feature-dash | Predictive Feature Intelligence

Target: -

Feature Distributions
Target Predictive Power (IV / WoE)
Spearman Correlation Matrix
Redundant Pairs (|r| ≥ 0.75)
2D Quantile Interaction Grids

Select a Feature

Pure single-variable distribution statistics, percentiles, and histogram.

-
-
-
-
-
-
-
-

Feature Frequency Histogram (Non-Null Values)

Distribution of valid non-null observations

Distribution Summary & Quartiles

Metric Value Description

Select a Feature

Click any feature in the left sidebar to view quantile distributions, IV, and WoE stats.

-
-
-
-
-
-
-

Target Positive Rate (%) per Quantile Bin

Weight of Evidence (WoE) per Quantile Bin

Quantile Bin Distributions & WoE Breakdown

Bin Range [Min, Max] Mean Value Sample Count Positives (Target=1) Positive Rate (%) WoE IV Contribution
No feature selected

Spearman Rank Correlation Matrix Heatmap

Evaluates pairwise rank correlations across top predictive features (Hierarchically Clustered by Ward Distance). Values near +1.0 or -1.0 indicate strong monotonic relationships.

Redundant Feature Pairs (|r| ≥ 0.75) & Action Recommendations

High feature collinearity adds redundant variance to predictive models. For each pair with |r| ≥ 0.75, the feature with higher Information Value (IV) is recommended for retention.

Pair # Feature 1 (IV) Feature 2 (IV) Spearman Correlation (|r|) Action Recommendation
No redundant pairs detected

2D Feature Interaction Grids (Top Predictive Pairs)

Evaluates 5x5 quantile grid target positive rates (%) across top feature combinations to detect non-linear interaction synergy.