ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Studio: validate vision_image_size in YAML/model-default loader

Browse source 
mapBackendModelConfigToTrainingPatch now mirrors the backend validator
at studio/backend/models/training.py:169 by dropping any value that is
not an integer in [256, 2048]. Pre-fix, an imported YAML like
vision_image_size: 4096 or 640.5 would land in the store and the UI
would happily display it, only to fail when Start Training posted to
the backend. With this guard the store never holds a value the backend
would reject.
This commit is contained in:
Daniel Han 2026-05-24 14:06:26 +00:00
commit d0bfb11e4c
1 changed files with 12 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

16
studio/frontend/src/features/training/lib/model-defaults.ts
View file

@ -131,10 +131,18 @@ export function mapBackendModelConfigToTrainingPatch(
if (randomSeed !== undefined) patch.randomSeed = randomSeed;
if (Object.hasOwn(training ?? {}, "vision_image_size")) {
const visionImageSize = training?.vision_image_size == null
? null
: toNumber(training.vision_image_size);
if (visionImageSize !== undefined) patch.visionImageSize = visionImageSize;
const raw = training?.vision_image_size;
if (raw == null) {
patch.visionImageSize = null;
} else {
// Mirror the backend validator at studio/backend/models/training.py:169.
// Anything not an integer in [256, 2048] is dropped so the store and UI
// never show a value the backend would reject.
const n = toNumber(raw);
if (n !== undefined && Number.isInteger(n) && n >= 256 && n <= 2048) {
patch.visionImageSize = n;
}
}
}
const packing = toBoolean(training?.packing);