url stringlengths 56 57 | repository_url stringclasses 1
value | labels_url stringlengths 70 71 | comments_url stringlengths 65 66 | events_url stringlengths 63 64 | html_url stringlengths 46 47 | id int64 1.88B 2.91B | node_id stringlengths 18 18 | number int64 906 2.42k | title stringlengths 3 380 | user dict | labels listlengths 0 3 | state stringclasses 2
values | locked bool 1
class | assignee dict | assignees listlengths 0 1 | milestone null | comments int64 0 42 | created_at timestamp[s] | updated_at timestamp[s] | closed_at timestamp[s] | author_association stringclasses 4
values | sub_issues_summary dict | active_lock_reason null | draft bool 0
classes | pull_request dict | body stringlengths 4 45.9k ⌀ | closed_by dict | reactions dict | timeline_url stringlengths 65 66 | performed_via_github_app null | state_reason stringclasses 3
values | is_pull_request bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://api.github.com/repos/huggingface/peft/issues/1298 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1298/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1298/comments | https://api.github.com/repos/huggingface/peft/issues/1298/events | https://github.com/huggingface/peft/issues/1298 | 2,056,130,139 | I_kwDOIf9iDM56jg5b | 1,298 | [Question] What is the main difference between "modules_to_save" and "target_modules"? | {
"login": "SatireY",
"id": 42096167,
"node_id": "MDQ6VXNlcjQyMDk2MTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/42096167?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SatireY",
"html_url": "https://github.com/SatireY",
"followers_url": "https://api.github.com/users/Satire... | [] | closed | false | null | [] | null | 5 | 2023-12-26T07:37:05 | 2024-02-03T15:03:27 | 2024-02-03T15:03:27 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, in my work I need to add some special token to LLAMA, so I need to train the parameter of ["embed_tokens", "lm_head"] for both layers, what confuses me is that should I add this parameter to LoraConfig's "modules_to_save " or "target_modules"? Looking forward to your reply! | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1298/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1298/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1296 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1296/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1296/comments | https://api.github.com/repos/huggingface/peft/issues/1296/events | https://github.com/huggingface/peft/issues/1296 | 2,055,422,901 | I_kwDOIf9iDM56g0O1 | 1,296 | using LORA with multi-gpu | {
"login": "wxyzjp123",
"id": 59395733,
"node_id": "MDQ6VXNlcjU5Mzk1NzMz",
"avatar_url": "https://avatars.githubusercontent.com/u/59395733?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wxyzjp123",
"html_url": "https://github.com/wxyzjp123",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 4 | 2023-12-25T06:23:06 | 2023-12-26T12:10:20 | 2023-12-26T12:10:20 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Pytorch DP
### Who can help?
@pacman100 @younesbelkada @sayakpaul
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
### Reproduction
model = Aut... | {
"login": "wxyzjp123",
"id": 59395733,
"node_id": "MDQ6VXNlcjU5Mzk1NzMz",
"avatar_url": "https://avatars.githubusercontent.com/u/59395733?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wxyzjp123",
"html_url": "https://github.com/wxyzjp123",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1296/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1296/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1293 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1293/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1293/comments | https://api.github.com/repos/huggingface/peft/issues/1293/events | https://github.com/huggingface/peft/issues/1293 | 2,054,606,813 | I_kwDOIf9iDM56ds_d | 1,293 | support 2bit quip# method | {
"login": "Minami-su",
"id": 76865636,
"node_id": "MDQ6VXNlcjc2ODY1NjM2",
"avatar_url": "https://avatars.githubusercontent.com/u/76865636?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Minami-su",
"html_url": "https://github.com/Minami-su",
"followers_url": "https://api.github.com/users/... | [
{
"id": 5192583244,
"node_id": "LA_kwDOIf9iDM8AAAABNYCITA",
"url": "https://api.github.com/repos/huggingface/peft/labels/PRs%20welcome%20to%20address%20this",
"name": "PRs welcome to address this",
"color": "0052cc",
"default": false,
"description": "contributions are welcome from commun... | closed | false | null | [] | null | 25 | 2023-12-23T01:10:56 | 2024-07-11T16:47:44 | 2024-06-04T15:03:50 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | https://github.com/Cornell-RelaxML/quip-sharp
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1293/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1293/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1292 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1292/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1292/comments | https://api.github.com/repos/huggingface/peft/issues/1292/events | https://github.com/huggingface/peft/issues/1292 | 2,054,354,695 | I_kwDOIf9iDM56cvcH | 1,292 | Selecting LoRA layers from PeftModel | {
"login": "PCanavelli",
"id": 43319759,
"node_id": "MDQ6VXNlcjQzMzE5NzU5",
"avatar_url": "https://avatars.githubusercontent.com/u/43319759?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PCanavelli",
"html_url": "https://github.com/PCanavelli",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 2 | 2023-12-22T18:23:48 | 2024-01-31T15:03:38 | 2024-01-31T15:03:38 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi everyone,
I've been strunggling with what may just be an API usage issue as of late.
The use case is a follows: I am training a diffusion model with LoRA adapters in both the text encoder and UNet. I would like to selectively change the dtype and activate/deactivate the gradients of the LoRA layers, without im... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1292/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1292/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1289 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1289/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1289/comments | https://api.github.com/repos/huggingface/peft/issues/1289/events | https://github.com/huggingface/peft/issues/1289 | 2,051,624,183 | I_kwDOIf9iDM56SUz3 | 1,289 | What's the default setting for "target_modules" for TaskType.CAUSAL_LM (LoraConfig) | {
"login": "suchunxie",
"id": 85145489,
"node_id": "MDQ6VXNlcjg1MTQ1NDg5",
"avatar_url": "https://avatars.githubusercontent.com/u/85145489?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/suchunxie",
"html_url": "https://github.com/suchunxie",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 4 | 2023-12-21T03:23:43 | 2023-12-22T04:37:29 | 2023-12-21T13:21:05 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I'm fine-tuning Llama with Lora, and today I found my code didn't specify the `target_modules` parameter when training, which makes me a bit worried. I would appreciate your help.
My code is as follows:
```
model = create_hf_model(LlamaForCausalLM, args.model_name_or_path, tokenizer)
peft_config = LoraConf... | {
"login": "suchunxie",
"id": 85145489,
"node_id": "MDQ6VXNlcjg1MTQ1NDg5",
"avatar_url": "https://avatars.githubusercontent.com/u/85145489?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/suchunxie",
"html_url": "https://github.com/suchunxie",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1289/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1289/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1284 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1284/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1284/comments | https://api.github.com/repos/huggingface/peft/issues/1284/events | https://github.com/huggingface/peft/issues/1284 | 2,049,371,796 | I_kwDOIf9iDM56Ju6U | 1,284 | Prompt Tuning documentation trains all parameters | {
"login": "Ben-Epstein",
"id": 22605641,
"node_id": "MDQ6VXNlcjIyNjA1NjQx",
"avatar_url": "https://avatars.githubusercontent.com/u/22605641?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Ben-Epstein",
"html_url": "https://github.com/Ben-Epstein",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 7 | 2023-12-19T20:15:44 | 2024-01-28T15:03:35 | 2024-01-28T15:03:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Google colab
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
It seems lik... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1284/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1284/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1283 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1283/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1283/comments | https://api.github.com/repos/huggingface/peft/issues/1283/events | https://github.com/huggingface/peft/issues/1283 | 2,048,642,574 | I_kwDOIf9iDM56G84O | 1,283 | Can not import in python file. | {
"login": "aaghiijnnuz",
"id": 30295884,
"node_id": "MDQ6VXNlcjMwMjk1ODg0",
"avatar_url": "https://avatars.githubusercontent.com/u/30295884?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aaghiijnnuz",
"html_url": "https://github.com/aaghiijnnuz",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 1 | 2023-12-19T13:17:35 | 2023-12-19T14:01:03 | 2023-12-19T14:01:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
python 3.10.12
ubuntu 22.04
peft 0.7.1
cuda version:12.3
nvidia driver version:545.23.08
### Who can help?
@pacman100 @younesbelkada
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` f... | {
"login": "aaghiijnnuz",
"id": 30295884,
"node_id": "MDQ6VXNlcjMwMjk1ODg0",
"avatar_url": "https://avatars.githubusercontent.com/u/30295884?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aaghiijnnuz",
"html_url": "https://github.com/aaghiijnnuz",
"followers_url": "https://api.github.com/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1283/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1283/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1280 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1280/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1280/comments | https://api.github.com/repos/huggingface/peft/issues/1280/events | https://github.com/huggingface/peft/issues/1280 | 2,048,221,691 | I_kwDOIf9iDM56FWH7 | 1,280 | How can we enable continuous learning with the LLM model ? | {
"login": "TapendraBaduwal",
"id": 87867535,
"node_id": "MDQ6VXNlcjg3ODY3NTM1",
"avatar_url": "https://avatars.githubusercontent.com/u/87867535?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/TapendraBaduwal",
"html_url": "https://github.com/TapendraBaduwal",
"followers_url": "https://api... | [] | closed | false | null | [] | null | 8 | 2023-12-19T09:04:18 | 2024-01-29T15:03:38 | 2024-01-29T15:03:38 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
If we merge adapters then this work for continuous learning ?
I expect the new train model to provide output for previously trained data as well.
```python
model_path = "model/path"
# Reload model in FP16 and merge it with LoRA weights
base_model = AutoModelForCausalLM.from_pretrained(
... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1280/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1280/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1278 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1278/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1278/comments | https://api.github.com/repos/huggingface/peft/issues/1278/events | https://github.com/huggingface/peft/issues/1278 | 2,045,062,207 | I_kwDOIf9iDM555Sw_ | 1,278 | How to add trainable parameters? (bugs in 'modules_to_save') | {
"login": "shawnricecake",
"id": 39080134,
"node_id": "MDQ6VXNlcjM5MDgwMTM0",
"avatar_url": "https://avatars.githubusercontent.com/u/39080134?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shawnricecake",
"html_url": "https://github.com/shawnricecake",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 5 | 2023-12-17T05:34:09 | 2024-01-29T15:03:39 | 2024-01-29T15:03:39 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Hi,
How can I train other weights in the model rather than fix them during lora training?
### Who can help?
@BenjaminBossan Hi, I find you are active recently so I @ you here..
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially su... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1278/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1278/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1272 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1272/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1272/comments | https://api.github.com/repos/huggingface/peft/issues/1272/events | https://github.com/huggingface/peft/issues/1272 | 2,043,392,123 | I_kwDOIf9iDM55y7B7 | 1,272 | Adding LoRA like additional trainable parameters for nn.Parameter() or nn.Embedding() | {
"login": "lawrence-cj",
"id": 55283454,
"node_id": "MDQ6VXNlcjU1MjgzNDU0",
"avatar_url": "https://avatars.githubusercontent.com/u/55283454?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lawrence-cj",
"html_url": "https://github.com/lawrence-cj",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 8 | 2023-12-15T10:17:03 | 2023-12-17T16:03:35 | 2023-12-17T16:03:19 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
When training LoRA with PEFT, if there is a learnable embedding or parameter variables in the model, PEFT won't add additional trainable parameters for them. Maybe trainable parameter for nn.embedding() and nn.Parameter() data should be added for model fine-tuning.
### Motivation
When I use LoraC... | {
"login": "lawrence-cj",
"id": 55283454,
"node_id": "MDQ6VXNlcjU1MjgzNDU0",
"avatar_url": "https://avatars.githubusercontent.com/u/55283454?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lawrence-cj",
"html_url": "https://github.com/lawrence-cj",
"followers_url": "https://api.github.com/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1272/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1272/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1270 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1270/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1270/comments | https://api.github.com/repos/huggingface/peft/issues/1270/events | https://github.com/huggingface/peft/issues/1270 | 2,042,840,597 | I_kwDOIf9iDM55w0YV | 1,270 | Can peft support singular value decomposition? | {
"login": "hobbitlzy",
"id": 35392624,
"node_id": "MDQ6VXNlcjM1MzkyNjI0",
"avatar_url": "https://avatars.githubusercontent.com/u/35392624?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hobbitlzy",
"html_url": "https://github.com/hobbitlzy",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 16 | 2023-12-15T03:45:03 | 2024-01-03T11:15:06 | 2023-12-25T03:32:43 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I am new to `peft`. I am wondering whether the `peft` supports use singular value decomposition to replace the orignal linear module. I find this is similar to the funciton of lora, but different at the `loar_A` and `lora_B` are decomposition of the original weights.
