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
-
Add Point (click on canvas)
-
Remove Point (click existing point)
-
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
-
Bin Size Slider
-
Dataset Selector (dropdown: normal, uniform, skewed)
-
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
-
Drag Points (mouse)
-
Add Outlier (button)
-
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
-
Drag Cluster (mouse)
-
Add Noise (slider)
-
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
-
Slope Slider
-
Intercept Slider
-
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
-
Noise Level Slider
-
Number of Points Slider
-
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
-
Train/Test Ratio Slider
-
Resample Dataset (button)
-
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
-
Number of Folds Slider
-
Dataset Size Slider
-
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
-
Polynomial Degree Slider
-
Noise Level Slider
-
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)
-
Noise Level Slider
-
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
-
Model Complexity Slider
-
Toggle Residual Colors
-
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
-
Distribution Type Selector
-
Skewness Slider
-
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
-
Drag Data Points
-
Add Outlier (button)
-
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
-
Population Distribution Selector
-
Sample Size Slider
-
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
-
Sampling Method Selector (random, biased)
-
Sample Size Slider
-
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
-
Population Mean Slider
-
Sample Mean Slider
-
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
-
Confidence Level Slider
-
Sample Size Slider
-
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
-
Mean Difference Slider
-
Sample Size Slider
-
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
-
Relationship Type Selector
-
Add Confounder Variable (button)
-
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
-
Highlight Missing Values (checkbox)
-
Fill Missing Values Method Selector
-
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
-
Imputation Method Selector
-
Preview Changes (toggle)
-
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
-
Category Count Slider
-
Toggle Encoding (checkbox)
-
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
-
Scaling Method Selector
-
Dataset Selector
-
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
-
Variable Selector (multi-select)
-
Highlight Correlated Pairs (toggle)
-
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
-
Add Feature (dropdown)
-
Remove Feature (click)
-
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
-
Learning Rate Slider
-
Start Position Selector
-
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
-
Loss Function Selector
-
Add Outlier (button)
-
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
-
Slope Slider
-
Intercept Slider
-
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
-
Threshold Slider
-
Toggle Misclassification Highlight (checkbox)
-
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
-
Move Threshold Point (mouse drag)
-
Toggle AUC Display (checkbox)
-
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
-
Threshold Slider
-
Show F1 Peak (toggle)
-
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
-
Feature Selector
-
Threshold Slider
-
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
-
Number of Trees Slider
-
Tree Depth Slider
-
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
-
Method Selector
-
Number of Estimators Slider
-
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
-
Number of Clusters Slider
-
Drag Cluster Centers
-
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
-
Max k Slider
-
Dataset Selector
-
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
-
Cut Height Slider
-
Dataset Selector
-
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
-
Number of Components Slider
-
Dataset Selector
-
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
-
Rotate Projection (mouse drag)
-
Highlight Class Labels (toggle)
-
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
-
Remove Feature (click bar)
-
Recalculate Accuracy (button)
-
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
-
Trend Slope Slider
-
Noise Level Slider
-
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
-
Amplitude Slider
-
Frequency Slider
-
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
-
Pattern Selector (trend, seasonal, white noise)
-
Series Length Slider
-
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
-
Window Size Slider
-
Toggle Original Series (checkbox)
-
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
-
Smoothing Factor Slider (0--1)
-
Toggle Forecast Values (checkbox)
-
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
-
AR Order Slider
-
I Order Slider
-
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
-
Split Point Slider
-
Model Selector
-
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
-
Insert Outlier (click point)
-
Outlier Magnitude Slider
-
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
-
Text Input Box
-
Toggle Stopword Removal (checkbox)
-
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
-
Text Input Box
-
Tokenizer Type Selector
-
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
-
Select Word to Highlight (dropdown)
-
Show Similar Words (toggle)
-
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
-
Threshold Slider
-
Show Confusion Matrix (checkbox)
-
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
-
Representation Selector
-
Dataset Selector
-
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
-
Number of Layers Slider
-
Activation Function Selector
-
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
-
Function Selector
-
Input Range Slider
-
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
-
Initialization Method Selector
-
Learning Rate Slider
-
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
-
Start Rate Slider
-
End Rate Slider
-
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
-
Filter Type Selector
-
Kernel Size Slider
-
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
-
Pooling Type Selector
-
Pool Size Slider
-
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
-
Number of Neurons Slider
-
Dropout Rate Slider
-
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
-
Regularization Type Selector
-
Penalty Strength Slider
-
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
-
Dropout Rate Slider
-
Toggle Training/Inference View (checkbox)
-
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
-
Patience Slider
-
Max Epochs Slider
-
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
-
Select Data Point (dropdown)
-
Show Positive/Negative Contributions (toggle)
-
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
-
Feature Selector
-
Value Slider
-
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
-
Feature Sliders
-
Toggle Prediction Probability (checkbox)
-
Reset Features (button)
67. Bias Detection Dashboard
Description
Compare model accuracy across demographic subgroups.
Learning Goals
-
Detect model bias
-
Understand fairness metrics
Input Controls
-
Group Selector
-
Metric Selector
-
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
-
Metric Selector
-
Group Selector
-
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
-
Perturbation Magnitude Slider
-
Noise Pattern Selector
-
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
-
Time Window Selector
-
Drift Metric Selector
-
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
-
Search Type Selector
-
Parameter Range Sliders
-
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
-
Base Model Selector
-
Meta-Learner Selector
-
Toggle Base Predictions (checkbox)
73. Pipeline Builder
Description
Chain preprocessing and modeling steps interactively.
Learning Goals
-
Understand ML pipelines
-
Ensure reproducible workflows
Input Controls
-
Add Step (dropdown)
-
Remove Step (click)
-
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
-
Save Model (button)
-
Load Model (button)
-
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
-
Input Data Field
-
Send Request (button)
-
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
-
Mode Selector
-
Upload Dataset (file input)
-
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
-
Version Selector A
-
Version Selector B
-
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
-
Traffic Split Slider
-
Run Experiment (button)
-
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
-
Model Size Slider
-
Requests per Minute Slider
-
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
-
Model Type Selector
-
Batch Size Slider
-
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
-
Data Point Selector
-
Interpretation Method Toggle
-
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
-
Data Point Selector
-
Number of Samples Slider
-
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
-
Feature Sliders
-
Reset to Original (button)
-
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
-
Mitigation Method Selector
-
Target Group Selector
-
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
-
Noise Level Slider
-
Missing Value Percentage Slider
-
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
-
Number of Models Slider
-
Model Type Selector
-
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
-
Pre-Trained Model Selector
-
Layer Output Selector
-
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
-
Learning Rate Slider
-
Epoch Count Slider
-
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
-
Task Weight Sliders
-
Epoch Count Slider
-
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
-
Input Sequence Field
-
Highlight Attention Matrix (checkbox)
-
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
-
Layer Stepper (next/prev)
-
Toggle Positional Encoding View
-
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
-
Parameter Range Sliders
-
Run Grid Search (button)
-
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
-
Pruning Percentage Slider
-
Quantization Level Selector
-
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
-
Hardware Profile Selector
-
Batch Size Slider
-
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
-
Stream Speed Slider
-
Pause/Resume Stream (button)
-
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
-
Learning Rate Slider
-
Batch Size Slider
-
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
-
Dataset Selector
-
Preview Stats (button)
-
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
-
Model Selector (multi-select)
-
Metric Selector
-
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
-
Stage Palette (drag items)
-
Connect Stages (mouse drag)
-
Run Workflow (button)
-
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
-
Impact Category Sliders
-
Generate Report (button)
-
Reset Assessment (button)