benefits over single train-test split

Input Controls

  1. Number of Folds Slider

  2. Dataset Size Slider

  3. Play/Pause Animation (button)

9. Overfitting vs. Underfitting Explorer

Description

Adjust polynomial degree to see bias--variance trade-off on train vs. test errors.

Learning Goals

  • Recognize overfitting and underfitting patterns

  • Connect complexity to generalization

Input Controls

  1. Polynomial Degree Slider

  2. Noise Level Slider

  3. Toggle Error Curves (checkbox)

10. $2Multiple Regression Plane

Description

Manipulate two independent variables in 3D space to see a regression plane fit to data.

Learning Goals

  • Visualize multivariate linear regression

  • See plane adjustment with variable changes

Input Controls

1. Rotate View (mouse drag)

  1. Noise Level Slider

  2. Add/Remove Points (click)

11. Residuals Heatmap Viewer

Description

Color-code residuals on a scatter plot to identify patterns and non-linearity.

Learning Goals

  • Understand residual analysis

  • Detect systematic errors in model predictions

Input Controls

  1. Model Complexity Slider

  2. Toggle Residual Colors

  3. Noise Level Slider

12. Distribution Shape Explorer

Description

Morph between uniform, normal, skewed, and bimodal distributions.

Learning Goals

  • Identify common data distributions

  • Understand skewness and kurtosis visually

Input Controls

  1. Distribution Type Selector

  2. Skewness Slider

  3. Kurtosis Slider

13. Box Plot Anatomy

Description

Interactively adjust dataset values to see quartiles, whiskers, and outliers update in real time.

Learning Goals

  • Interpret box plot components

  • Relate box plot features to dataset properties

Input Controls

  1. Drag Data Points

  2. Add Outlier (button)

  3. Reset Data (button)

14. Central Limit Theorem Animator

Description

Sample repeatedly from various population distributions and watch sampling distribution approach normality.

Learning Goals

  • Visualize the CLT in action

  • Understand why normality emerges

Input Controls

  1. Population Distribution Selector

  2. Sample Size Slider

  3. Number of Samples Slider

15. Sampling Bias Demonstrator

Description

Draw samples from skewed or representative datasets to see effect on mean/median estimates.

Learning Goals

  • Recognize sampling bias

  • Relate bias to poor generalization

Input Controls

  1. Sampling Method Selector (random, biased)

  2. Sample Size Slider

  3. Reset Data (button)

16. Hypothesis Testing Visualizer

Description

Adjust population mean and see how p-values change for given sample statistics.

Learning Goals

  • Understand null/alternative hypotheses

  • Interpret p-values visually

Input Controls

  1. Population Mean Slider

  2. Sample Mean Slider

  3. Sample Size Slider

17. Confidence Interval Explorer

Description

Show multiple sample means with confidence intervals and see coverage percentage.

Learning Goals

  • Understand confidence interval interpretation

  • See how sample size affects interval width

Input Controls

  1. Confidence Level Slider

  2. Sample Size Slider

  3. Number of Samples Slider

18. t-Test Simulator

Description

Compare means of two groups with adjustable overlap and see t-statistic and p-value.

Learning Goals

  • Perform and interpret t-tests

  • Relate group separation to statistical significance

Input Controls

  1. Mean Difference Slider

  2. Sample Size Slider

  3. Variance Slider

19. Correlation vs. Causation Scenario Builder

Description

Toggle between linked and independent variables with visual storytelling elements.

Learning Goals

  • Distinguish correlation from causation

  • Recognize spurious correlations

Input Controls

  1. Relationship Type Selector

  2. Add Confounder Variable (button)

  3. Noise Level Slider

20. Data Cleaning Sandbox

Description

Interactively identify and fix missing values, duplicates, and inconsistencies in a small dataset.

Learning Goals

  • Practice data cleaning operations

  • Recognize data quality issues

Input Controls

  1. Highlight Missing Values (checkbox)

  2. Fill Missing Values Method Selector

  3. Remove Duplicates (button)

21. Missing Data Imputation Lab

Description

Students choose different strategies to fill in missing values and compare how summaries change.