If this does not support, is there any advic... | {
"login": "hobbitlzy",
"id": 35392624,
"node_id": "MDQ6VXNlcjM1MzkyNjI0",
"avatar_url": "https://avatars.githubusercontent.com/u/35392624?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hobbitlzy",
"html_url": "https://github.com/hobbitlzy",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1270/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1270/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1277 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1277/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1277/comments | https://api.github.com/repos/huggingface/peft/issues/1277/events | https://github.com/huggingface/peft/issues/1277 | 2,044,847,804 | I_kwDOIf9iDM554ea8 | 1,277 | Mistral: Expected all tensors to be on the same device, but found at least two devices, cuda:6 and cuda:7! (when checking argument for argument mat2 in method wrapper_CUDA_mm) | {
"login": "amritgupta98",
"id": 23277271,
"node_id": "MDQ6VXNlcjIzMjc3Mjcx",
"avatar_url": "https://avatars.githubusercontent.com/u/23277271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/amritgupta98",
"html_url": "https://github.com/amritgupta98",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | null | 11 | 2023-12-14T23:03:53 | 2024-04-25T17:15:50 | 2024-01-22T15:03:44 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | # System Info
- ```transformers``` version: 4.36.1
- Platform: Amazon Sagemaker Studio
- Python version: 3.10
- PyTorch version (GPU?): 2.0.1 (True)
- Tensorflow version (GPU?): not installed (NA)
- Flax version (CPU?/GPU?/TPU?): not installed (NA)
- Jax version: not installed
- JaxLib version: not installed
... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1277/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1277/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1265 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1265/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1265/comments | https://api.github.com/repos/huggingface/peft/issues/1265/events | https://github.com/huggingface/peft/issues/1265 | 2,041,159,478 | I_kwDOIf9iDM55qZ82 | 1,265 | When generate outputs, how to get the probility of the outputs? Is there any param to let the model output probility ? | {
"login": "ShawnALiu",
"id": 45930473,
"node_id": "MDQ6VXNlcjQ1OTMwNDcz",
"avatar_url": "https://avatars.githubusercontent.com/u/45930473?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ShawnALiu",
"html_url": "https://github.com/ShawnALiu",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 1 | 2023-12-14T08:05:34 | 2023-12-14T10:37:19 | 2023-12-14T10:37:19 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
xx
### Motivation
xx
### Your contribution
xx | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1265/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1265/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1264 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1264/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1264/comments | https://api.github.com/repos/huggingface/peft/issues/1264/events | https://github.com/huggingface/peft/issues/1264 | 2,040,857,525 | I_kwDOIf9iDM55pQO1 | 1,264 | done | {
"login": "aohan237",
"id": 3992281,
"node_id": "MDQ6VXNlcjM5OTIyODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/3992281?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aohan237",
"html_url": "https://github.com/aohan237",
"followers_url": "https://api.github.com/users/aohan... | [] | closed | false | null | [] | null | 0 | 2023-12-14T03:56:29 | 2023-12-14T08:58:13 | 2023-12-14T08:58:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | null | {
"login": "aohan237",
"id": 3992281,
"node_id": "MDQ6VXNlcjM5OTIyODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/3992281?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aohan237",
"html_url": "https://github.com/aohan237",
"followers_url": "https://api.github.com/users/aohan... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1264/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1264/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1260 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1260/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1260/comments | https://api.github.com/repos/huggingface/peft/issues/1260/events | https://github.com/huggingface/peft/issues/1260 | 2,039,136,261 | I_kwDOIf9iDM55isAF | 1,260 | This regex is confusing | {
"login": "s-k-yx",
"id": 39121401,
"node_id": "MDQ6VXNlcjM5MTIxNDAx",
"avatar_url": "https://avatars.githubusercontent.com/u/39121401?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/s-k-yx",
"html_url": "https://github.com/s-k-yx",
"followers_url": "https://api.github.com/users/s-k-yx/fo... | [] | closed | false | null | [] | null | 3 | 2023-12-13T07:41:33 | 2024-01-10T14:24:07 | 2024-01-10T14:24:06 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | https://github.com/huggingface/peft/blob/c9df262d69b41d36c7e199219605420be03dfbca/src/peft/tuners/tuners_utils.py#L502
It works now, but seems has a wrong format.
`r".*\.{pattern}\.(\d+)\..*".format(pattern=pattern)` this is a better format, if only layer_index is needed `r".*\.{pattern}\.(\d+)".format(pattern=patter... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1260/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1260/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1259 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1259/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1259/comments | https://api.github.com/repos/huggingface/peft/issues/1259/events | https://github.com/huggingface/peft/issues/1259 | 2,038,851,491 | I_kwDOIf9iDM55hmej | 1,259 | PEFT Model returns same results like original model after fine tuning using LORA. | {
"login": "SachinLimbachiya",
"id": 112546107,
"node_id": "U_kgDOBrVROw",
"avatar_url": "https://avatars.githubusercontent.com/u/112546107?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SachinLimbachiya",
"html_url": "https://github.com/SachinLimbachiya",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | null | 3 | 2023-12-13T02:42:10 | 2024-12-30T19:54:22 | 2024-01-20T15:03:32 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I have fine tune the FlanT5 with target module Q and V base model using my data but during inference, it returns exact same results as the original model. Upon checking the weight in these target modules before and after, they are same which means during fine tuning weight is not updated. Does anyone ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1259/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1259/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1253 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1253/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1253/comments | https://api.github.com/repos/huggingface/peft/issues/1253/events | https://github.com/huggingface/peft/issues/1253 | 2,036,410,566 | I_kwDOIf9iDM55YSjG | 1,253 | Retraining Fine tune model | {
"login": "AliHaider20",
"id": 42317018,
"node_id": "MDQ6VXNlcjQyMzE3MDE4",
"avatar_url": "https://avatars.githubusercontent.com/u/42317018?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AliHaider20",
"html_url": "https://github.com/AliHaider20",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 12 | 2023-12-11T19:47:05 | 2024-01-27T03:26:23 | 2024-01-27T03:26:23 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Model saving:
```
base_model_reload = AutoModelForCausalLM.from_pretrained(
base_model, low_cpu_mem_usage=True, trust_remote_code=True,
return_dict=True,torch_dtype=torch.bfloat16,
device_map= {"": 0})
model = PeftModel.from_pretrained(base_model_reload, new_model)
model = model.... | {
"login": "sayakpaul",
"id": 22957388,
"node_id": "MDQ6VXNlcjIyOTU3Mzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/22957388?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sayakpaul",
"html_url": "https://github.com/sayakpaul",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1253/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1253/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1249 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1249/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1249/comments | https://api.github.com/repos/huggingface/peft/issues/1249/events | https://github.com/huggingface/peft/issues/1249 | 2,035,073,773 | I_kwDOIf9iDM55TMLt | 1,249 | About the dtype of trainable params | {
"login": "hiyouga",
"id": 16256802,
"node_id": "MDQ6VXNlcjE2MjU2ODAy",
"avatar_url": "https://avatars.githubusercontent.com/u/16256802?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hiyouga",
"html_url": "https://github.com/hiyouga",
"followers_url": "https://api.github.com/users/hiyoug... | [] | closed | false | null | [] | null | 19 | 2023-12-11T08:14:36 | 2024-01-21T15:03:42 | 2024-01-21T15:03:42 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft 0.7.0
transformers 4.34.0
torch 2.0.1
### Who can help?
@pacman100 @younesbelkada @sayakpaul
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1249/reactions",
"total_count": 2,
"+1": 2,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1249/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1248 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1248/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1248/comments | https://api.github.com/repos/huggingface/peft/issues/1248/events | https://github.com/huggingface/peft/issues/1248 | 2,034,999,381 | I_kwDOIf9iDM55S6BV | 1,248 | Type Error: '<=' not supported between instances of 'LoraConfig' and 'int' | {
"login": "qy1026",
"id": 49086900,
"node_id": "MDQ6VXNlcjQ5MDg2OTAw",
"avatar_url": "https://avatars.githubusercontent.com/u/49086900?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/qy1026",
"html_url": "https://github.com/qy1026",
"followers_url": "https://api.github.com/users/qy1026/fo... | [] | closed | false | null | [] | null | 1 | 2023-12-11T07:25:06 | 2023-12-12T06:38:41 | 2023-12-12T06:38:41 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | peft = 0.7.0
When I use the [ unsloth](https://github.com/unslothai/unsloth/tree/main) github repo to faster my qlora finetune code, a TypeError would occur. And the error refers to: https://github.com/huggingface/peft/blob/504d3c832968358498f02b2beac032ed95ef8467/src/peft/tuners/lora/bnb.py#L202C13-L202C92
`self... | {
"login": "qy1026",
"id": 49086900,
"node_id": "MDQ6VXNlcjQ5MDg2OTAw",
"avatar_url": "https://avatars.githubusercontent.com/u/49086900?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/qy1026",
"html_url": "https://github.com/qy1026",
"followers_url": "https://api.github.com/users/qy1026/fo... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1248/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1248/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1247 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1247/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1247/comments | https://api.github.com/repos/huggingface/peft/issues/1247/events | https://github.com/huggingface/peft/issues/1247 | 2,034,717,537 | I_kwDOIf9iDM55R1Nh | 1,247 | How to save parameters in prompt_encoder layers in p-tuning? | {
"login": "lyt719",
"id": 58416622,
"node_id": "MDQ6VXNlcjU4NDE2NjIy",
"avatar_url": "https://avatars.githubusercontent.com/u/58416622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lyt719",
"html_url": "https://github.com/lyt719",
"followers_url": "https://api.github.com/users/lyt719/fo... | [] | closed | false | null | [] | null | 1 | 2023-12-11T02:44:59 | 2024-01-19T15:03:32 | 2024-01-19T15:03:32 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I want to resume training from checkpoint in p-tuning, but the model only save parameters in prompt_embeddings.
<img width="370" alt="image" src="https://github.com/huggingface/peft/assets/58416622/a085224f-32f2-409c-9a51-77c7438bc6a2">
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1247/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1247/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1245 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1245/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1245/comments | https://api.github.com/repos/huggingface/peft/issues/1245/events | https://github.com/huggingface/peft/issues/1245 | 2,034,198,932 | I_kwDOIf9iDM55P2mU | 1,245 | [Question] What is the main difference between "AutoModelForCasualLM" and "PeftModelForCausalLM"? | {
"login": "daehuikim",
"id": 40377750,
"node_id": "MDQ6VXNlcjQwMzc3NzUw",
"avatar_url": "https://avatars.githubusercontent.com/u/40377750?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/daehuikim",
"html_url": "https://github.com/daehuikim",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 1 | 2023-12-10T03:08:54 | 2023-12-11T11:15:25 | 2023-12-11T11:15:24 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Because This is is related to "transformers". Therefore I wrote this question in transformers repo either.
issue is here in transformers(https://github.com/huggingface/transformers/issues/27928)
Hello, Sorry for naive question.
I noticed that the``model.generate()`` function performed differently when inferrence r... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1245/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1245/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1240 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1240/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1240/comments | https://api.github.com/repos/huggingface/peft/issues/1240/events | https://github.com/huggingface/peft/issues/1240 | 2,032,105,873 | I_kwDOIf9iDM55H3mR | 1,240 | [LoftQConfig + LoraConfig] throws size matmul mismatch error | {
"login": "SoundProvider",
"id": 48939336,
"node_id": "MDQ6VXNlcjQ4OTM5MzM2",
"avatar_url": "https://avatars.githubusercontent.com/u/48939336?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SoundProvider",
"html_url": "https://github.com/SoundProvider",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 8 | 2023-12-08T07:42:37 | 2024-01-10T12:48:19 | 2023-12-11T11:16:31 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
- docker image: `pytorch/pytorch:2.1.1-cuda12.1-cudnn8-devel`
- pip list
```
Package Version
----------------------- ------------
accelerate 0.25.0
aiohttp 3.9.1
aiosignal 1.3.1
asttokens 2.0.5
astunparse 1... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1240/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1240/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1239 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1239/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1239/comments | https://api.github.com/repos/huggingface/peft/issues/1239/events | https://github.com/huggingface/peft/issues/1239 | 2,032,021,529 | I_kwDOIf9iDM55HjAZ | 1,239 | error while loading lora | {
"login": "chris111110",
"id": 35104867,
"node_id": "MDQ6VXNlcjM1MTA0ODY3",
"avatar_url": "https://avatars.githubusercontent.com/u/35104867?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chris111110",
"html_url": "https://github.com/chris111110",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 2 | 2023-12-08T06:42:04 | 2024-04-23T17:11:10 | 2023-12-11T10:08:50 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
A800, multiple loras
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
```p... | {
"login": "chris111110",
"id": 35104867,
"node_id": "MDQ6VXNlcjM1MTA0ODY3",
"avatar_url": "https://avatars.githubusercontent.com/u/35104867?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chris111110",
"html_url": "https://github.com/chris111110",
"followers_url": "https://api.github.com/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1239/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/peft/issues/1239/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1235 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1235/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1235/comments | https://api.github.com/repos/huggingface/peft/issues/1235/events | https://github.com/huggingface/peft/issues/1235 | 2,031,479,728 | I_kwDOIf9iDM55Feuw | 1,235 | AttributeError: 'Linear4bit' object has no attribute 'compute_dtype' | {
"login": "Margaretmmh",
"id": 97282300,
"node_id": "U_kgDOBcxo_A",
"avatar_url": "https://avatars.githubusercontent.com/u/97282300?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Margaretmmh",
"html_url": "https://github.com/Margaretmmh",
"followers_url": "https://api.github.com/users/Ma... | [] | closed | false | null | [] | null | 9 | 2023-12-07T20:31:01 | 2024-01-08T05:56:10 | 2024-01-08T05:55:50 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
System: Google colab V100 High RAM
### Who can help?
@pacman100 @younesbelkada @sayakpaul
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
### R... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1235/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1235/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1229 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1229/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1229/comments | https://api.github.com/repos/huggingface/peft/issues/1229/events | https://github.com/huggingface/peft/issues/1229 | 2,030,039,614 | I_kwDOIf9iDM54__I- | 1,229 | Is there a TaskType just for origin model, not other fixed task. | {
"login": "EeyoreLee",
"id": 49790022,
"node_id": "MDQ6VXNlcjQ5NzkwMDIy",
"avatar_url": "https://avatars.githubusercontent.com/u/49790022?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/EeyoreLee",
"html_url": "https://github.com/EeyoreLee",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 2 | 2023-12-07T07:00:09 | 2023-12-07T08:06:42 | 2023-12-07T08:06:41 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
I'm training a reward model and I just need Lora some linear for the base model. It seems that any one of TaskType is suitable.
