ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
701 commits 391 branches 58 tags 304 MiB
Commit graph

40 commits

Author SHA1 Message Date
Roland Tannous
01082b84e5 Merge branch 'nightly' into feat/gguf-llama-cpp-inference 2026-02-25 16:06:03 +04:00
Roland Tannous
6f0b7bc38a fix: use raw github URL for vision.py patch + add VLM processor diagnostic logging 2026-02-25 10:29:05 +00:00
Roland Tannous
2149bc74ee Merge pull request #232 from unslothai/fix/disable-eval-by-default 
# fix/disable eval by default
 2026-02-24 13:35:11 +04:00
Roland Tannous
2be2933846 skip eval split and HF split detection when eval_steps is disabled 2026-02-24 09:26:54 +00:00
Leo Borcherding
cdeed53a97 fix: disable eval by default, set eval_steps to 0.0 
- Changed default eval_steps from 0.01 to 0.0 across backend and frontend
- Fixed UI to allow eval_steps=0 (removed min=0.001 constraint)
- Added conditional eval logic with helpful console messages
- Updated tooltip to explain how to disable evaluation
- Tested: confirmed eval disabled by default with eval_steps=0.0
 2026-02-23 13:07:47 -06:00
Roland Tannous
d74174f7f5 Cap dataset.map num_proc on multi-GPU machines to prevent fork deadlocks 2026-02-23 14:25:31 +00:00
Roland Tannous
3015916d26 fix: error on >30% sample drop after train_on_responses_only instead of silent DataLoader crash 2026-02-23 12:21:06 +00:00
Roland Tannous
dbbcdb4f09 feat: clear unsloth_compiled_cache on startup, shutdown, and between model loads 2026-02-23 07:26:22 +00:00
Roland Tannous
202b7cdfa7 fix: pass full Processor as processing_class for VLM SFTTrainer 2026-02-22 14:11:12 +00:00
Manan17
798bfb8f6f Setting it to total cpu_count // 4 2026-02-20 06:32:01 +00:00
Roland Tannous
adc0c78dbc reduce dataset_num_proc to 1/4 of cpu_count 2026-02-18 20:53:32 +00:00
Roland Tannous
e33920974b fix: unwrap ProcessorMixin to raw tokenizer for text-only SFTTrainer on VLM-architecture models 2026-02-18 19:13:20 +00:00
Roland Tannous
a6e2fa5b3a Merge remote-tracking branch 'origin/nightly' into fix/dataset-mapping-vlm-text-datasets 2026-02-18 18:11:45 +04:00
Manan17
76cd1dc24c fixing the hangup of training after multiple back to back training processes 2026-02-18 08:18:13 +00:00
Manan17
c37bf686a6 Dividing the total cpu_count // 3 2026-02-18 07:59:57 +00:00
Manan17
58116e7e7a fix the linear path on backend 2026-02-18 07:08:32 +00:00
Roland Tannous
d7853efd21 debug statements 2026-02-18 00:37:09 +00:00
Roland Tannous
0d8b67b706 fix: normalize target_modules [all-linear] list to string for Unsloth/PEFT compatibility 2026-02-18 00:32:21 +00:00
Roland Tannous
d2332622d1 fix: defensively rename VLM chat column to match model's forward() signature 2026-02-17 23:49:13 +00:00
Shine1i
dc0cec772d feat: enhance training stop and reset flow with detailed checks 2026-02-17 23:32:22 +01:00
Shine1i
0be3e6f525 feat: integrate gradient norm tracking in training runtime and metrics 
- Enhanced chart logic to filter and visualize finite gradient norm values.
 2026-02-17 18:26:59 +01:00
Roland Tannous
ff0aec180a Merge branch 'nightly' into feature/eval-split-auto-detection 2026-02-17 01:11:30 +04:00
Roland Tannous
fa0ca59215 feat: auto-detect model+dataset compatibility to select VLM vs LLM training path 2026-02-16 19:18:49 +00:00
Roland Tannous
5df3a0b250 feat: add eval_enabled flag and format-first-then-split for eval dataset 2026-02-16 14:13:55 +00:00
Roland Tannous
0aea3f149d feat: add eval split auto-detection, eval_steps hyperparam, and eval_loss chart integration 2026-02-16 13:51:10 +00:00
Roland Tannous
37452d56cf feat: add eval split auto-detection, eval_steps hyperparam, and eval_loss chart integration 2026-02-16 13:38:54 +00:00
Roland Tannous
6ecc03485d Merge pull request #97 from unslothai/fix/progress-metics 
Resolved the progress metrics
 2026-02-16 11:55:01 +04:00
sshah229
63b34660ed modified the num_tokens logic 2026-02-16 00:38:40 -07:00
Roland Tannous
d0964652af feat: thread dataset subset/split params from API routes through to load_dataset calls 2026-02-16 03:56:22 +00:00
sshah229
0b1c635b43 resolved the prgress metrics 2026-02-15 05:35:32 -07:00
Manan17
6e4cde3bf8 Adding save-steps to the SFTConfig 2026-02-15 09:37:54 +00:00
Manan17
6ccbc4edce Fixing stuck training processes 2026-02-15 05:38:06 +00:00
Manan17
97c6a09b84 feat: add cancel or save and stop training 2026-02-15 00:00:22 +00:00
Roland Tannous
4f0fad2156 fix: increase SSE progress timeout to 30min and allow step-0 updates 2026-02-14 05:47:22 +00:00
Roland Tannous
67edebfeb3 feat: wire custom_format_mapping through training pipeline to format_and_template_dataset 2026-02-13 21:07:36 +00:00
Roland Tannous
f52bddc23f refactor: remove gradio dependency from training backend 2026-02-13 09:25:49 +00:00
Roland Tannous
75f775d088 fix: change epoch type from int to float to match TrainerState 2026-02-13 06:51:55 +00:00
Roland Tannous
da1cde971c use get_device() for device selection and clear_gpu_cache() for GPU memory cleanup in inference, trainer, and export 2026-02-11 16:56:52 +00:00
Roland Tannous
62ddcfa019 Refactor [dataset_utils.py](cci:7://file:///home/support/new-ui-prototype/studio/backend/utils/datasets/dataset_utils.py:0:0-0:0) into focused modules 2026-02-03 14:38:02 +00:00
Roland Tannous
544d6944d1 root studio folder 2026-02-02 09:13:49 +00:00