huggingface peft issues

huggingface / peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

https://huggingface.co/docs/peft

Apache License 2.0

16.53k stars 1.63k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Bump version of MacOS runners from 12 to 13

#2235 githubnemo opened 12 hours ago
1
TST: Skip test on multi-GPU as DataParallel fails

#2234 BenjaminBossan opened 16 hours ago
1
new version Bone

#2233 JL-er opened 21 hours ago
6
CI: Fix failing torchao test

#2232 BenjaminBossan closed 14 hours ago
1
Adding CorDA as an optional initialization method of LoRA

#2231 iboing opened 4 days ago
2
FIX: Prevent CUDA context initialization due to AWQ

#2230 BenjaminBossan opened 4 days ago
1
Update CPT documentation

#2229 tsachiblau opened 4 days ago
2
tp_layer.py lora_a,b init_method is different from the method in lora paper

#2228 aeeeeeep closed 4 days ago
2
FIX Correctly set device of input data in bnb test

#2227 BenjaminBossan closed 4 days ago
1
CI: Skip EETQ tests while broken

#2226 BenjaminBossan closed 4 days ago
1
[FEAT] EVA: ensure deterministic behavior of SVD on multi gpu setups

#2225 sirluk closed 4 days ago
2
TST: Eva: Speed up consistency tests

#2224 BenjaminBossan closed 3 days ago
9
TST: Move slow compile tests to nightly CI

#2223 BenjaminBossan closed 6 days ago
1
CI Update AutoAWQ version to fix CI

#2222 BenjaminBossan closed 6 days ago
3
Can not free GPU memory after Trainer.train() a Peft lora model

#2221 Deno-V closed 6 days ago
3
FIX: Avoid needless copy from modules_to_save

#2220 BenjaminBossan opened 1 week ago
1
Bug: BOFT forward/merging with CUDA

#2219 BenjaminBossan opened 1 week ago
6
[FIX] EVA `meta` device check bug + add multi-gpu functionality

#2218 sirluk closed 1 week ago
1
How to specify the coefficients of loading lora during inference?

#2216 laolongboy closed 1 week ago
1
[FIX] Invalid `None` check for `loftq_config` attribute in `LoraConfig`

#2215 sirluk closed 6 days ago
2
[FIX] Invalid none check for `loftq_config` attribute in `LoraConfig`

#2214 sirluk closed 1 week ago
0
训练时使用的QLoRA 4rank，进行cuda模型合并导出时出现，KeyError: 'base_model.model.model.model.layers.14.mlp.down_proj'

#2213 xiaoheiyue closed 1 day ago
19
Documentation for LoRAConfig.

#2212 brynhayder opened 1 week ago
1
the lack of adapter_model.bin and adapter_config.json after fine-tuning

#2211 TracyGuo2001 closed 1 week ago
14
Add Validation for Invalid `task_type` in PEFT Configurations

#2210 d-kleine closed 4 days ago
15
Unable to merge lora into base model properly?

#2209 hgftrdw45ud67is8o89 closed 2 weeks ago
2
TypeError: LoraConfig.__init__() got an unexpected keyword argument 'exclude_modules'

#2208 imrankh46 opened 2 weeks ago
25
update load_dataset for examples/feature_extraction

#2207 sinchir0 closed 2 weeks ago
2
modules_to_save Incorrect Overlap in Multiple LoRA Adapters

#2206 saeid93 opened 2 weeks ago
3
KeyError: Parameter containing

#2205 Amerehei opened 2 weeks ago
16
KeyError: 'messages'

#2204 rickeyhhh closed 1 week ago
9
Add Assertions for `task_type` in `LoraConfig`

#2203 d-kleine closed 4 days ago
1
FIX: LoRA & DoRA for depthwise-convolutional layers

#2202 gslama12 closed 2 weeks ago
2
Add new feature of SafeLoRA

#2201 chiayi-hsu opened 2 weeks ago
2
RuntimeError: element 0 of tensors.. OpenCLIP model

#2200 EngEmmanuel opened 2 weeks ago
4
CI: MacOS seems to be canceled, investigating

#2199 BenjaminBossan closed 2 weeks ago
1
WIP: Implement CorDA

#2198 5eqn closed 3 weeks ago
1
Dora_datacollector_updated

#2197 shirinyamani closed 3 weeks ago
3
Memory Inefficiency for LoRA & DoRA during fine-tuning.

#2196 gslama12 closed 3 weeks ago
3
[BUG] Issue with using `rank_pattern` and `alpha_pattern` together in `LoraConfig`

#2195 sirluk closed 3 weeks ago
1
[BUG] Issue with using `rank_pattern` and `alpha_pattern` together in `LoraConfig`

#2194 sirluk closed 6 days ago
2
Strange GPU MEM Occupation on GPU0 when using torchrun

#2193 ma787639046 closed 3 weeks ago
9
”peft_prefix_tuning_seq2seq.ipynb“RuntimeError Due to Tensor Dimension Mismatch

#2192 1hb6s7t closed 3 weeks ago
2
FIX: Check for prefix tuning + gradient checkpointing fails

#2191 BenjaminBossan closed 3 weeks ago
1
evaluation of peft model using lm-eval-harness toolkit

#2190 JINO-ROHIT closed 3 weeks ago
7
FIX: Prefix tuning with model on multiple devices

#2189 BenjaminBossan closed 3 weeks ago
2
How to change 'modules_to_save' setting when reloading a lora finetuned model

#2188 dengchengxifrank opened 3 weeks ago
1
TST: Skip AQLM test that is incompatible with torch 2.5

#2187 BenjaminBossan closed 3 weeks ago
1
ENH: Warn when loading PiSSA/OLoRA together with other adapters

#2186 BenjaminBossan closed 3 weeks ago
2
Xlora cannot reload model from last checkpoint by using trainer.train(resume_from_checkpoint="checkpp")

#2185 SongHanKen opened 4 weeks ago
0