### Motivation
no
### Your contribution
no | {
"login": "EeyoreLee",
"id": 49790022,
"node_id": "MDQ6VXNlcjQ5NzkwMDIy",
"avatar_url": "https://avatars.githubusercontent.com/u/49790022?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/EeyoreLee",
"html_url": "https://github.com/EeyoreLee",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1229/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1229/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1228 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1228/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1228/comments | https://api.github.com/repos/huggingface/peft/issues/1228/events | https://github.com/huggingface/peft/issues/1228 | 2,029,805,276 | I_kwDOIf9iDM54_F7c | 1,228 | Using Trainer to store checkpoint raise NotADirectoryError. | {
"login": "ZionDoki",
"id": 16078801,
"node_id": "MDQ6VXNlcjE2MDc4ODAx",
"avatar_url": "https://avatars.githubusercontent.com/u/16078801?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ZionDoki",
"html_url": "https://github.com/ZionDoki",
"followers_url": "https://api.github.com/users/Zio... | [
{
"id": 5148671019,
"node_id": "LA_kwDOIf9iDM8AAAABMuJ8Kw",
"url": "https://api.github.com/repos/huggingface/peft/labels/solved",
"name": "solved",
"color": "d4c5f9",
"default": false,
"description": "solved"
}
] | closed | false | null | [] | null | 2 | 2023-12-07T03:21:47 | 2024-01-14T15:03:28 | 2024-01-14T15:03:28 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
**error message**:
```shell
Traceback (most recent call last):
File "/workspace/xxx/make/run_peft.py", line 252, in <module>
trainer.train()
File "/home/xxx/anaconda3/envs/llama/lib/python3.10/site-packages/transformers/trainer.py", line 1555, in train
return inner_training_loop... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1228/reactions",
"total_count": 3,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 3,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1228/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1226 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1226/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1226/comments | https://api.github.com/repos/huggingface/peft/issues/1226/events | https://github.com/huggingface/peft/issues/1226 | 2,027,650,006 | I_kwDOIf9iDM5423vW | 1,226 | Enable Adapter is not working exactly the same as when the adapter is merged to the Codellama Model | {
"login": "akhilamartia",
"id": 54582540,
"node_id": "MDQ6VXNlcjU0NTgyNTQw",
"avatar_url": "https://avatars.githubusercontent.com/u/54582540?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/akhilamartia",
"html_url": "https://github.com/akhilamartia",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | null | 11 | 2023-12-06T04:51:44 | 2024-06-24T05:00:59 | 2024-01-30T15:03:38 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
the codellama model from hugging face :(https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) is Peft Lora tuned to have an adapter.
So i have 2 versions, one where i add the adapter to the model dynamically, enable/set it and test it with a sample code completion
```python
BASE_MODEL_ID ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1226/reactions",
"total_count": 2,
"+1": 2,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1226/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1225 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1225/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1225/comments | https://api.github.com/repos/huggingface/peft/issues/1225/events | https://github.com/huggingface/peft/issues/1225 | 2,027,606,409 | I_kwDOIf9iDM542tGJ | 1,225 | ModulesToSaveWrapper has no 'weight' attritude | {
"login": "zhangsheng377",
"id": 3692247,
"node_id": "MDQ6VXNlcjM2OTIyNDc=",
"avatar_url": "https://avatars.githubusercontent.com/u/3692247?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhangsheng377",
"html_url": "https://github.com/zhangsheng377",
"followers_url": "https://api.github.... | [] | closed | false | null | [] | null | 5 | 2023-12-06T04:04:09 | 2023-12-19T18:10:44 | 2023-12-19T18:10:43 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft=0.4.0
model=Megatron's GPTModel
### Who can help?
@BenjaminBossan
We are recently using peft=0.4.0+Megatron to run GPTModel. There is a code of the model `self.language_model.output_layer.weight`, which directly takes the weight of the output_layer, but the output_layer has already been `modu... | {
"login": "zhangsheng377",
"id": 3692247,
"node_id": "MDQ6VXNlcjM2OTIyNDc=",
"avatar_url": "https://avatars.githubusercontent.com/u/3692247?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhangsheng377",
"html_url": "https://github.com/zhangsheng377",
"followers_url": "https://api.github.... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1225/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1225/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1223 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1223/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1223/comments | https://api.github.com/repos/huggingface/peft/issues/1223/events | https://github.com/huggingface/peft/issues/1223 | 2,027,414,469 | I_kwDOIf9iDM541-PF | 1,223 | Whisper V3 PEFT finetuning is broken | {
"login": "nshmyrev",
"id": 2886672,
"node_id": "MDQ6VXNlcjI4ODY2NzI=",
"avatar_url": "https://avatars.githubusercontent.com/u/2886672?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nshmyrev",
"html_url": "https://github.com/nshmyrev",
"followers_url": "https://api.github.com/users/nshmy... | [] | closed | false | null | [] | null | 6 | 2023-12-06T00:46:15 | 2023-12-09T14:46:00 | 2023-12-06T05:58:49 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
transformers 4.36.0.dev0,
peft 0.6.3.dev0
accelerate 0.25.0
bitsandbytes 0.41.2.post2
### Who can help?
@sanchit-gandhi
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] A... | {
"login": "nshmyrev",
"id": 2886672,
"node_id": "MDQ6VXNlcjI4ODY2NzI=",
"avatar_url": "https://avatars.githubusercontent.com/u/2886672?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nshmyrev",
"html_url": "https://github.com/nshmyrev",
"followers_url": "https://api.github.com/users/nshmy... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1223/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1223/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1222 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1222/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1222/comments | https://api.github.com/repos/huggingface/peft/issues/1222/events | https://github.com/huggingface/peft/issues/1222 | 2,026,830,843 | I_kwDOIf9iDM54zvv7 | 1,222 | get_peft_model for quantized EsmForSequenceClassification fails due to "only Tensors of floating point dtype can require gradients" | {
"login": "brianloyal",
"id": 94650726,
"node_id": "U_kgDOBaRBZg",
"avatar_url": "https://avatars.githubusercontent.com/u/94650726?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/brianloyal",
"html_url": "https://github.com/brianloyal",
"followers_url": "https://api.github.com/users/brian... | [] | closed | false | null | [] | null | 3 | 2023-12-05T18:05:22 | 2024-03-25T11:40:38 | 2023-12-05T22:47:13 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
python==3.10.8
accelerate==0.25.0
bitsandbytes==0.41.2.post2
datasets==2.15.0
peft==0.6.2
torch==2.1.1
transformers==4.35.2
ml.g5.2xlarge SageMaker notebook instance (8 vCPU + 32 GB Memory + 1 A10G GPU)
Amazon Linux 2
JupyterLab 3
### Who can help?
@BenjaminBossan
### Informati... | {
"login": "brianloyal",
"id": 94650726,
"node_id": "U_kgDOBaRBZg",
"avatar_url": "https://avatars.githubusercontent.com/u/94650726?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/brianloyal",
"html_url": "https://github.com/brianloyal",
"followers_url": "https://api.github.com/users/brian... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1222/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1222/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1209 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1209/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1209/comments | https://api.github.com/repos/huggingface/peft/issues/1209/events | https://github.com/huggingface/peft/issues/1209 | 2,024,403,755 | I_kwDOIf9iDM54qfMr | 1,209 | 'Linear4bit' object has no attribute 'compute_dtype' | {
"login": "d1pankarmedhi",
"id": 136924835,
"node_id": "U_kgDOCClOow",
"avatar_url": "https://avatars.githubusercontent.com/u/136924835?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/d1pankarmedhi",
"html_url": "https://github.com/d1pankarmedhi",
"followers_url": "https://api.github.com/... | [] | closed | false | {
"login": "Titus-von-Koeller",
"id": 9048635,
"node_id": "MDQ6VXNlcjkwNDg2MzU=",
"avatar_url": "https://avatars.githubusercontent.com/u/9048635?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Titus-von-Koeller",
"html_url": "https://github.com/Titus-von-Koeller",
"followers_url": "https:/... | [
{
"login": "Titus-von-Koeller",
"id": 9048635,
"node_id": "MDQ6VXNlcjkwNDg2MzU=",
"avatar_url": "https://avatars.githubusercontent.com/u/9048635?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Titus-von-Koeller",
"html_url": "https://github.com/Titus-von-Koeller",
"foll... | null | 17 | 2023-12-04T17:48:43 | 2024-02-17T15:04:03 | 2024-02-17T15:04:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
**Instance**: T4 (Colab)
Occured while running the example notebook `bnb-4bit-training-with-inference.ipynb` forQLora training.
