optimized predict() loop using groupby

[ChaitanyaChute]

Fixes Issue #74

This PR optimizes the data aggregation step inside predict() by removing row-wise DataFrame iteration and repeated filtering operations. The refactor improves performance by processing grouped data in a single pass instead of repeatedly scanning the DataFrame.

Changes Made

Replaced

• df_data.iterrows() loop
• Per-row DataFrame filtering based on (True X, True Y)

Added

• Single-pass iteration using:

df_data.groupby("True XY")

• Direct extraction of grouped predictions:

group["Predicted X"].tolist()
group["Predicted Y"].tolist()

• Reuse of existing precomputed metrics:

  • precision_xy
  • accuracy_xy

Behavior Verification

• Output dictionary structure remains unchanged
• Precision and accuracy values remain unchanged
• Filtering of negative predictions is still respected