feat(imputation): add physics-constrained MICE pipeline (Issue #86)

[Krishna200608]

Issue: #86

Implemented an advanced data imputation pipeline to handle missing values in the 30 approved planetary features. The goal was to produce a scientifically valid dataset by preserving the statistical and physical properties of exoplanets.

Logic & Methodology

• Multivariate Imputation (MICE): utilized IterativeImputer to estimate missing values based on correlations between features (e.g., Mass vs. Radius) rather than using simple averages, which can distort data structure.
• Physics-Informed Constraints: configured the regression model with min_value=0 and added a post-processing safety lock (clip(lower=0)). This ensures all imputed physical attributes (like Radius and Mass) remain non-negative, adhering to physical laws.
• Dynamic Pipeline Architecture: designed the preprocessor to automatically detect and exclude 100% empty columns at runtime, ensuring the pipeline remains robust regardless of input sparsity.
• Type-Safe Visualization: updated KDE plotting scripts to adhere to strict type-checking standards.

Validation

Generated Kernel Density Estimation (KDE) plots to verify that the imputed data follows the natural distribution of observed planets without introducing artificial artifacts or impossible values.

[OpenGitBot]

Hey @Krishna200608

PR Template is not followed. Kindly follow the template & edit it again for consideration as a valid pull request.

Thanks for contributing in OpenCode'25 ✨✨!

Correct Tempelate:
Issue : #<Issue_Number>

[OpenGitBot]

Hey @Krishna200608

Thanks for opening this PR 🚀. Mentor will review your pull request soon and till then, keep contributing and stay calm.

Thanks for contributing in OpenCode'25 ✨✨!

[OpenGitBot]

Hey @Krishna200608

Your PR has been merged 🥳🥳 and you have earned 30 points.

Thanks for contributing in OpenCode'25✨✨