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `ex... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1209/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1209/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1206 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1206/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1206/comments | https://api.github.com/repos/huggingface/peft/issues/1206/events | https://github.com/huggingface/peft/issues/1206 | 2,022,464,601 | I_kwDOIf9iDM54jFxZ | 1,206 | ERROR "Target modules not found in the base model" in PEFT-0.6.2 | {
"login": "chenjunyi1999",
"id": 49756696,
"node_id": "MDQ6VXNlcjQ5NzU2Njk2",
"avatar_url": "https://avatars.githubusercontent.com/u/49756696?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chenjunyi1999",
"html_url": "https://github.com/chenjunyi1999",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 3 | 2023-12-03T10:49:06 | 2023-12-04T11:21:26 | 2023-12-04T11:21:25 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
python == 3.8.10
torch == 2.0.0+cu118
transformers == 4.35.2
peft == 0.6.2
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [x] My own task or dataset (give... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1206/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1206/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1203 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1203/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1203/comments | https://api.github.com/repos/huggingface/peft/issues/1203/events | https://github.com/huggingface/peft/issues/1203 | 2,019,279,706 | I_kwDOIf9iDM54W8Na | 1,203 | Got error when trying to merge qlora to model | {
"login": "khiempm209",
"id": 119430607,
"node_id": "U_kgDOBx5dzw",
"avatar_url": "https://avatars.githubusercontent.com/u/119430607?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/khiempm209",
"html_url": "https://github.com/khiempm209",
"followers_url": "https://api.github.com/users/khi... | [] | closed | false | null | [] | null | 8 | 2023-11-30T18:31:36 | 2024-02-06T14:31:25 | 2023-12-18T09:27:45 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
When i tried to merge qlora to model, i got this error. It seem some arguments are not added to the function
```
File /opt/conda/lib/python3.10/site-packages/peft/peft_model.py:304, in PeftModel.from_pretrained(cls, model, model_id, adapter_name, is_trainable, config, **kwargs)
302 # load t... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1203/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1203/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1199 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1199/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1199/comments | https://api.github.com/repos/huggingface/peft/issues/1199/events | https://github.com/huggingface/peft/issues/1199 | 2,018,424,593 | I_kwDOIf9iDM54TrcR | 1,199 | [`torch.compile()`] fails on PEFT LoRA in `diffusers` | {
"login": "sayakpaul",
"id": 22957388,
"node_id": "MDQ6VXNlcjIyOTU3Mzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/22957388?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sayakpaul",
"html_url": "https://github.com/sayakpaul",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 15 | 2023-11-30T11:18:10 | 2024-01-05T01:37:42 | 2023-12-11T06:18:04 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ```bash
Unsupported: call_method ListIteratorVariable() __contains__ [ConstantVariable(str)] {}
from user code:
File "/usr/local/lib/python3.10/dist-packages/diffusers/models/unet_2d_condition.py", line 1047, in forward
scale_lora_layers(self, lora_scale)
File "/usr/local/lib/python3.10/dist-packages/di... | {
"login": "sayakpaul",
"id": 22957388,
"node_id": "MDQ6VXNlcjIyOTU3Mzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/22957388?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sayakpaul",
"html_url": "https://github.com/sayakpaul",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1199/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1199/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1198 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1198/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1198/comments | https://api.github.com/repos/huggingface/peft/issues/1198/events | https://github.com/huggingface/peft/issues/1198 | 2,018,014,440 | I_kwDOIf9iDM54SHTo | 1,198 | Potentially a bug | {
"login": "ngocbh",
"id": 14000067,
"node_id": "MDQ6VXNlcjE0MDAwMDY3",
"avatar_url": "https://avatars.githubusercontent.com/u/14000067?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ngocbh",
"html_url": "https://github.com/ngocbh",
"followers_url": "https://api.github.com/users/ngocbh/fo... | [] | closed | false | null | [] | null | 3 | 2023-11-30T07:07:20 | 2023-12-11T11:35:56 | 2023-12-11T11:35:56 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | https://github.com/huggingface/peft/blob/2b901ee57230559aaf39867c7698f6aca3617162/src/peft/tuners/tuners_utils.py#L260C32-L260C32
inject_adapters have its own `model` argument but, in line 260 and 263, it accesses model parameters using `self.model`. I guess it's not a safe choice. | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1198/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1198/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1197 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1197/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1197/comments | https://api.github.com/repos/huggingface/peft/issues/1197/events | https://github.com/huggingface/peft/issues/1197 | 2,017,834,501 | I_kwDOIf9iDM54RbYF | 1,197 | Error while working with custom model in trainer | {
"login": "funger1990",
"id": 13957946,
"node_id": "MDQ6VXNlcjEzOTU3OTQ2",
"avatar_url": "https://avatars.githubusercontent.com/u/13957946?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/funger1990",
"html_url": "https://github.com/funger1990",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 7 | 2023-11-30T04:07:14 | 2024-01-02T10:02:18 | 2024-01-02T10:02:18 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
python=3.8.16
torch=1.13.0+cu117
transformers=4.33.2
peft=0.6.2
accelerate=0.23.0
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [x] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [x] My o... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1197/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1197/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1194 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1194/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1194/comments | https://api.github.com/repos/huggingface/peft/issues/1194/events | https://github.com/huggingface/peft/issues/1194 | 2,016,617,893 | I_kwDOIf9iDM54MyWl | 1,194 | Figure out failing stable diffusion tests | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | [
{
"id": 4838806417,
"node_id": "LA_kwDOIf9iDM8AAAABIGpTkQ",
"url": "https://api.github.com/repos/huggingface/peft/labels/bug",
"name": "bug",
"color": "d73a4a",
"default": true,
"description": "Something isn't working"
},
{
"id": 4838806438,
"node_id": "LA_kwDOIf9iDM8AAAABIGp... | closed | false | null | [] | null | 2 | 2023-11-29T13:38:04 | 2023-12-05T17:12:41 | 2023-12-05T17:12:41 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Tests may only fail on Windows, only with CPU, or only for specific PyTorch versions.
### Who can help?
Anyone.
### Information
- [x] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1194/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1194/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1192 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1192/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1192/comments | https://api.github.com/repos/huggingface/peft/issues/1192/events | https://github.com/huggingface/peft/issues/1192 | 2,015,882,289 | I_kwDOIf9iDM54J-wx | 1,192 | model_kwargs["input_ids"].shape[0], peft_config.num_virtual_tokens KeyError: 'input_ids' | {
"login": "mengqianshuang",
"id": 79396498,
"node_id": "MDQ6VXNlcjc5Mzk2NDk4",
"avatar_url": "https://avatars.githubusercontent.com/u/79396498?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mengqianshuang",
"html_url": "https://github.com/mengqianshuang",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | null | 3 | 2023-11-29T06:05:51 | 2023-11-29T12:44:06 | 2023-11-29T12:44:06 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft 0.6.2
pytorch 2.0.1
transformers 4.35.2
python 3.11
### Who can help?
_No response_
### Information
- [x] The official example scripts
- [x] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
###... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1192/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1192/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1188 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1188/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1188/comments | https://api.github.com/repos/huggingface/peft/issues/1188/events | https://github.com/huggingface/peft/issues/1188 | 2,011,111,423 | I_kwDOIf9iDM533x__ | 1,188 | prefix tuning for encoder-decoder models in peft different from original implementation? | {
"login": "mickeymickeymonkey",
"id": 142171154,
"node_id": "U_kgDOCHlcEg",
"avatar_url": "https://avatars.githubusercontent.com/u/142171154?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mickeymickeymonkey",
"html_url": "https://github.com/mickeymickeymonkey",
"followers_url": "https://... | [] | closed | false | null | [] | null | 2 | 2023-11-26T16:41:57 | 2024-01-04T15:03:36 | 2024-01-04T15:03:36 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft==0.6.2
### Who can help?
@pacman @younesbelkada @BenjaminBossan
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
```
mo... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1188/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 2
} | https://api.github.com/repos/huggingface/peft/issues/1188/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1184 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1184/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1184/comments | https://api.github.com/repos/huggingface/peft/issues/1184/events | https://github.com/huggingface/peft/issues/1184 | 2,009,911,148 | I_kwDOIf9iDM53zM9s | 1,184 | LoRA applied to encoder and decoder block of a seq-to-seq model | {
"login": "BSharmi",
"id": 6493020,
"node_id": "MDQ6VXNlcjY0OTMwMjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6493020?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BSharmi",
"html_url": "https://github.com/BSharmi",
"followers_url": "https://api.github.com/users/BSharmi/... | [] | closed | false | null | [] | null | 3 | 2023-11-24T15:39:49 | 2024-01-03T15:03:32 | 2024-01-03T15:03:32 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
Hi there!
Is it possible to take the encoder and decoder block of a seq-to-seq model like T5 and apply LoRA to each module instead of loading an ` AutoModelForSeq2SeqLM` directly?
### Motivation
If we want to have a conditional T5 model where I want to have a conditional module in between ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1184/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1184/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1182 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1182/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1182/comments | https://api.github.com/repos/huggingface/peft/issues/1182/events | https://github.com/huggingface/peft/issues/1182 | 2,009,837,233 | I_kwDOIf9iDM53y66x | 1,182 | Toggle inference_mode during training | {
"login": "Vectorrent",
"id": 94832312,
"node_id": "U_kgDOBacGuA",
"avatar_url": "https://avatars.githubusercontent.com/u/94832312?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Vectorrent",
"html_url": "https://github.com/Vectorrent",
"followers_url": "https://api.github.com/users/Vecto... | [] | closed | false | null | [] | null | 2 | 2023-11-24T14:43:37 | 2023-11-24T16:12:29 | 2023-11-24T16:12:29 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
After loading a model for training with `get_peft_model()`, I would like to have the ability to temporarily put that model into `inference_mode=True` - so I can test generations while actively training.
### Motivation
As part of my training process, I like to periodically log test generations to ... | {
"login": "Vectorrent",
"id": 94832312,
"node_id": "U_kgDOBacGuA",
"avatar_url": "https://avatars.githubusercontent.com/u/94832312?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Vectorrent",
"html_url": "https://github.com/Vectorrent",
"followers_url": "https://api.github.com/users/Vecto... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1182/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1182/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1180 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1180/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1180/comments | https://api.github.com/repos/huggingface/peft/issues/1180/events | https://github.com/huggingface/peft/issues/1180 | 2,009,028,993 | I_kwDOIf9iDM53v1mB | 1,180 | Using QLoRA to fine-tune BaiChuan-13B reports AttributeError: 'Parameter' object has no attribute 'weight' | {
"login": "Aitejiu",
"id": 73458598,
"node_id": "MDQ6VXNlcjczNDU4NTk4",
"avatar_url": "https://avatars.githubusercontent.com/u/73458598?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Aitejiu",
"html_url": "https://github.com/Aitejiu",
"followers_url": "https://api.github.com/users/Aiteji... | [] | closed | false | null | [] | null | 2 | 2023-11-24T03:10:50 | 2023-11-25T13:16:52 | 2023-11-25T13:16:52 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
`peft==0.6.2`
`accelerate==0.24.1`
`transformers==4.33.2`
`platform:linux`
`python==3.9.0`
Using `QLoRA` to fine-tune `BaiChuan-13B` reports `AttributeError: 'Parameter' object has no attribute 'weight'.`
The error location code is
![image](https://github.com/huggingface/peft/assets/73458598/0e1... | {
"login": "Aitejiu",
"id": 73458598,
"node_id": "MDQ6VXNlcjczNDU4NTk4",
"avatar_url": "https://avatars.githubusercontent.com/u/73458598?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Aitejiu",
"html_url": "https://github.com/Aitejiu",
"followers_url": "https://api.github.com/users/Aiteji... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1180/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1180/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1178 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1178/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1178/comments | https://api.github.com/repos/huggingface/peft/issues/1178/events | https://github.com/huggingface/peft/issues/1178 | 2,008,538,416 | I_kwDOIf9iDM53t90w | 1,178 | Weird Result between different implementation of LoRA (theoretically should be the same, but the results are drastically different) | {
"login": "nbasyl",
"id": 45586614,
"node_id": "MDQ6VXNlcjQ1NTg2NjE0",
"avatar_url": "https://avatars.githubusercontent.com/u/45586614?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nbasyl",
"html_url": "https://github.com/nbasyl",
"followers_url": "https://api.github.com/users/nbasyl/fo... | [] | closed | false | null | [] | null | 4 | 2023-11-23T16:28:43 | 2024-01-02T15:03:28 | 2024-01-02T15:03:28 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
transformers == 4.36.0.dev0
torch == 2.1.1
cuda version 12.2
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1178/reactions",
"total_count": 2,
"+1": 2,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1178/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1175 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1175/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1175/comments | https://api.github.com/repos/huggingface/peft/issues/1175/events | https://github.com/huggingface/peft/issues/1175 | 2,007,589,043 | I_kwDOIf9iDM53qWCz | 1,175 | Inquiry about the initialization method of adaption_prompt | {
"login": "callanwu",
"id": 63695429,
"node_id": "MDQ6VXNlcjYzNjk1NDI5",
"avatar_url": "https://avatars.githubusercontent.com/u/63695429?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/callanwu",
"html_url": "https://github.com/callanwu",
"followers_url": "https://api.github.com/users/cal... | [] | closed | false | null | [] | null | 3 | 2023-11-23T06:57:41 | 2023-11-24T11:19:46 | 2023-11-24T11:08:59 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | In current repo, adaption_prompt initializes the tokens with standard normal values by using `nn.Parameter()`
https://github.com/huggingface/peft/blob/dd4771b2f42a72b9d7a54fbf9815530659305718/src/peft/tuners/adaption_prompt/layer.py#L52-L54
The comments about these code is https://github.com/huggingface/peft/blob/dd4... | {
"login": "callanwu",
"id": 63695429,
"node_id": "MDQ6VXNlcjYzNjk1NDI5",
"avatar_url": "https://avatars.githubusercontent.com/u/63695429?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/callanwu",
"html_url": "https://github.com/callanwu",
"followers_url": "https://api.github.com/users/cal... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1175/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1175/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1173 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1173/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1173/comments | https://api.github.com/repos/huggingface/peft/issues/1173/events | https://github.com/huggingface/peft/issues/1173 | 2,007,514,523 | I_kwDOIf9iDM53qD2b | 1,173 | add token for phi-1.5 tokenizer | {
"login": "lastrei",
"id": 32331131,
"node_id": "MDQ6VXNlcjMyMzMxMTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/32331131?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lastrei",
"html_url": "https://github.com/lastrei",
"followers_url": "https://api.github.com/users/lastre... | [] | closed | false | null | [] | null | 5 | 2023-11-23T05:35:06 | 2023-11-27T15:17:31 | 2023-11-27T15:17:31 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
