dmlc/gluon-cv

Add Tutorial: Finetuning an existing SSD model

Open

#216 opened on Jul 20, 2018

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (5,610 stars) (1,224 forks)batch import
enhancementgood first issue

Description

Hello! First of all, great work on this awesome lib!

I'm training my own SSDs for object detection, and I noticed there are no tutorials in the docs for fine-tuning a model in the zoo with a different dataset. I think it would be great for newcomers to have a starting point for quick training and results.

There is code for a ToySSD in the Object Detection sector of The Straight Dope, but you can't simply substitute the net in the Notebook for a net from the model_zoo.

Contributor guide

Tech stack
python
Domain
documentationmachine learning
Issue type
documentation
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.
half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
PythonObject detection basics
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
70
Research direction
Review existing object detection tutorials in gluon cv docs, especially the ToySSD example in The Straight Dope. Examine the model zoo APIs for loading pre trained SSD models. Check recent updates to the repository for any changes in fine tuning workflows. Consult issue comments for user experiences and maintainer guidance.
Add Tutorial: Finetuning an existing SSD model · dmlc/gluon-cv#216 | Good First Issue