Learning Goals

  • Explore mean, median, mode, and model-based imputation

  • See effects of imputation on dataset statistics

Input Controls

  1. Imputation Method Selector

  2. Preview Changes (toggle)

  3. Apply Changes (button)

22. One-Hot Encoding Demonstrator

Description

Convert categorical variables into binary columns and see the dataset shape change.

Learning Goals

  • Understand one-hot encoding

  • Recognize dataset expansion with categorical variables

Input Controls

  1. Category Count Slider

  2. Toggle Encoding (checkbox)

  3. Reset Categories (button)

23. Feature Scaling Visualizer

Description

Scale features using min-max, standardization, or robust scaling, and compare results.

Learning Goals

  • Understand scaling methods

  • Recognize scaling's impact on model training

Input Controls

  1. Scaling Method Selector

  2. Dataset Selector

  3. Apply Scaling (button)

24. Scatter Plot Matrix Explorer

Description

Select variables to display in an interactive scatter plot matrix.

Learning Goals

  • Visualize pairwise relationships

  • Identify potential multicollinearity

Input Controls

  1. Variable Selector (multi-select)

  2. Highlight Correlated Pairs (toggle)

  3. Reset Matrix (button)

25. Multicollinearity Detector

Description

Add or remove features and see the correlation heatmap update in real time.

Learning Goals

  • Recognize multicollinearity

  • Learn its impact on regression models

Input Controls

  1. Add Feature (dropdown)

  2. Remove Feature (click)

  3. Threshold Slider for correlation warning

26. Gradient Descent Animation

Description

Visualize gradient descent steps on a 3D loss surface.

Learning Goals

  • Understand optimization paths

  • See effects of learning rate changes

Input Controls

  1. Learning Rate Slider

  2. Start Position Selector

  3. Play/Pause Steps (button)

27. Loss Function Comparator

Description

Compare MSE, MAE, and Huber loss on the same dataset.

Learning Goals

  • Understand different loss functions

  • Recognize how they respond to outliers

Input Controls

  1. Loss Function Selector

  2. Add Outlier (button)

  3. Reset Data (button)

28. Logistic Regression Probability Curve

Description

Adjust slope and intercept to see how the logistic curve shifts and steepens.

Learning Goals

  • Understand logistic regression shape

  • Relate slope to classification threshold sharpness

Input Controls

  1. Slope Slider

  2. Intercept Slider

  3. Show Decision Boundary (toggle)

29. Confusion Matrix Builder

Description

Manually adjust predictions to see confusion matrix cells update and metrics recalculate.

Learning Goals

  • Interpret precision, recall, F1-score

  • See trade-offs in prediction thresholds

Input Controls

  1. Threshold Slider

  2. Toggle Misclassification Highlight (checkbox)

  3. Reset Predictions (button)

30. ROC Curve Interactive Plotter

Description

Drag threshold point along the curve to see corresponding confusion matrix metrics.

Learning Goals

  • Understand ROC curves

  • Relate AUC to model performance

Input Controls

  1. Move Threshold Point (mouse drag)

  2. Toggle AUC Display (checkbox)

  3. Dataset Selector

31. Precision-Recall Trade-off Tool

Description

Visualize precision and recall lines as threshold changes, highlighting the F1-score peak.

Learning Goals

  • Recognize trade-offs between precision and recall

  • Identify optimal balance point

Input Controls

  1. Threshold Slider

  2. Show F1 Peak (toggle)

  3. Reset Chart (button)

32. Decision Tree Split Explorer

Description

Select split features and thresholds to see how data partitions change.

Learning Goals

  • Understand feature-based splitting

  • Recognize overfitting in deep trees

Input Controls

  1. Feature Selector

  2. Threshold Slider

  3. Add Split (button)

33. Random Forest Voting Visualizer

Description

Show predictions from individual trees and how majority vote determines the final prediction.

Learning Goals

  • Understand ensemble voting

  • See stability from multiple models

Input Controls

  1. Number of Trees Slider

  2. Tree Depth Slider

  3. Noise Level Slider

34. Bagging vs. Boosting Simulator

Description

Switch between bagging and boosting to compare error reduction over iterations.