sentencepiece 0.1.99
bitsandbytes 0.41.1
cryptography 40.0.2
peft 0.4.0
transformers 4.30.2
### Who can help?
@pacman100 @younesbelkada
### Information
- [ ] The official example scripts
- [X] My ... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1173/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1173/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1172 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1172/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1172/comments | https://api.github.com/repos/huggingface/peft/issues/1172/events | https://github.com/huggingface/peft/issues/1172 | 2,007,341,149 | I_kwDOIf9iDM53pZhd | 1,172 | change dtype of output after passing through lora_A | {
"login": "huseyinatahaninan",
"id": 89604977,
"node_id": "MDQ6VXNlcjg5NjA0OTc3",
"avatar_url": "https://avatars.githubusercontent.com/u/89604977?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/huseyinatahaninan",
"html_url": "https://github.com/huseyinatahaninan",
"followers_url": "https... | [] | closed | false | null | [] | null | 8 | 2023-11-23T01:24:29 | 2023-12-21T16:33:01 | 2023-12-21T16:33:01 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft 0.6.2
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
I am not sure if this is really a ... | {
"login": "huseyinatahaninan",
"id": 89604977,
"node_id": "MDQ6VXNlcjg5NjA0OTc3",
"avatar_url": "https://avatars.githubusercontent.com/u/89604977?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/huseyinatahaninan",
"html_url": "https://github.com/huseyinatahaninan",
"followers_url": "https... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1172/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1172/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1168 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1168/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1168/comments | https://api.github.com/repos/huggingface/peft/issues/1168/events | https://github.com/huggingface/peft/issues/1168 | 2,005,936,194 | I_kwDOIf9iDM53kChC | 1,168 | TypeError: PeftConfig.__init__() got an unexpected keyword argument '_from_model_config' | {
"login": "GrayZ77",
"id": 76860116,
"node_id": "MDQ6VXNlcjc2ODYwMTE2",
"avatar_url": "https://avatars.githubusercontent.com/u/76860116?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GrayZ77",
"html_url": "https://github.com/GrayZ77",
"followers_url": "https://api.github.com/users/GrayZ7... | [] | closed | false | null | [] | null | 9 | 2023-11-22T09:39:04 | 2024-01-03T15:03:35 | 2024-01-03T15:03:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I was loading my own local lora model in Langchain-Chatchat, and I encountered with this error:
TypeError: PeftConfig.__init__() got an unexpected keyword argument '_from_model_config'
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modifi... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1168/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1168/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1161 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1161/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1161/comments | https://api.github.com/repos/huggingface/peft/issues/1161/events | https://github.com/huggingface/peft/issues/1161 | 2,004,338,796 | I_kwDOIf9iDM53d8hs | 1,161 | I want to use PEFT to load the Lora model and keep the original model | {
"login": "LDLINGLINGLING",
"id": 47373076,
"node_id": "MDQ6VXNlcjQ3MzczMDc2",
"avatar_url": "https://avatars.githubusercontent.com/u/47373076?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LDLINGLINGLING",
"html_url": "https://github.com/LDLINGLINGLING",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | null | 2 | 2023-11-21T13:52:44 | 2023-12-29T15:03:44 | 2023-12-29T15:03:44 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft ,transformer
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
```pyth... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1161/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1161/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1158 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1158/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1158/comments | https://api.github.com/repos/huggingface/peft/issues/1158/events | https://github.com/huggingface/peft/issues/1158 | 2,004,033,389 | I_kwDOIf9iDM53cx9t | 1,158 | Training is successfull, output image does not contain the person whose 10 images i used to train it | {
"login": "nayan-dhabarde",
"id": 24761808,
"node_id": "MDQ6VXNlcjI0NzYxODA4",
"avatar_url": "https://avatars.githubusercontent.com/u/24761808?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nayan-dhabarde",
"html_url": "https://github.com/nayan-dhabarde",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | null | 5 | 2023-11-21T11:10:30 | 2023-12-04T09:38:33 | 2023-11-22T08:21:33 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | command = [f"accelerate launch --config_file \"config.yaml\" train_dreambooth.py \
--pretrained_model_name_or_path=\"SG161222/Realistic_Vision_V4.0_noVAE\" \
--instance_data_dir=\"smart_cropped_images\" \
--class_data_dir=\"{dataset}\" \
--output_dir=\"trained_model\" \
--... | {
"login": "nayan-dhabarde",
"id": 24761808,
"node_id": "MDQ6VXNlcjI0NzYxODA4",
"avatar_url": "https://avatars.githubusercontent.com/u/24761808?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nayan-dhabarde",
"html_url": "https://github.com/nayan-dhabarde",
"followers_url": "https://api.gi... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1158/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1158/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1156 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1156/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1156/comments | https://api.github.com/repos/huggingface/peft/issues/1156/events | https://github.com/huggingface/peft/issues/1156 | 2,003,820,938 | I_kwDOIf9iDM53b-GK | 1,156 | MoLoRA on PEFT | {
"login": "alexrs",
"id": 1732642,
"node_id": "MDQ6VXNlcjE3MzI2NDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/1732642?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alexrs",
"html_url": "https://github.com/alexrs",
"followers_url": "https://api.github.com/users/alexrs/foll... | [] | closed | false | null | [] | null | 12 | 2023-11-21T09:21:57 | 2024-02-20T04:37:38 | 2024-02-07T15:03:48 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
The paper ["Pushing Mixture of Experts to the Limit: Extremely Parameter Efficient MoE for Instruction Tuning
"](https://arxiv.org/abs/2309.05444) introduced MoLoRA, a Mixutre-of-Experts approach using LoRA adapters. I am using it to conduct some research for my MSc thesis, and have implemented it... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1156/reactions",
"total_count": 8,
"+1": 8,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1156/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1155 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1155/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1155/comments | https://api.github.com/repos/huggingface/peft/issues/1155/events | https://github.com/huggingface/peft/issues/1155 | 2,003,678,836 | I_kwDOIf9iDM53bbZ0 | 1,155 | A (possible) bug in lora merging method: add_weighted_adapter | {
"login": "jihuishan",
"id": 151612440,
"node_id": "U_kgDOCQlsGA",
"avatar_url": "https://avatars.githubusercontent.com/u/151612440?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jihuishan",
"html_url": "https://github.com/jihuishan",
"followers_url": "https://api.github.com/users/jihuis... | [] | closed | false | null | [] | null | 8 | 2023-11-21T08:02:44 | 2023-12-04T08:48:32 | 2023-11-22T12:08:33 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft 0.6.1
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [x] My own task or dataset (give details below)
### Reproduction
In both line 565 and 578 of add_we... | {
"login": "jihuishan",
"id": 151612440,
"node_id": "U_kgDOCQlsGA",
"avatar_url": "https://avatars.githubusercontent.com/u/151612440?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jihuishan",
"html_url": "https://github.com/jihuishan",
"followers_url": "https://api.github.com/users/jihuis... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1155/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1155/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1154 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1154/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1154/comments | https://api.github.com/repos/huggingface/peft/issues/1154/events | https://github.com/huggingface/peft/issues/1154 | 2,003,539,383 | I_kwDOIf9iDM53a5W3 | 1,154 | saftetensor in subfolder not supported | {
"login": "yxli2123",
"id": 69247082,
"node_id": "MDQ6VXNlcjY5MjQ3MDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/69247082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yxli2123",
"html_url": "https://github.com/yxli2123",
"followers_url": "https://api.github.com/users/yxl... | [] | closed | false | null | [] | null | 11 | 2023-11-21T06:15:21 | 2024-01-26T15:04:00 | 2024-01-26T15:04:00 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
When I put `adapter_model.safetensor` in a subfolder of a Huggingface Hub, for example, `LoftQ/Llama-2-7b-hf-4bit-64rank`, `PeftModel.from_pretrained("oftQ/Llama-2-7b-hf-4bit-64rank", subfolder='loftq_init')` is not able to find the `adapter_model.safetensor` properly. It only supports `adapter_model... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1154/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1154/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1151 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1151/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1151/comments | https://api.github.com/repos/huggingface/peft/issues/1151/events | https://github.com/huggingface/peft/issues/1151 | 2,000,982,784 | I_kwDOIf9iDM53RJMA | 1,151 | Mention that Mistral is supported as well | {
"login": "osanseviero",
"id": 7246357,
"node_id": "MDQ6VXNlcjcyNDYzNTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/7246357?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/osanseviero",
"html_url": "https://github.com/osanseviero",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | null | 0 | 2023-11-19T18:37:04 | 2023-11-28T10:04:58 | 2023-11-28T10:04:58 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | In https://huggingface.co/docs/peft/index and the README, we list supported models. I think Mistral is supported as well (and probably few others) | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1151/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1151/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1148 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1148/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1148/comments | https://api.github.com/repos/huggingface/peft/issues/1148/events | https://github.com/huggingface/peft/issues/1148 | 2,000,334,032 | I_kwDOIf9iDM53OqzQ | 1,148 | RecursionError: maximum recursion depth exceeded | {
"login": "estuday",
"id": 45581281,
"node_id": "MDQ6VXNlcjQ1NTgxMjgx",
"avatar_url": "https://avatars.githubusercontent.com/u/45581281?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/estuday",
"html_url": "https://github.com/estuday",
"followers_url": "https://api.github.com/users/estuda... | [] | closed | false | null | [] | null | 2 | 2023-11-18T09:15:05 | 2023-12-27T15:03:32 | 2023-12-27T15:03:32 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft==0.6.1
transformers==4.35.0
accelerate==0.24.1
ubuntu22.04
python==3.9
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1148/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1148/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1142 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1142/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1142/comments | https://api.github.com/repos/huggingface/peft/issues/1142/events | https://github.com/huggingface/peft/issues/1142 | 1,998,681,425 | I_kwDOIf9iDM53IXVR | 1,142 | How to do Gradient Checkpoint + LoRA | {
"login": "tcapelle",
"id": 18441985,
"node_id": "MDQ6VXNlcjE4NDQxOTg1",
"avatar_url": "https://avatars.githubusercontent.com/u/18441985?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tcapelle",
"html_url": "https://github.com/tcapelle",
"followers_url": "https://api.github.com/users/tca... | [] | closed | false | null | [] | null | 11 | 2023-11-17T09:34:16 | 2024-04-15T19:08:29 | 2024-01-26T15:04:02 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
<img width="570" alt="image" src="https://github.com/huggingface/peft/assets/18441985/9b3ae040-d78a-477b-a9ec-6ab26b687a68">
### Who can help?
I need help with using LoRA + gradient checkpointing.
Using the reentrant option appears to be the solution, but it slows down training a lot, for LLam... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1142/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1142/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1138 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1138/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1138/comments | https://api.github.com/repos/huggingface/peft/issues/1138/events | https://github.com/huggingface/peft/issues/1138 | 1,997,147,842 | I_kwDOIf9iDM53Cg7C | 1,138 | Linear4bit.merge() got an unexpected keyword argument 'adapter_names' | {
"login": "khiempm209",
"id": 119430607,
"node_id": "U_kgDOBx5dzw",
"avatar_url": "https://avatars.githubusercontent.com/u/119430607?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/khiempm209",
"html_url": "https://github.com/khiempm209",
"followers_url": "https://api.github.com/users/khi... | [] | closed | false | null | [] | null | 2 | 2023-11-16T15:36:19 | 2023-11-16T16:13:17 | 2023-11-16T16:12:40 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I got this error when i use the 'merge_and_unload()' to merge my own trained lora apdapter to the Mistral model. I thought you guys forgot to put the 'adapter_names' argument to the merge function in Linear4bit class.
File /opt/conda/lib/python3.10/site-packages/peft/tuners/lora/model.py:390, in Lor... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1138/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1138/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1134 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1134/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1134/comments | https://api.github.com/repos/huggingface/peft/issues/1134/events | https://github.com/huggingface/peft/issues/1134 | 1,993,857,234 | I_kwDOIf9iDM5219jS | 1,134 | Error occur when load model with PeftModel.from_pretrained on a cpu machine | {
"login": "woaipichuli",
"id": 15848251,
"node_id": "MDQ6VXNlcjE1ODQ4MjUx",
"avatar_url": "https://avatars.githubusercontent.com/u/15848251?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/woaipichuli",
"html_url": "https://github.com/woaipichuli",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 0 | 2023-11-15T01:31:34 | 2023-11-15T03:14:07 | 2023-11-15T03:14:07 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I trained a lora model with peft and saved the checkpoint for inference. When I used the check point for inference on a cpu machine, some error occur and report no cuda and gpu is found.
Traceback (most recent call last):
File "/app_service.py", line 15, in <module>
TEXT_CLASSIFY = TextCla... | {
"login": "woaipichuli",
"id": 15848251,
"node_id": "MDQ6VXNlcjE1ODQ4MjUx",
"avatar_url": "https://avatars.githubusercontent.com/u/15848251?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/woaipichuli",
"html_url": "https://github.com/woaipichuli",
"followers_url": "https://api.github.com/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1134/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1134/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1133 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1133/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1133/comments | https://api.github.com/repos/huggingface/peft/issues/1133/events | https://github.com/huggingface/peft/issues/1133 | 1,993,556,088 | I_kwDOIf9iDM5200B4 | 1,133 | Add Orthogonal Finetuning (OFT) | {
"login": "lukaskuhn-lku",
"id": 23077250,
"node_id": "MDQ6VXNlcjIzMDc3MjUw",
"avatar_url": "https://avatars.githubusercontent.com/u/23077250?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lukaskuhn-lku",
"html_url": "https://github.com/lukaskuhn-lku",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 0 | 2023-11-14T20:55:23 | 2023-11-14T21:43:54 | 2023-11-14T21:43:54 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
The paper [Controlling Text-to-Image Diffusion by Orthogonal Finetuning](https://arxiv.org/abs/2306.07280) proposes a new method of fine-tuning text-to-image diffusion models by using multiple learned orthogonal transformations on the layers of the pretrained model.
