issues
search
finegrain-ai
/
refiners
A microframework on top of PyTorch with first-class citizen APIs for foundation model adaptation
https://refine.rs
MIT License
584
stars
43
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add ControlNet Tile e2e test
#381
limiteinductive
closed
2 months ago
0
update dependencies and fix issues in tests
#380
catwell
closed
2 months ago
0
Add scale_decay parameter to ControlNet
#379
limiteinductive
closed
2 months ago
0
fix typing problems
#378
catwell
closed
2 months ago
0
Injecting/ejecting multiple controlnet models results in an error
#377
holwech
closed
3 months ago
2
generate doc for LatentDiffusionModel
#376
catwell
closed
3 months ago
0
add a docstring for set_inference_steps
#375
catwell
closed
3 months ago
0
training_utils: export neptune config and mixin
#373
deltheil
closed
3 months ago
0
Neptune callback pydantic fix
#372
Laurent2916
closed
3 months ago
0
add NeptuneCallback
#371
Laurent2916
closed
3 months ago
0
Fix clock log order
#369
limiteinductive
closed
3 months ago
0
Refactor Trainer with `data_iterable`
#368
limiteinductive
closed
4 months ago
0
Fix `StableDiffusion_1_Inpainting` init
#367
Laurent2916
closed
4 months ago
0
Add event configs to Callback
#366
limiteinductive
closed
4 months ago
0
Refactor TimeValue
#365
limiteinductive
closed
4 months ago
0
Add `sample_noise` and modify `add_noise`
#364
Laurent2916
closed
4 months ago
0
Fix training_utils Requirements Check
#363
limiteinductive
closed
4 months ago
0
SAM: image_to_scaled_tensor gray images
#362
piercus
closed
5 months ago
0
Add init file to SAM tests
#361
limiteinductive
closed
5 months ago
0
Add new ForceCommit callback in training utils
#360
Laurent2916
closed
5 months ago
0
Add `DataloaderConfig` to `Trainer`
#359
limiteinductive
closed
5 months ago
0
Turn `scoped_seed` into a Context Manager
#358
limiteinductive
closed
5 months ago
0
TrainerClock : assert dataset_length >= batch_size
#357
piercus
closed
5 months ago
0
SAM: expose sizing helpers
#356
piercus
closed
5 months ago
0
Support DINOv2 Giant flavor
#355
Laurent2916
closed
5 months ago
0
Add Static Typing to __call__ method for LatentDiffusion models
#354
limiteinductive
closed
5 months ago
0
hq sam: add constructor args to docstring
#353
deltheil
closed
5 months ago
0
SAM/HQSAMAdapter: docstring examples
#352
piercus
closed
5 months ago
0
use factories in context example
#351
catwell
closed
5 months ago
0
Fix clock step inconsistencies on batch end
#350
hugojarkoff
closed
5 months ago
0
add DINOv2-FD metric
#349
catwell
closed
5 months ago
0
get rid of invisible-watermark test dependency
#348
deltheil
closed
5 months ago
0
Fix DINOv2 interpolation batching
#347
Laurent2916
closed
5 months ago
0
Remove typos "arange" ignore
#346
Laurent2916
closed
5 months ago
0
update pyright and fix Pillow 10.3 typing issues
#345
catwell
closed
5 months ago
0
Fix typos ignores
#344
Laurent2916
closed
5 months ago
0
(DINOv2) Implement positional encoding interpolation
#343
Laurent2916
closed
5 months ago
0
HQSAM : weights/load_weights
#342
piercus
closed
5 months ago
0
add missing word in documentation
#341
catwell
closed
5 months ago
0
use uv for Rye
#340
catwell
closed
5 months ago
0
test weights: fix control-lora expected hashes
#339
deltheil
closed
5 months ago
0
tweak auto_attach_loras so debugging is easier when it fails
#338
catwell
closed
5 months ago
0
Fix IPAdapter injection
#337
Laurent2916
closed
5 months ago
0
do not call `__getattr__` with keyword argument
#336
deltheil
closed
5 months ago
0
SAM init with mask_decoder argument
#335
piercus
closed
5 months ago
0
HQ-SAM logit torch.equal test
#334
piercus
closed
5 months ago
0
Mistral 7b 0.1 Implementation
#333
AyoubMDL
closed
2 months ago
9
support no CFG in compute_clip_text_embedding
#332
catwell
closed
5 months ago
0
SAM e2e test tolerance explained
#331
piercus
closed
5 months ago
0
Update README.md with HQ-SAM news
#330
hugojarkoff
closed
5 months ago
0
Previous
Next