Learning Goals

  • Contrast two ensemble methods

  • Understand impact on bias and variance

Input Controls

  1. Method Selector

  2. Number of Estimators Slider

  3. Learning Rate Slider (for boosting)

35. k-Means Clustering Playground

Description

Move cluster centers and see point assignments change instantly.

Learning Goals

  • Understand k-means mechanics

  • Recognize sensitivity to initialization

Input Controls

  1. Number of Clusters Slider

  2. Drag Cluster Centers

  3. Reset Clusters (button)

36. Elbow Method Visualizer

Description

Generate k-means cost curve to find optimal k.

Learning Goals

  • Apply elbow method for cluster selection

  • Interpret inertia curve

Input Controls

  1. Max k Slider

  2. Dataset Selector

  3. Recalculate Curve (button)

37. Hierarchical Clustering Dendrogram

Description

Cut dendrogram at different heights to form clusters.

Learning Goals

  • Interpret dendrograms

  • Relate cut height to cluster count

Input Controls

  1. Cut Height Slider

  2. Dataset Selector

  3. Toggle Leaf Labels (checkbox)

38. PCA Variance Explorer

Description

Adjust number of principal components and see variance explained update.

Learning Goals

  • Understand dimensionality reduction

  • Relate components to variance retention

Input Controls

  1. Number of Components Slider

  2. Dataset Selector

  3. Show Projection (toggle)

39. PCA Projection Visualizer

Description

Project high-dimensional data into 2D and explore structure.

Learning Goals

  • Visualize principal component projections

  • Detect patterns in reduced space

Input Controls

  1. Rotate Projection (mouse drag)

  2. Highlight Class Labels (toggle)

  3. Reset View (button)

40. Feature Importance Bar Chart

Description

Interactively remove features and see model accuracy update.

Learning Goals

  • Rank feature contributions

  • Recognize redundancy in predictors

Input Controls

  1. Remove Feature (click bar)

  2. Recalculate Accuracy (button)

  3. Reset Features (button)

41. Time Series Trend Explorer

Description

Students add or remove long-term upward or downward trends to see their effect on time series plots.

Learning Goals

  • Recognize trends in time series data

  • Separate trend from noise visually

Input Controls

  1. Trend Slope Slider

  2. Noise Level Slider

  3. Reset Series (button)

42. Seasonality Animator

Description

Add seasonal patterns to time series and adjust amplitude/frequency.

Learning Goals

  • Understand seasonality components

  • Differentiate seasonal effects from trends

Input Controls

  1. Amplitude Slider

  2. Frequency Slider

  3. Toggle Seasonal Component (checkbox)

43. Autocorrelation Plot Builder

Description

Interactively generate autocorrelation plots for different time series patterns.

Learning Goals

  • Recognize autocorrelation signatures

  • Link patterns to time lags

Input Controls

  1. Pattern Selector (trend, seasonal, white noise)

  2. Series Length Slider

  3. Recalculate Plot (button)

44. Moving Average Filter

Description

Smooth noisy time series using different window sizes.

Learning Goals

  • Apply moving average smoothing

  • Understand trade-off between smoothing and responsiveness

Input Controls

  1. Window Size Slider

  2. Toggle Original Series (checkbox)

  3. Reset Filter (button)

45. Exponential Smoothing Explorer

Description

Adjust smoothing factor to see effect on responsiveness to new data.

Learning Goals

  • Understand exponential smoothing

  • Compare to simple moving average

Input Controls

  1. Smoothing Factor Slider (0--1)

  2. Toggle Forecast Values (checkbox)

  3. Reset Data (button)

46. ARIMA Model Simulator

Description

Experiment with AR, I, and MA parameters to fit simple time series.

Learning Goals

  • Recognize ARIMA components

  • See parameter effects on forecast shape

Input Controls

  1. AR Order Slider

  2. I Order Slider

  3. MA Order Slider

47. Train-Test Split for Time Series

Description

Split data chronologically and compare forecasting performance.