![](https://oft.wyliu.com/... | {
"login": "lukaskuhn-lku",
"id": 23077250,
"node_id": "MDQ6VXNlcjIzMDc3MjUw",
"avatar_url": "https://avatars.githubusercontent.com/u/23077250?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lukaskuhn-lku",
"html_url": "https://github.com/lukaskuhn-lku",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1133/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1133/timeline | null | not_planned | false |
https://api.github.com/repos/huggingface/peft/issues/1127 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1127/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1127/comments | https://api.github.com/repos/huggingface/peft/issues/1127/events | https://github.com/huggingface/peft/issues/1127 | 1,991,866,535 | I_kwDOIf9iDM52uXin | 1,127 | Clarify the best practice to continue training after loading from a lora checkpoint | {
"login": "hanyin88",
"id": 59374800,
"node_id": "MDQ6VXNlcjU5Mzc0ODAw",
"avatar_url": "https://avatars.githubusercontent.com/u/59374800?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hanyin88",
"html_url": "https://github.com/hanyin88",
"followers_url": "https://api.github.com/users/han... | [] | closed | false | null | [] | null | 12 | 2023-11-14T02:30:36 | 2024-07-07T18:12:04 | 2024-03-20T01:22:36 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
N/A
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
N/A
### Expected b... | {
"login": "sayakpaul",
"id": 22957388,
"node_id": "MDQ6VXNlcjIyOTU3Mzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/22957388?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sayakpaul",
"html_url": "https://github.com/sayakpaul",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1127/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1127/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1124 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1124/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1124/comments | https://api.github.com/repos/huggingface/peft/issues/1124/events | https://github.com/huggingface/peft/issues/1124 | 1,989,824,289 | I_kwDOIf9iDM52mk8h | 1,124 | Prompt distillation loss in MPT | {
"login": "junzhang-zj",
"id": 52874161,
"node_id": "MDQ6VXNlcjUyODc0MTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/52874161?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/junzhang-zj",
"html_url": "https://github.com/junzhang-zj",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 2 | 2023-11-13T03:33:03 | 2023-11-21T03:10:15 | 2023-11-21T03:10:15 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I have trouble positioning the **distillation process** in the source training of MPT. I observed in the test that the output loss is only the classification loss of the T5 model. I want to confirm whether this part has not been implemented or if I have missed it. Thank you very much! | {
"login": "junzhang-zj",
"id": 52874161,
"node_id": "MDQ6VXNlcjUyODc0MTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/52874161?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/junzhang-zj",
"html_url": "https://github.com/junzhang-zj",
"followers_url": "https://api.github.com/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1124/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1124/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1120 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1120/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1120/comments | https://api.github.com/repos/huggingface/peft/issues/1120/events | https://github.com/huggingface/peft/issues/1120 | 1,989,101,280 | I_kwDOIf9iDM52j0bg | 1,120 | compute_metrics() does not seem to be called when fine-tuning with peft LoRA | {
"login": "thepluck",
"id": 54402720,
"node_id": "MDQ6VXNlcjU0NDAyNzIw",
"avatar_url": "https://avatars.githubusercontent.com/u/54402720?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thepluck",
"html_url": "https://github.com/thepluck",
"followers_url": "https://api.github.com/users/the... | [] | closed | false | null | [] | null | 14 | 2023-11-11T20:03:25 | 2024-09-30T16:41:16 | 2023-12-21T15:03:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
transformers==4.35.0
peft==0.6.1
python==3.9
WSL 2
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details b... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1120/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1120/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1119 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1119/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1119/comments | https://api.github.com/repos/huggingface/peft/issues/1119/events | https://github.com/huggingface/peft/issues/1119 | 1,989,041,625 | I_kwDOIf9iDM52jl3Z | 1,119 | AttributeError: module 'numpy.linalg._umath_linalg' has no attribute '_ilp64' | {
"login": "gunaxprofessional",
"id": 66107066,
"node_id": "MDQ6VXNlcjY2MTA3MDY2",
"avatar_url": "https://avatars.githubusercontent.com/u/66107066?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gunaxprofessional",
"html_url": "https://github.com/gunaxprofessional",
"followers_url": "https... | [] | closed | false | null | [] | null | 7 | 2023-11-11T17:02:18 | 2024-03-13T08:47:10 | 2023-12-18T11:33:29 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Google Colab Free Version
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1119/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1119/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1118 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1118/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1118/comments | https://api.github.com/repos/huggingface/peft/issues/1118/events | https://github.com/huggingface/peft/issues/1118 | 1,988,641,738 | I_kwDOIf9iDM52iEPK | 1,118 | Don’tfix | {
"login": "kizzaBillionz",
"id": 150114830,
"node_id": "U_kgDOCPKSDg",
"avatar_url": "https://avatars.githubusercontent.com/u/150114830?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kizzaBillionz",
"html_url": "https://github.com/kizzaBillionz",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 1 | 2023-11-11T01:47:59 | 2023-11-13T20:10:40 | 2023-11-13T20:10:40 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Trading views… imposing others privacy invalid
### Who can help?
Kizza_billionz
### Information
- [X] The official example scripts
- [X] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
### Reproductio... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1118/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1118/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1114 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1114/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1114/comments | https://api.github.com/repos/huggingface/peft/issues/1114/events | https://github.com/huggingface/peft/issues/1114 | 1,987,947,041 | I_kwDOIf9iDM52faoh | 1,114 | peft==0.6.1 implicitly requires transformers>=4.30.0, but transformers version is not specified | {
"login": "minotru",
"id": 6318579,
"node_id": "MDQ6VXNlcjYzMTg1Nzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/6318579?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/minotru",
"html_url": "https://github.com/minotru",
"followers_url": "https://api.github.com/users/minotru/... | [] | closed | false | null | [] | null | 4 | 2023-11-10T16:25:22 | 2023-11-15T09:54:07 | 2023-11-10T17:37:39 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Hi peft community,
We use `peft` in our project and yesterday tests started to fail in our CI with an error like:
```
/opt/hostedtoolcache/Python/3.10.13/x64/lib/python3.10/site-packages/peft/peft_model.py:35: in <module>
from transformers.pytorch_utils import id_tensor_storage
E ImportE... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1114/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1114/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1113 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1113/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1113/comments | https://api.github.com/repos/huggingface/peft/issues/1113/events | https://github.com/huggingface/peft/issues/1113 | 1,987,934,840 | I_kwDOIf9iDM52fXp4 | 1,113 | AdaLora + bnb not working | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | [] | closed | false | null | [] | null | 0 | 2023-11-10T16:17:55 | 2023-11-17T14:18:36 | 2023-11-17T14:18:36 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
-
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
The issue is this line:... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1113/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1113/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1110 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1110/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1110/comments | https://api.github.com/repos/huggingface/peft/issues/1110/events | https://github.com/huggingface/peft/issues/1110 | 1,986,859,775 | I_kwDOIf9iDM52bRL_ | 1,110 | More fine-grained control on LoRA | {
"login": "HenryCai11",
"id": 24936331,
"node_id": "MDQ6VXNlcjI0OTM2MzMx",
"avatar_url": "https://avatars.githubusercontent.com/u/24936331?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HenryCai11",
"html_url": "https://github.com/HenryCai11",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 1 | 2023-11-10T04:32:46 | 2023-11-10T05:58:13 | 2023-11-10T05:58:12 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
In my case,
- I want to customize learning rates for different module I want to tune with LoRA. I think this one is relevant to issue #1006
- Also, I want to only train specific layers, say, something like `model.layer.1.self_attn.q` (not sure if the q_value is represented this way)
### Motivat... | {
"login": "HenryCai11",
"id": 24936331,
"node_id": "MDQ6VXNlcjI0OTM2MzMx",
"avatar_url": "https://avatars.githubusercontent.com/u/24936331?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HenryCai11",
"html_url": "https://github.com/HenryCai11",
"followers_url": "https://api.github.com/use... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1110/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1110/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1108 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1108/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1108/comments | https://api.github.com/repos/huggingface/peft/issues/1108/events | https://github.com/huggingface/peft/issues/1108 | 1,986,791,318 | I_kwDOIf9iDM52bAeW | 1,108 | install perft get error by PyYAML-5.4.1 | {
"login": "631068264",
"id": 8144089,
"node_id": "MDQ6VXNlcjgxNDQwODk=",
"avatar_url": "https://avatars.githubusercontent.com/u/8144089?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/631068264",
"html_url": "https://github.com/631068264",
"followers_url": "https://api.github.com/users/63... | [] | closed | false | null | [] | null | 1 | 2023-11-10T03:32:05 | 2023-11-13T02:36:34 | 2023-11-13T02:36:34 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
python 3.10
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
```
Collecting pyyaml (from peft... | {
"login": "631068264",
"id": 8144089,
"node_id": "MDQ6VXNlcjgxNDQwODk=",
"avatar_url": "https://avatars.githubusercontent.com/u/8144089?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/631068264",
"html_url": "https://github.com/631068264",
"followers_url": "https://api.github.com/users/63... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1108/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1108/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1098 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1098/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1098/comments | https://api.github.com/repos/huggingface/peft/issues/1098/events | https://github.com/huggingface/peft/issues/1098 | 1,984,267,442 | I_kwDOIf9iDM52RYSy | 1,098 | Support for Ladder Side Tuning (LST) | {
"login": "yeshwanthv5",
"id": 8533303,
"node_id": "MDQ6VXNlcjg1MzMzMDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/8533303?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yeshwanthv5",
"html_url": "https://github.com/yeshwanthv5",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | null | 2 | 2023-11-08T19:33:13 | 2023-12-18T15:04:29 | 2023-12-18T15:04:29 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
Hello
Are there any plans for integrating Ladder Side Tuning (LST) into this library?
References
Repo: https://github.com/ylsung/Ladder-Side-Tuning/tree/main
Paper: https://arxiv.org/abs/2206.06522
### Motivation
The paper proposes a method that is more efficient in terms of GPU memory ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1098/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1098/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1097 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1097/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1097/comments | https://api.github.com/repos/huggingface/peft/issues/1097/events | https://github.com/huggingface/peft/issues/1097 | 1,983,524,375 | I_kwDOIf9iDM52Oi4X | 1,097 | ValueError: Target module QuantLinear() is not supported. | {
"login": "manideep-paa",
"id": 97121476,
"node_id": "U_kgDOBcn0xA",
"avatar_url": "https://avatars.githubusercontent.com/u/97121476?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/manideep-paa",
"html_url": "https://github.com/manideep-paa",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | null | 4 | 2023-11-08T12:44:26 | 2023-12-17T15:03:33 | 2023-12-17T15:03:33 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Hi,I was able to finetune a GPTQ model till 2 days back, but suddenly I have been getting the following error when executing the line: model = get_peft_model(model, config)
ERROR:
ValueError: Target module QuantLinear() is not supported. Currently, only the following modules are supported: `torch.n... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1097/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1097/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1096 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1096/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1096/comments | https://api.github.com/repos/huggingface/peft/issues/1096/events | https://github.com/huggingface/peft/issues/1096 | 1,983,466,563 | I_kwDOIf9iDM52OUxD | 1,096 | Failed to load PeftModel with a pretrained LLM and a trained qlora | {
"login": "NgoDoanKien",
"id": 110718429,
"node_id": "U_kgDOBplt3Q",
"avatar_url": "https://avatars.githubusercontent.com/u/110718429?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/NgoDoanKien",
"html_url": "https://github.com/NgoDoanKien",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 2 | 2023-11-08T12:15:02 | 2023-12-17T15:03:35 | 2023-12-17T15:03:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
All imported ones are latest
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My own task or dataset (give details below)
### Reproduction... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1096/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1096/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1095 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1095/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1095/comments | https://api.github.com/repos/huggingface/peft/issues/1095/events | https://github.com/huggingface/peft/issues/1095 | 1,982,908,865 | I_kwDOIf9iDM52MMnB | 1,095 | Error while loading from PeftModel. | {
"login": "pranavbhat12",
"id": 54463581,
"node_id": "MDQ6VXNlcjU0NDYzNTgx",
"avatar_url": "https://avatars.githubusercontent.com/u/54463581?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pranavbhat12",
"html_url": "https://github.com/pranavbhat12",
"followers_url": "https://api.github.c... | [
{
"id": 5148671019,
"node_id": "LA_kwDOIf9iDM8AAAABMuJ8Kw",
"url": "https://api.github.com/repos/huggingface/peft/labels/solved",
"name": "solved",
"color": "d4c5f9",
"default": false,
"description": "solved"
}
] | closed | false | null | [] | null | 15 | 2023-11-08T07:14:36 | 2023-12-17T15:03:37 | 2023-12-17T15:03:37 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am trying the load the finetuned adapter on the llama-2-7b model.But getting the following error.This was working fine till yesterday.Tried to downgrade Peft version to 0.5.0 and transformers to 4.32.0.
