Praxis Model Arena

Compare detection outputs against ground truth.

Load one labeled reference and two to four result files. Arena scores each run, identifies exact error frames, and exports the comparison without uploading the data.

Local evaluator readyJSON and CSV inputs
Local browser runDetection files stay on this device.

Arena scores result files. It does not execute model weights.

Inputs

Load detections

01 Ground truthOne JSON or CSV file
02 Model results0 of 4 files
Accepted detection format

Use integer frame values and matching coordinate units in every file. JSON accepts a detection array, COCO annotations, a detections array, or frames with nested detections. CSV accepts x1, y1, x2, y2 or x, y, width, height. Confidence defaults to 1. Labels default to object.

Method

Set the match rule

A prediction matches one ground-truth box when overlap meets this threshold.

Predictions below this score are excluded before matching.

Scoring rule

Predictions are ordered by confidence, then matched once within the same frame and class to the available ground-truth box with the highest IoU. Scores aggregate every loaded frame.

Choose one ground-truth file and two to four model result files.