Learning Goals

  • Understand why random splits don't work in time series

  • Practice chronological evaluation

Input Controls

  1. Split Point Slider

  2. Model Selector

  3. Show Forecast Horizon (toggle)

48. Outlier Impact on Time Series

Description

Insert outliers into time series and see how forecasts change.

Learning Goals

  • Recognize sensitivity to anomalies

  • Understand need for preprocessing

Input Controls

  1. Insert Outlier (click point)

  2. Outlier Magnitude Slider

  3. Remove Outliers (button)

49. TF-IDF Text Weighting Tool

Description

Type or paste text and see term frequencies and TF-IDF scores update live.

Learning Goals

  • Understand term frequency weighting

  • Recognize the role of inverse document frequency

Input Controls

  1. Text Input Box

  2. Toggle Stopword Removal (checkbox)

  3. Recalculate Scores (button)

50. Tokenization Visualizer

Description

See text split into tokens using different tokenization rules.

Learning Goals

  • Understand tokenization

  • Compare word vs. subword methods

Input Controls

  1. Text Input Box

  2. Tokenizer Type Selector

  3. Show Token IDs (toggle)

51. Word Embedding Explorer

Description

Plot word embeddings in 2D space and explore semantic similarity.

Learning Goals

  • Understand word vector representations

  • See clusters of related words

Input Controls

  1. Select Word to Highlight (dropdown)

  2. Show Similar Words (toggle)

  3. Reset Embeddings (button)

52. Sentiment Classification Threshold Tool

Description

Adjust sentiment score threshold to see how classifications change.

Learning Goals

  • Understand sentiment score distributions

  • See trade-offs in precision and recall for sentiment tasks

Input Controls

  1. Threshold Slider

  2. Show Confusion Matrix (checkbox)

  3. Dataset Selector

53. Bag-of-Words vs. Embeddings

Description

Switch between BoW and embedding-based representations to compare classification accuracy.

Learning Goals

  • Contrast sparse vs. dense text features

  • Recognize embedding advantages

Input Controls

  1. Representation Selector

  2. Dataset Selector

  3. Recalculate Accuracy (button)

54. Neural Network Layer Visualizer

Description

Show how inputs propagate through fully connected layers with activation functions.

Learning Goals

  • Understand forward propagation

  • Visualize activation transformations

Input Controls

  1. Number of Layers Slider

  2. Activation Function Selector

  3. Reset Network (button)

55. Activation Function Explorer

Description

Compare sigmoid, ReLU, and tanh shapes and outputs for input ranges.

Learning Goals

  • Recognize activation function behaviors

  • See saturation and dead neuron effects

Input Controls

  1. Function Selector

  2. Input Range Slider

  3. Toggle Derivative Curve (checkbox)

56. Weight Initialization Impact

Description

Initialize neural network weights differently and observe training convergence.

Learning Goals

  • Understand initialization strategies

  • See effect on loss curve and accuracy

Input Controls

  1. Initialization Method Selector

  2. Learning Rate Slider

  3. Reset Training (button)

57. Learning Rate Finder

Description

Gradually increase learning rate to see where loss diverges or minimizes fastest.

Learning Goals

  • Tune learning rate

  • Recognize underfitting and instability from wrong rates

Input Controls

  1. Start Rate Slider

  2. End Rate Slider

  3. Run LR Finder (button)

58. Convolution Filter Visualizer

Description

Apply filters to images and see resulting feature maps.

Learning Goals

  • Understand convolution in CNNs

  • Recognize edge and texture detection

Input Controls

  1. Filter Type Selector

  2. Kernel Size Slider

  3. Toggle Original Image (checkbox)

59. Pooling Layer Explorer

Description

Compare max pooling and average pooling effects on feature maps.

Learning Goals

  • Understand pooling

  • See dimensionality reduction effects

Input Controls

  1. Pooling Type Selector

  2. Pool Size Slider

  3. Toggle Feature Map Overlay (checkbox)

60. Overfitting in Deep Networks

Description

Increase network capacity and watch training vs. validation loss diverge.

