huggingface / huggingface/diffusers

Training on unconditional image generation creates colorized images

Offen
#11,055 1 Kommentar 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen
bug stale
Vorherrschende Sprache
Python
Sterne
34.5k
Forks
7.3k
Ø Merge
3 T. 3 Std.
Gemergte PRs (30 T.)
91

Beschreibung

### Describe the bug

Hi, I'm trying to follow the tutorial from unconditional image generation on my own dataset, and I'm getting weirdly colored images. I originally thought it was due to RGB/BGR channel order, but I've switched it around and got the same result. Do you have any suggestions of how to fix it?

### Reproduction

NA

### Logs

```shell

```

### System Info

NA

### Who can help?

_No response_

Beitragsleitfaden

Beitragsleitfaden öffnen

Rechercherichtung

Start with the unconditional image generation tutorial and compare its dataset preprocessing with the reporter’s own dataset, including the stated RGB/BGR concern. First obtain a reproducible example with concrete dataset and system details, since the issue provides none; done means identifying the cause of the colorized output and verifying a fix.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Bewertung

Tech-Stack
python, pytorch
Bereich
computer-vision, machine-learning
Issue-Typ
Bug
Schwierigkeit
4/5
Geschätzter Aufwand
3-5 Tage
Aktivitätsstatus
Veraltet
Klarheit
Muss geklärt werden
Anfängerfreundlichkeit
20/100

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.