AutoPyDep is a recommendation system for Python dependency management offering four key features: 

• Library Importance Analysis: Implements Eigenvector Centrality to identify critical libraries requiring timely updates, with color-coded urgency indicators (red for critical, green for non-critical) and one-click pip installation commands

• Community Impact Assessment: Visualizes library interdependencies using the Louvain community detection algorithm, enabling developers to understand the broader impact of specific library updates through an interactive dependency graph. The View Impact Graph feature provides an intuitive, dynamic visualization of direct connections between libraries, allowing developers to immediately identify dependency relationships 

• Release Category Prediction: Employs a Stacking Ensemble technique combining SVM, Random Forest, and Gradient Boosting classifiers to categorize releases (Bug Fix, New Feature, Performance Improvement, Security), validated through 10-fold cross-validation

• Release Date Forecasting: Uses a hybrid approach combining linear regression with ARIMA time-series modeling to predict release dates with a Mean Absolute Error of 1.8 months

Identified and implemented seven specific Metamorphic Relations (MRs) for mobile AR testing:

1. Raycast within Boundary - ensuring spawned objects remain within plane boundaries despite camera transformations, 

2. Overlapping with Game Object - preventing duplicate object spawning by using layer masking,

3. Visibility and Occlusion - maintaining proper occlusion adjustments as perspective changes

4. Object Scaling with Distance - correctly scaling objects based on camera distance

5. Orientation Consistency Relative to Gravity - preserving object orientation relative to gravity during device movement,

6.  Varying Rotation - maintaining rotation angles within specified thresholds, and

7. Check Correct Object Instantiation - verifying touch-instantiated objects match expectations.