Learning Goals

  • Recognize overfitting in neural nets

  • See regularization benefits

Input Controls

  1. Number of Neurons Slider

  2. Dropout Rate Slider

  3. Toggle Validation Curve (checkbox)

61. L1 vs. L2 Regularization Visualizer

Description

Toggle between L1 and L2 regularization and see coefficient shrinkage effects.

Learning Goals

  • Understand Lasso vs. Ridge regression

  • See how regularization affects weights

Input Controls

  1. Regularization Type Selector

  2. Penalty Strength Slider

  3. Reset Coefficients (button)

62. Dropout Effect Simulator

Description

Adjust dropout rates and watch neuron activations disappear during training.

Learning Goals

  • Understand dropout regularization

  • Recognize its role in preventing overfitting

Input Controls

  1. Dropout Rate Slider

  2. Toggle Training/Inference View (checkbox)

  3. Reset Network (button)

63. Early Stopping Demonstrator

Description

Visualize training and validation loss to determine optimal stop point.

Learning Goals

  • Understand early stopping criteria

  • Avoid overtraining a model

Input Controls

  1. Patience Slider

  2. Max Epochs Slider

  3. Toggle Loss Curves (checkbox)

64. SHAP Value Explorer

Description

Show feature contributions to individual predictions using SHAP values.

Learning Goals

  • Interpret model predictions

  • Recognize key contributing features

Input Controls

  1. Select Data Point (dropdown)

  2. Show Positive/Negative Contributions (toggle)

  3. Reset View (button)

65. Partial Dependence Plot Tool

Description

Adjust a single feature and see average prediction change while holding others constant.

Learning Goals

  • Interpret partial dependence

  • Detect feature impact trends

Input Controls

  1. Feature Selector

  2. Value Slider

  3. Toggle Confidence Interval (checkbox)

66. Counterfactual Example Generator

Description

Change features to flip a prediction outcome.

Learning Goals

  • Understand counterfactual reasoning

  • Identify decision boundaries

Input Controls

  1. Feature Sliders

  2. Toggle Prediction Probability (checkbox)

  3. Reset Features (button)

67. Bias Detection Dashboard

Description

Compare model accuracy across demographic subgroups.

Learning Goals

  • Detect model bias

  • Understand fairness metrics

Input Controls

  1. Group Selector

  2. Metric Selector

  3. Show Disparity Alert (checkbox)

68. Fairness Metric Comparator

Description

Compare demographic parity, equalized odds, and other fairness metrics.

Learning Goals

  • Interpret multiple fairness definitions

  • Recognize trade-offs between them

Input Controls

  1. Metric Selector

  2. Group Selector

  3. Highlight Best Metric (checkbox)

69. Adversarial Example Creator

Description

Add small perturbations to input data and see if predictions change.

Learning Goals

  • Understand adversarial vulnerability

  • Recognize security risks in ML

Input Controls

  1. Perturbation Magnitude Slider

  2. Noise Pattern Selector

  3. Reset Data (button)

70. Model Drift Monitor

Description

Compare live data predictions to historical model performance.

Learning Goals

  • Detect concept and data drift

  • Understand retraining triggers

Input Controls

  1. Time Window Selector

  2. Drift Metric Selector

  3. Refresh Data (button)

71. Hyperparameter Search Playground

Description

Run grid/random searches and compare performance heatmaps.

Learning Goals

  • Understand hyperparameter optimization

  • Interpret search results

Input Controls

  1. Search Type Selector

  2. Parameter Range Sliders

  3. Run Search (button)

72. Model Stacking Visualizer

Description

Show predictions from multiple base models and meta-learner output.

Learning Goals

  • Understand stacking ensembles

  • See diversity benefits

Input Controls

  1. Base Model Selector

  2. Meta-Learner Selector

  3. Toggle Base Predictions (checkbox)

73. Pipeline Builder

Description

Chain preprocessing and modeling steps interactively.

Learning Goals

  • Understand ML pipelines

  • Ensure reproducible workflows

Input Controls

  1. Add Step (dropdown)

  2. Remove Step (click)

  3. Run Pipeline (button)

74. Model Export and Import Simulator

Description

Save and reload trained models to demonstrate persistence.