**Code**:
`finetune_model=PeftModel.from_pretrained(model,"llama-finetuned/llama-7b-hf-4bit-3epochs/",device_ma... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1095/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1095/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1093 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1093/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1093/comments | https://api.github.com/repos/huggingface/peft/issues/1093/events | https://github.com/huggingface/peft/issues/1093 | 1,982,580,432 | I_kwDOIf9iDM52K8bQ | 1,093 | Possible issue with prompt tuning with left padding | {
"login": "SumanthRH",
"id": 39546518,
"node_id": "MDQ6VXNlcjM5NTQ2NTE4",
"avatar_url": "https://avatars.githubusercontent.com/u/39546518?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SumanthRH",
"html_url": "https://github.com/SumanthRH",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 5 | 2023-11-08T02:07:13 | 2024-01-10T15:03:48 | 2024-01-10T15:03:48 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
peft version 0.7.0
### Who can help?
@pacman100 @BenjaminBossan
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [X] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reprod... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1093/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1093/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1090 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1090/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1090/comments | https://api.github.com/repos/huggingface/peft/issues/1090/events | https://github.com/huggingface/peft/issues/1090 | 1,981,525,709 | I_kwDOIf9iDM52G67N | 1,090 | Adapter weights are cast in float16 unexpectedly after load_adapter | {
"login": "hiyouga",
"id": 16256802,
"node_id": "MDQ6VXNlcjE2MjU2ODAy",
"avatar_url": "https://avatars.githubusercontent.com/u/16256802?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hiyouga",
"html_url": "https://github.com/hiyouga",
"followers_url": "https://api.github.com/users/hiyoug... | [] | closed | false | null | [] | null | 1 | 2023-11-07T14:41:22 | 2023-12-13T09:23:36 | 2023-12-13T09:23:36 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Ubuntu + NVIDIA V100
transformers 4.34.1
peft 0.6.0
### Who can help?
@pacman100 @younesbelkada
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [x] My own task or dataset ... | {
"login": "hiyouga",
"id": 16256802,
"node_id": "MDQ6VXNlcjE2MjU2ODAy",
"avatar_url": "https://avatars.githubusercontent.com/u/16256802?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hiyouga",
"html_url": "https://github.com/hiyouga",
"followers_url": "https://api.github.com/users/hiyoug... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1090/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/peft/issues/1090/timeline | null | not_planned | false |
https://api.github.com/repos/huggingface/peft/issues/1089 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1089/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1089/comments | https://api.github.com/repos/huggingface/peft/issues/1089/events | https://github.com/huggingface/peft/issues/1089 | 1,981,168,322 | I_kwDOIf9iDM52FjrC | 1,089 | DOC: List of TODOs for the documentation | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | [
{
"id": 4838806421,
"node_id": "LA_kwDOIf9iDM8AAAABIGpTlQ",
"url": "https://api.github.com/repos/huggingface/peft/labels/documentation",
"name": "documentation",
"color": "0075ca",
"default": true,
"description": "Improvements or additions to documentation"
},
{
"id": 4838806434,... | closed | false | null | [] | null | 8 | 2023-11-07T11:47:25 | 2023-12-21T16:20:52 | 2023-12-21T16:20:51 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
This issue serves to list areas in which the documentation can be improved. It can be extended in the future if we find more gaps.
Contributions are welcome, if you want to pick up one of the open topics, let us know beforehand so that we can avoid duplicate work.
- [x] Conceptual guides:
... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1089/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/peft/issues/1089/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1087 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1087/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1087/comments | https://api.github.com/repos/huggingface/peft/issues/1087/events | https://github.com/huggingface/peft/issues/1087 | 1,980,864,302 | I_kwDOIf9iDM52EZcu | 1,087 | Error when loading Lora weights in example notebook. | {
"login": "kelsey-um",
"id": 103369622,
"node_id": "U_kgDOBilLlg",
"avatar_url": "https://avatars.githubusercontent.com/u/103369622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kelsey-um",
"html_url": "https://github.com/kelsey-um",
"followers_url": "https://api.github.com/users/kelsey... | [] | closed | false | null | [] | null | 6 | 2023-11-07T09:01:40 | 2024-02-14T02:17:04 | 2023-11-07T15:01:12 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Running on Colab using T4 GPU but I've run into this issue aswell on a RTX4090 with other models. Using the example found in _examples/int8_training/Finetune_flan_t5_large_bnb_peft.ipynb_, specifically the cell where the adapters are loaded from the hub.
### Who can help?
_No response_
### Informati... | {
"login": "kelsey-um",
"id": 103369622,
"node_id": "U_kgDOBilLlg",
"avatar_url": "https://avatars.githubusercontent.com/u/103369622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kelsey-um",
"html_url": "https://github.com/kelsey-um",
"followers_url": "https://api.github.com/users/kelsey... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1087/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1087/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1083 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1083/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1083/comments | https://api.github.com/repos/huggingface/peft/issues/1083/events | https://github.com/huggingface/peft/issues/1083 | 1,978,376,063 | I_kwDOIf9iDM51659_ | 1,083 | Use AutoPeftModelForCausalLM to merge and save model cause crashed in the low VRAM situation | {
"login": "ytchen175",
"id": 58975450,
"node_id": "MDQ6VXNlcjU4OTc1NDUw",
"avatar_url": "https://avatars.githubusercontent.com/u/58975450?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ytchen175",
"html_url": "https://github.com/ytchen175",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 9 | 2023-11-06T06:19:36 | 2025-01-11T20:30:55 | 2023-11-08T10:17:49 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Platform
---
* Ubuntu 20 LTS
Python
---
* 3.11.5
Lib
---
* peft=0.5.0
* accelerate=0.21.0
* transformers=4.31.0
### Who can help?
@pacman100 @younesbelkada @sayakpaul
### Information
- [ ] The official example scripts
- [X] My own modified scripts
### Tasks
- [ ] A... | {
"login": "ytchen175",
"id": 58975450,
"node_id": "MDQ6VXNlcjU4OTc1NDUw",
"avatar_url": "https://avatars.githubusercontent.com/u/58975450?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ytchen175",
"html_url": "https://github.com/ytchen175",
"followers_url": "https://api.github.com/users/... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1083/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1083/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1082 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1082/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1082/comments | https://api.github.com/repos/huggingface/peft/issues/1082/events | https://github.com/huggingface/peft/issues/1082 | 1,978,238,004 | I_kwDOIf9iDM516YQ0 | 1,082 | 'Embedding' object has no attribute 'device' for MPT | {
"login": "junzhang-zj",
"id": 52874161,
"node_id": "MDQ6VXNlcjUyODc0MTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/52874161?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/junzhang-zj",
"html_url": "https://github.com/junzhang-zj",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 11 | 2023-11-06T03:59:24 | 2024-02-19T12:54:04 | 2024-02-11T15:03:34 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I followed the test process of MPT notebook, but encountered an error during target training: 'Embedding' object has no attribute 'device'. I want to ask how to solve this. | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1082/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1082/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1081 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1081/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1081/comments | https://api.github.com/repos/huggingface/peft/issues/1081/events | https://github.com/huggingface/peft/issues/1081 | 1,977,502,546 | I_kwDOIf9iDM513ktS | 1,081 | PEFT with quantized LLMa | {
"login": "Sai-Ashish",
"id": 30151156,
"node_id": "MDQ6VXNlcjMwMTUxMTU2",
"avatar_url": "https://avatars.githubusercontent.com/u/30151156?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sai-Ashish",
"html_url": "https://github.com/Sai-Ashish",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 3 | 2023-11-04T20:52:12 | 2023-12-13T15:03:47 | 2023-12-13T15:03:47 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
I would like to inquire if prompt tuning feature is available with 4-bit quantized LLAMA-2 model.
### Motivation
4-bit quantization makes the models easy to use and hardware friendly. Can they be used along with prompt tuning using this library?
### Your contribution
I believe it shou... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1081/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1081/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1080 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1080/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1080/comments | https://api.github.com/repos/huggingface/peft/issues/1080/events | https://github.com/huggingface/peft/issues/1080 | 1,977,265,514 | I_kwDOIf9iDM512q1q | 1,080 | Add docs on how to merge adapters after 4bit QLoRA with PEFT 0.6 | {
"login": "geronimi73",
"id": 141400217,
"node_id": "U_kgDOCG2YmQ",
"avatar_url": "https://avatars.githubusercontent.com/u/141400217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/geronimi73",
"html_url": "https://github.com/geronimi73",
"followers_url": "https://api.github.com/users/ger... | [] | closed | false | null | [] | null | 4 | 2023-11-04T10:07:16 | 2023-11-17T22:22:06 | 2023-11-07T15:46:28 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
there has been some controversy on how to correctly **merge the adapters with the base model after 4bit LoRA** training.
to me it seems there are two ways to merge and save:
- ChrisHayduk https://gist.github.com/ChrisHayduk/1a53463331f52dca205e55982baf9930
- TheBloke https://github.com/Th... | {
"login": "geronimi73",
"id": 141400217,
"node_id": "U_kgDOCG2YmQ",
"avatar_url": "https://avatars.githubusercontent.com/u/141400217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/geronimi73",
"html_url": "https://github.com/geronimi73",
"followers_url": "https://api.github.com/users/ger... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1080/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1080/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1079 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1079/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1079/comments | https://api.github.com/repos/huggingface/peft/issues/1079/events | https://github.com/huggingface/peft/issues/1079 | 1,977,071,746 | I_kwDOIf9iDM5117iC | 1,079 | Token classification LoRA example broken | {
"login": "srama2512",
"id": 31301440,
"node_id": "MDQ6VXNlcjMxMzAxNDQw",
"avatar_url": "https://avatars.githubusercontent.com/u/31301440?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/srama2512",
"html_url": "https://github.com/srama2512",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 4 | 2023-11-04T00:18:57 | 2023-11-09T16:38:57 | 2023-11-09T13:50:37 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Python version: 3.9.18
PEFT version: 0.5.0
Transformers: 4.35.0
Safetensors: 0.4.0
System: Ubuntu 22.04, CUDA 11.8
### Who can help?
_No response_
### Information
- [X] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` ... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1079/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1079/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1073 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1073/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1073/comments | https://api.github.com/repos/huggingface/peft/issues/1073/events | https://github.com/huggingface/peft/issues/1073 | 1,975,386,885 | I_kwDOIf9iDM51vgMF | 1,073 | Device map is not as configured | {
"login": "xinyual",
"id": 74362153,
"node_id": "MDQ6VXNlcjc0MzYyMTUz",
"avatar_url": "https://avatars.githubusercontent.com/u/74362153?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xinyual",
"html_url": "https://github.com/xinyual",
"followers_url": "https://api.github.com/users/xinyua... | [] | closed | false | null | [] | null | 5 | 2023-11-03T03:05:57 | 2023-12-11T15:03:41 | 2023-12-11T15:03:41 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I want to load a model on a fixed GPU. But I find it always cost some memory on GPU:0 by default. For example, if I set device = torch.device("cuda:1"), and load a lora-head Mistral model, it will always cost about 700M in GPU:0.
### Who can help?
_No response_
### Information
- [ ] The official e... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1073/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1073/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1070 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1070/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1070/comments | https://api.github.com/repos/huggingface/peft/issues/1070/events | https://github.com/huggingface/peft/issues/1070 | 1,972,917,950 | I_kwDOIf9iDM51mFa- | 1,070 | modules_to_save not working for AutoModelForSequenceClassification | {
"login": "drei34",
"id": 2657501,
"node_id": "MDQ6VXNlcjI2NTc1MDE=",
"avatar_url": "https://avatars.githubusercontent.com/u/2657501?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/drei34",
"html_url": "https://github.com/drei34",
"followers_url": "https://api.github.com/users/drei34/foll... | [] | closed | false | null | [] | null | 9 | 2023-11-01T18:51:30 | 2024-05-13T18:29:47 | 2023-12-11T15:03:43 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Hi I am using LLAMA2 and GPT2 for sequence classification. Both models add a "score" layer on top to transform the last embedding of the tokens into a vector of class logits. If I specify the layers are below for GPT2 or LLAMA2, I see NaN for the rain and validation accuracy. If I use "classifier", w... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1070/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1070/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1067 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1067/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1067/comments | https://api.github.com/repos/huggingface/peft/issues/1067/events | https://github.com/huggingface/peft/issues/1067 | 1,971,038,362 | I_kwDOIf9iDM51e6ia | 1,067 | Huge performance hit adding in a LoRA | {
"login": "praguepride",
"id": 116130178,
"node_id": "U_kgDOBuwBgg",
"avatar_url": "https://avatars.githubusercontent.com/u/116130178?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/praguepride",
"html_url": "https://github.com/praguepride",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 3 | 2023-10-31T18:07:56 | 2023-12-09T15:03:47 | 2023-12-09T15:03:47 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I am running BLIP-2 on an AWS CPU ECS instance: (16 vGPU, 64 GB of ram). Having BLIP-2 caption images comes out to about 1 image every 3 seconds given the code I am running.
