huggingface peft issues

huggingface / peft

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

https://huggingface.co/docs/peft

Apache License 2.0

15.98k stars 1.56k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

PEFT Config checking update request

#2112 lemingshen opened 5 hours ago
0
could not finetune gemma 2 9b with lora and fsdp

#2111 imadoualid opened 18 hours ago
0
Update install.md

#2110 Salehbigdeli opened 1 day ago
0
add missed requirement

#2109 Salehbigdeli closed 1 day ago
0
Add missed requirement

#2108 Salehbigdeli closed 1 day ago
0
Optimize DoRA computation when there is no dropout

#2107 BenjaminBossan opened 2 days ago
0
FIX: Change check if past_key_values is empty

#2106 BenjaminBossan closed 2 days ago
3
merge_and_unload docs do not clarify behaviour for quantized base models

#2105 RonanKMcGovern opened 3 days ago
1
FIX: Transpose weight matrix based on fan_in_fan_out condition in PiSSA initialization (#2103)

#2104 suyang160 opened 3 days ago
0
Lora PISSA init: not support gpt2

#2103 suyang160 opened 3 days ago
2
FEAT: Adding exclude modules param(#2044)

#2102 JINO-ROHIT opened 3 days ago
2
adaption for moe models

#2101 dhrhank187 opened 3 days ago
5
Questions about original_module and modules_to_save.default

#2100 dengchengxifrank opened 4 days ago
1
Using module_to_save to save parameters inited by nn.parameters dose't work!

#2099 minmie opened 4 days ago
6
Add new features: Safe LoRA

#2098 chiayi-hsu opened 4 days ago
2
loftq_utils.py depdends on huggingface_hub.errors, which doesn't appear in some versions of huggingface_hub

#2097 mashoutsider opened 4 days ago
1
[WIP] Fix to prefix tuning to fit transformers

#2096 BenjaminBossan opened 4 days ago
1
Bump version to 0.13.1.dev0

#2094 BenjaminBossan closed 4 days ago
1
Release v0.13.0

#2093 BenjaminBossan closed 4 days ago
1
Why original layer weight is saved for LoRA adapter?

#2092 leosongwei closed 5 days ago
1
Abnormal performance of training LLaMA3.1-70 via LoRA

#2091 junzhang-zj opened 6 days ago
3
FIX Raise an error when performing mixed adapter inference and passing non-existing adapter names

#2090 BenjaminBossan opened 6 days ago
1
ENH: Better DoRA check in mixed adapter batch inference

#2089 BenjaminBossan closed 5 days ago
1
Fix func docstring

#2087 kwonmha closed 6 days ago
1
Update setup.py to update contact info

#2086 sayakpaul closed 6 days ago
1
Prompt-Tuning for text-to-image diffusion models

#2085 AHHHZ975 opened 1 week ago
9
Fix Inconsistent Missing Keys Warning for Adapter Weights in PEFT

#2084 yaswanth19 closed 4 days ago
11
FIX: Bug in find_minimal_target_modules

#2083 BenjaminBossan closed 6 days ago
1
Support Conv3d layer in LoRA and IA3

#2082 jsilter closed 4 days ago
6
Expose bias to to ModulesToSaveWrapper

#2081 dengdifan closed 1 week ago
1
make RMSNorm or other small parameters trainable with lora

#2080 IvanSedykh closed 1 week ago
2
Support Conv3d layer

#2079 jsilter opened 1 week ago
2
ENH: Add default target layers for gemma2 architecture

#2078 BenjaminBossan closed 6 days ago
1
ENH: PiSSA/OLoRA: Preserve original config on save

#2077 BenjaminBossan closed 1 week ago
2
FEAT: Support quantization for VeRA using bitsandbytes (#2070)

#2076 ZiadHelal opened 1 week ago
25
lora_r is double when converting olora to lora.

#2075 JaheimLee closed 1 week ago
4
[tests] skip some tests for XPU devices

#2074 faaany closed 1 week ago
4
Add scaling option to loftq

#2073 sparsh2 opened 1 week ago
1
ImportError: cannot import name 'VBLoRAConfig' from 'peft'

#2072 KQDtianxiaK closed 1 week ago
4
RuntimeError: Error(s) in loading state_dict for PeftModelForCausalLM: size mismatch when I load using adapter path but not checkpoint

#2071 manitadayon closed 1 week ago
9
[Feature] Add Quantization Support for VeRA Method

#2070 ZiadHelal opened 1 week ago
1
Unaligned blit request with RoBERTa

#2069 vrmer opened 2 weeks ago
1
FIX: Bug that prevents BOFT from loading multiple adapters

#2068 BenjaminBossan closed 1 week ago
3
Does peft supports the custom setting of trainable parameters(for example, some params in word_embeddings)

#2067 dongdongzhaoUP opened 2 weeks ago
2
About merge lora weight and lora dropout

#2066 hhnqqq closed 2 weeks ago
2
Merge LoRA into 405B

#2065 junzhang-zj opened 2 weeks ago
6
MAINT: Give stale bot permissions for PRs too

#2064 BenjaminBossan closed 2 weeks ago
1
question about training time

#2063 harborsarah opened 2 weeks ago
4
FEAT: Support torchao

#2062 BenjaminBossan opened 2 weeks ago
3
Update permissions for githubtoken stale.yml

#2061 glegendre01 closed 2 weeks ago
1