JEE Rank Predictor

Statistical normalization · Neural prediction · Batch-trained

Loading model...

Test parameters

180 · 198 · 300
leave blank for auto

Prediction

—
Predicted percentile
Estimated rank
—
—
Est. topper score
—
stat. estimate
Difficulty
—
avg / max marks
Above avg by
—
raw marks

Normalization pipeline

difficulty = avg / max —
est. topper gap norm —
est. topper —
dynamic k —
sigma σ —
z-score —
x norm (vs topper) —
↓ RankNet
percentile —
⚠ Score is at or below class average. Model has limited data in this range — treat with caution.
⚠ Score exceeds estimated topper. Clamped to topper — you're likely in the top 1%.
⚠ Paper difficulty is outside training range. Prediction confidence is lower.
✓ Score is well above average. Prediction confidence is high.