deeplearning4j/deeplearning4j-examples

how to compute priorBoxes in HouseNumberDetection example

Open

#828 opened on Mar 22, 2019

View on GitHub
 (0 comments) (2 reactions) (0 assignees)Java (2,395 stars) (1,841 forks)batch import
enhancementhelp wanted

Description

Issue Description

  • As I know, before training yolo2, we need provide priorBoxes for our model, but the HouseNumberDetection example did not explain how to compute it. I hope dl4j can provide some method that can compute priorBoxes by k-means algorithm base training data.

Version Information

  • 1.0.0-beta2

Contributor guide

Tech stack
java
Domain
machine learningdata
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
mostly clear
Prerequisites
YOLOk meansJavaDeeplearning4j
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
20
Research direction
Investigate the HouseNumberDetection example in the deeplearning4j examples repository to understand the current dataset and model structure. Research the YOLOv2 prior box computation using k means clustering on bounding box dimensions from the training data. Explore existing implementations in other frameworks (e.g., Darknet) and propose a Java method that can be added to the example or the core DL4J library. No existing PRs or comments provide further direction.
how to compute priorBoxes in HouseNumberDetection example · deeplearning4j/deeplearning4j-examples#828 | Good First Issue