Learning Goals

  • Understand model serialization

  • Practice deployment readiness

Input Controls

  1. Save Model (button)

  2. Load Model (button)

  3. Reset Session (button)

75. API Endpoint Tester

Description

Send requests to a mock ML API and view JSON responses.

Learning Goals

  • Understand model serving endpoints

  • Practice request formatting

Input Controls

  1. Input Data Field

  2. Send Request (button)

  3. View Raw Response (toggle)

76. Batch vs. Real-Time Prediction Tool

Description

Switch between batch file processing and live API predictions.

Learning Goals

  • Understand latency differences

  • Recognize trade-offs in deployment modes

Input Controls

  1. Mode Selector

  2. Upload Dataset (file input)

  3. Simulate Real-Time (button)

77. Model Version Comparator

Description

Load two model versions and compare accuracy and latency.

Learning Goals

  • Track performance over versions

  • Make informed upgrade decisions

Input Controls

  1. Version Selector A

  2. Version Selector B

  3. Compare Now (button)

78. A/B Testing Simulator

Description

Split traffic between two models and track conversions.

Learning Goals

  • Understand online experimentation

  • Interpret statistical significance

Input Controls

  1. Traffic Split Slider

  2. Run Experiment (button)

  3. View p-Value (toggle)

79. Cost of Prediction Calculator

Description

Estimate compute cost for different model sizes and usage levels.

Learning Goals

  • Relate model complexity to cost

  • Make cost-aware deployment decisions

Input Controls

  1. Model Size Slider

  2. Requests per Minute Slider

  3. Region Selector

80. Energy Efficiency Meter

Description

Track power consumption estimates during model inference.

Learning Goals

  • Recognize environmental impact of ML

  • Optimize for efficiency

Input Controls

  1. Model Type Selector

  2. Batch Size Slider

  3. Toggle Energy Display (checkbox)

81. Model Interpretability Dashboard

Description

Combine SHAP, partial dependence, and counterfactuals in one view for a selected prediction.

Learning Goals

  • Integrate multiple interpretability methods

  • Develop storytelling skills for predictions

Input Controls

  1. Data Point Selector

  2. Interpretation Method Toggle

  3. Export Dashboard (button)

82. LIME Explainer Tool

Description

Generate local linear approximations for individual predictions.

Learning Goals

  • Understand local interpretability

  • Compare to global feature importance

Input Controls

  1. Data Point Selector

  2. Number of Samples Slider

  3. Toggle Highlighted Features (checkbox)

83. What-If Analysis Playground

Description

Change feature values and watch prediction changes in real time.

Learning Goals

  • Explore "what-if" scenarios

  • Understand sensitivity of predictions

Input Controls

  1. Feature Sliders

  2. Reset to Original (button)

  3. Show Probability Curve (toggle)

84. Bias Mitigation Simulator

Description

Apply pre-processing or in-processing bias mitigation and measure impact.

Learning Goals

  • Evaluate fairness interventions

  • Compare accuracy before and after

Input Controls

  1. Mitigation Method Selector

  2. Target Group Selector

  3. Recalculate Metrics (button)

85. Model Robustness Tester

Description

Add noise, missing values, or feature shifts to test model stability.

Learning Goals

  • Assess robustness under real-world conditions

  • Identify fragile models

Input Controls

  1. Noise Level Slider

  2. Missing Value Percentage Slider

  3. Feature Shift Toggle

86. Ensemble Diversity Visualizer

Description

Plot decision boundaries of ensemble members to show diversity benefits.

Learning Goals

  • Understand why diversity improves ensembles

  • Detect overcorrelated base models

Input Controls

  1. Number of Models Slider

  2. Model Type Selector

  3. Toggle Overlay Boundaries (checkbox)

87. Transfer Learning Feature Explorer

Description

Load pre-trained model features and visualize them for a custom dataset.

Learning Goals

  • Understand feature reuse

  • See adaptation benefits

Input Controls

  1. Pre-Trained Model Selector

  2. Layer Output Selector

  3. Toggle Feature Map Display

88. Fine-Tuning Tracker

Description

Compare base and fine-tuned model accuracy/loss curves.

