pytorch/examples

Is total variation loss necessary in fast_neural_style?

Open

#278 opened on Jan 3, 2018

View on GitHub
 (0 comments) (1 reaction) (0 assignees)Python (21,634 stars) (9,429 forks)batch import
good first issuequestion

Description

I notice that there is no total variation loss regularization implemented in the example of fast_neural_style. But the paper declared it and their torch version use it. I'm wondering if total variation loss is necessary or not in style transfer.

Contributor guide

Tech stack
pythonpytorch
Domain
machine learningai
Issue type
research
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.
1-3 hours
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.
clear
Prerequisites
basic knowledge of neural style transferPyTorch basics
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
Investigate the fast neural style example in pytorch/examples. Compare the implementation with the original torch version referenced in the paper. Check if total variation loss is absent and whether it affects results. Answer the question by examining the code and possibly running experiments.