Learning Goals

  • Understand fine-tuning process

  • Evaluate improvements over baseline

Input Controls

  1. Learning Rate Slider

  2. Epoch Count Slider

  3. Toggle Base Model Curve (checkbox)

89. Multi-Task Learning Visualizer

Description

Train on two tasks simultaneously and track performance for each.

Learning Goals

  • Understand shared representations

  • Recognize trade-offs in multi-task setups

Input Controls

  1. Task Weight Sliders

  2. Epoch Count Slider

  3. Toggle Shared Layers (checkbox)

90. Attention Mechanism Explorer

Description

Visualize attention weights for sequence-to-sequence models.

Learning Goals

  • Understand how models focus on parts of input

  • Interpret attention heatmaps

Input Controls

  1. Input Sequence Field

  2. Highlight Attention Matrix (checkbox)

  3. Reset Example (button)

91. Transformer Architecture Flow

Description

Step through encoder and decoder layers with visual activations.

Learning Goals

  • See data flow in transformer models

  • Recognize role of each sublayer

Input Controls

  1. Layer Stepper (next/prev)

  2. Toggle Positional Encoding View

  3. Reset Sequence (button)

92. Hyperparameter Sensitivity Map

Description

Generate heatmaps showing accuracy changes across parameter ranges.

Learning Goals

  • Identify sensitive parameters

  • Focus tuning efforts effectively

Input Controls

  1. Parameter Range Sliders

  2. Run Grid Search (button)

  3. Toggle Best Point Marker (checkbox)

93. Model Compression Simulator

Description

Prune weights and quantize parameters, tracking accuracy drop.

Learning Goals

  • Understand trade-offs between size and performance

  • Recognize deployment benefits of smaller models

Input Controls

  1. Pruning Percentage Slider

  2. Quantization Level Selector

  3. Apply Compression (button)

94. Edge Deployment Emulator

Description

Simulate running a model on constrained hardware.

Learning Goals

  • Understand latency and memory constraints

  • Optimize for edge environments

Input Controls

  1. Hardware Profile Selector

  2. Batch Size Slider

  3. Toggle Latency Display (checkbox)

95. Streaming Data Dashboard

Description

Stream incoming data and update predictions in real time.

Learning Goals

  • Handle continuous inputs

  • Recognize challenges in online learning

Input Controls

  1. Stream Speed Slider

  2. Pause/Resume Stream (button)

  3. Reset Dashboard (button)

96. Online Learning Visualizer

Description

Update model incrementally with new data and track evolving accuracy.

Learning Goals

  • Understand incremental training

  • Monitor performance drift

Input Controls

  1. Learning Rate Slider

  2. Batch Size Slider

  3. Toggle History Chart (checkbox)

97. Capstone Project Data Selector

Description

Choose dataset for final project from curated sources and preview statistics.

Learning Goals

  • Practice dataset selection skills

  • Evaluate dataset suitability

Input Controls

  1. Dataset Selector

  2. Preview Stats (button)

  3. Download Data (button)

98. Model Comparison Dashboard

Description

Compare multiple models across accuracy, latency, and fairness metrics.

Learning Goals

  • Perform multi-metric evaluation

  • Select best model for deployment

Input Controls

  1. Model Selector (multi-select)

  2. Metric Selector

  3. Toggle Best Model Highlight

99. End-to-End Workflow Builder

Description

Drag-and-drop stages to build a full ML pipeline from data to deployment.

Learning Goals

  • Integrate all learned concepts

  • Visualize project workflow

Input Controls

  1. Stage Palette (drag items)

  2. Connect Stages (mouse drag)

  3. Run Workflow (button)

  4. Ethical Impact Assessment Tool

Description

Rate model across transparency, fairness, privacy, and societal impact dimensions.

Learning Goals

  • Incorporate ethics into data science projects

  • Balance technical and social factors

Input Controls

  1. Impact Category Sliders

  2. Generate Report (button)

  3. Reset Assessment (button)