I was able to train a LoRA very quickly however when I run this section of code (taken from the sample google colab):
http... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1067/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1067/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1066 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1066/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1066/comments | https://api.github.com/repos/huggingface/peft/issues/1066/events | https://github.com/huggingface/peft/issues/1066 | 1,968,495,636 | I_kwDOIf9iDM51VNwU | 1,066 | test_prompt_encoder_warning_num_layers test failing | {
"login": "M-Waszkiewicz-Anaconda",
"id": 125905367,
"node_id": "U_kgDOB4Ep1w",
"avatar_url": "https://avatars.githubusercontent.com/u/125905367?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/M-Waszkiewicz-Anaconda",
"html_url": "https://github.com/M-Waszkiewicz-Anaconda",
"followers_url... | [] | closed | false | null | [] | null | 10 | 2023-10-30T14:22:23 | 2023-12-07T15:03:56 | 2023-12-07T15:03:56 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Running package build in conda env on linux64, list of installed dependencies:
The following NEW packages will be INSTALLED:
```
_libgcc_mutex: 0.1-main
_openmp_mutex: 5.1-1_gnu
abseil-cpp: 20211102.0-hd... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1066/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1066/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1065 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1065/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1065/comments | https://api.github.com/repos/huggingface/peft/issues/1065/events | https://github.com/huggingface/peft/issues/1065 | 1,968,291,572 | I_kwDOIf9iDM51Ub70 | 1,065 | PEFT LoHa to kohya-ss LoHa conversion scripts | {
"login": "Xynonners",
"id": 63993670,
"node_id": "MDQ6VXNlcjYzOTkzNjcw",
"avatar_url": "https://avatars.githubusercontent.com/u/63993670?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Xynonners",
"html_url": "https://github.com/Xynonners",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 3 | 2023-10-30T12:54:27 | 2023-12-07T15:03:58 | 2023-12-07T15:03:58 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
To have a conversion script from peft (lycoris) to kohya-ss especially for SDXL.
### Motivation
Currently, it's possible to use peft to centralize training of multiple different adapters (LoRA, LoCon, LoHa, LoKr) for Stable Diffusion, however the peft format is not accepted in some webuis. ... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1065/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1065/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1064 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1064/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1064/comments | https://api.github.com/repos/huggingface/peft/issues/1064/events | https://github.com/huggingface/peft/issues/1064 | 1,967,695,703 | I_kwDOIf9iDM51SKdX | 1,064 | Clarifications on merge_and_unload() as well as load_adapter() and add_adapter(). | {
"login": "yousenwang",
"id": 10716228,
"node_id": "MDQ6VXNlcjEwNzE2MjI4",
"avatar_url": "https://avatars.githubusercontent.com/u/10716228?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yousenwang",
"html_url": "https://github.com/yousenwang",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 7 | 2023-10-30T07:36:11 | 2023-12-18T11:31:38 | 2023-12-18T11:31:37 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I fine-tuned two PeftModel using LoraConfig on two different datasets.
What should I do to achieve the incremental effect?
Case 1:
```py
base_model.load_adapter("./my_first_peft_model_local_path")
## and then
base_model.add_adapter("./my_second_peft_model_local_path")
```
Is case 1 not what I want since ... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1064/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1064/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1062 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1062/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1062/comments | https://api.github.com/repos/huggingface/peft/issues/1062/events | https://github.com/huggingface/peft/issues/1062 | 1,967,020,626 | I_kwDOIf9iDM51PlpS | 1,062 | IA3 import/loading error | {
"login": "LeeChuh",
"id": 60937990,
"node_id": "MDQ6VXNlcjYwOTM3OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/60937990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LeeChuh",
"html_url": "https://github.com/LeeChuh",
"followers_url": "https://api.github.com/users/LeeChu... | [] | closed | false | null | [] | null | 1 | 2023-10-29T14:47:42 | 2023-10-29T14:53:22 | 2023-10-29T14:53:14 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
<img width="1056" alt="image" src="https://github.com/huggingface/peft/assets/60937990/0b7135ec-9395-462f-92b9-88a0dc2eb075">
Hi: I'm using default code from IA3Config, it seems that there's bug in it. And I cannot import ia3Config from Transformers.
### Who can help?
@pacman100 @younesbelkada @sa... | {
"login": "LeeChuh",
"id": 60937990,
"node_id": "MDQ6VXNlcjYwOTM3OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/60937990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LeeChuh",
"html_url": "https://github.com/LeeChuh",
"followers_url": "https://api.github.com/users/LeeChu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1062/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1062/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1060 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1060/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1060/comments | https://api.github.com/repos/huggingface/peft/issues/1060/events | https://github.com/huggingface/peft/issues/1060 | 1,966,591,680 | I_kwDOIf9iDM51N87A | 1,060 | There seems to be an error in LoRA's initialization | {
"login": "ZexiLee",
"id": 49397774,
"node_id": "MDQ6VXNlcjQ5Mzk3Nzc0",
"avatar_url": "https://avatars.githubusercontent.com/u/49397774?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ZexiLee",
"html_url": "https://github.com/ZexiLee",
"followers_url": "https://api.github.com/users/ZexiLe... | [] | closed | false | null | [] | null | 3 | 2023-10-28T13:14:45 | 2023-12-04T09:43:34 | 2023-12-04T09:43:07 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
In lora.py Lines 205-213, the code is
```
def reset_lora_parameters(self, adapter_name):
if adapter_name in self.lora_A.keys():
# initialize A the same way as the default for nn.Linear and B to zero
nn.init.kaiming_uniform_(self.lora_A[adapter_name].weight, a=... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1060/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1060/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1059 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1059/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1059/comments | https://api.github.com/repos/huggingface/peft/issues/1059/events | https://github.com/huggingface/peft/issues/1059 | 1,966,405,395 | I_kwDOIf9iDM51NPcT | 1,059 | For a multi-adapter model save_pretrained saves all adapters, from_pretrained only loads one | {
"login": "denizyuret",
"id": 1822118,
"node_id": "MDQ6VXNlcjE4MjIxMTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/1822118?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/denizyuret",
"html_url": "https://github.com/denizyuret",
"followers_url": "https://api.github.com/users... | [
{
"id": 4838806431,
"node_id": "LA_kwDOIf9iDM8AAAABIGpTnw",
"url": "https://api.github.com/repos/huggingface/peft/labels/enhancement",
"name": "enhancement",
"color": "a2eeef",
"default": true,
"description": "New feature or request"
}
] | closed | false | null | [] | null | 4 | 2023-10-28T04:06:56 | 2024-01-01T07:17:12 | 2023-12-05T15:03:49 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | When save_pretrained is called on a multi-adapter model it saves all the adapters.
When from_pretrained is called on the resulting directory, only the default adapter is loaded.
Is there a recommended way to get back the same multi-adapter model when loading? | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1059/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1059/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1058 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1058/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1058/comments | https://api.github.com/repos/huggingface/peft/issues/1058/events | https://github.com/huggingface/peft/issues/1058 | 1,965,736,204 | I_kwDOIf9iDM51KsEM | 1,058 | Numerical instabilities with LoHa | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | [] | closed | false | null | [] | null | 4 | 2023-10-27T15:17:32 | 2023-11-10T16:45:24 | 2023-11-08T11:09:15 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
Latest PEFT main.
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [ ] My own task or dataset (give details below)
### Reproduction
```pyth... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1058/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1058/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1057 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1057/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1057/comments | https://api.github.com/repos/huggingface/peft/issues/1057/events | https://github.com/huggingface/peft/issues/1057 | 1,965,361,034 | I_kwDOIf9iDM51JQeK | 1,057 | Forward Methods seems weird | {
"login": "KimJaehee0725",
"id": 43404665,
"node_id": "MDQ6VXNlcjQzNDA0NjY1",
"avatar_url": "https://avatars.githubusercontent.com/u/43404665?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/KimJaehee0725",
"html_url": "https://github.com/KimJaehee0725",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 0 | 2023-10-27T11:52:06 | 2023-10-27T11:53:01 | 2023-10-27T11:53:00 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | As you can see below code, forward method in PeftModel class uses self.base_model(which is the original model without any peft modules). In this way, there is no change between base model's forward and PeftModel's forward result.
can i know the reason?
https://github.com/huggingface/peft/blob/207229ad5e1854cd897... | {
"login": "KimJaehee0725",
"id": 43404665,
"node_id": "MDQ6VXNlcjQzNDA0NjY1",
"avatar_url": "https://avatars.githubusercontent.com/u/43404665?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/KimJaehee0725",
"html_url": "https://github.com/KimJaehee0725",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1057/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1057/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1055 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1055/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1055/comments | https://api.github.com/repos/huggingface/peft/issues/1055/events | https://github.com/huggingface/peft/issues/1055 | 1,964,429,933 | I_kwDOIf9iDM51FtJt | 1,055 | Fine tune models in the exl2 format | {
"login": "psych0v0yager",
"id": 105936906,
"node_id": "U_kgDOBlB4Cg",
"avatar_url": "https://avatars.githubusercontent.com/u/105936906?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/psych0v0yager",
"html_url": "https://github.com/psych0v0yager",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | null | 3 | 2023-10-26T22:01:05 | 2024-01-07T15:03:27 | 2024-01-07T15:03:27 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
This request is based on the following colab https://colab.research.google.com/drive/1_TIrmuKOFhuRRiTWN94iLKUFu6ZX4ceb?usp=sharing#scrollTo=RkU58PaRjMXC and following pages in the docs https://huggingface.co/docs/transformers/main_classes/quantization
Specifically huggingface offers a way to f... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1055/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1055/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1052 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1052/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1052/comments | https://api.github.com/repos/huggingface/peft/issues/1052/events | https://github.com/huggingface/peft/issues/1052 | 1,961,337,242 | I_kwDOIf9iDM5056Ga | 1,052 | RuntimeError: "svd_cuda_gesvdj" not implemented for 'BFloat16' | {
"login": "Vectorrent",
"id": 94832312,
"node_id": "U_kgDOBacGuA",
"avatar_url": "https://avatars.githubusercontent.com/u/94832312?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Vectorrent",
"html_url": "https://github.com/Vectorrent",
"followers_url": "https://api.github.com/users/Vecto... | [] | closed | false | null | [] | null | 2 | 2023-10-25T12:43:12 | 2023-12-02T15:03:44 | 2023-12-02T15:03:44 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
- `transformers` version: 4.34.0
- Platform: Linux-6.5.6-arch2-1-x86_64-with-glibc2.35
- Python version: 3.10.12
- Huggingface_hub version: 0.17.3
- Safetensors version: 0.4.0
- Accelerate version: 0.23.0
- Accelerate config: not found
- PyTorch version (GPU?): 2.1.0+cu121 (True)
- Tensorflo... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1052/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1052/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1048 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1048/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1048/comments | https://api.github.com/repos/huggingface/peft/issues/1048/events | https://github.com/huggingface/peft/issues/1048 | 1,960,445,936 | I_kwDOIf9iDM502gfw | 1,048 | Question about GPU Memory Usage with LoRA in VisionEncoderDecoderModel | {
"login": "eclickECNU",
"id": 37464335,
"node_id": "MDQ6VXNlcjM3NDY0MzM1",
"avatar_url": "https://avatars.githubusercontent.com/u/37464335?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/eclickECNU",
"html_url": "https://github.com/eclickECNU",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | null | 4 | 2023-10-25T03:12:47 | 2023-12-02T15:03:45 | 2023-12-02T15:03:45 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
- transformers 4.34.0
- torch 2.0.1
- peft 0.5.0
### Who can help?
@BenjaminBossan
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples` folder
- [X] My... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1048/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1048/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1045 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1045/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1045/comments | https://api.github.com/repos/huggingface/peft/issues/1045/events | https://github.com/huggingface/peft/issues/1045 | 1,956,066,795 | I_kwDOIf9iDM50lzXr | 1,045 | add_weighted_adapter() is unusable, throws error: "Invalid type <class 'list'> found in target_modules" | {
"login": "Vectorrent",
"id": 94832312,
"node_id": "U_kgDOBacGuA",
"avatar_url": "https://avatars.githubusercontent.com/u/94832312?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Vectorrent",
"html_url": "https://github.com/Vectorrent",
"followers_url": "https://api.github.com/users/Vecto... | [] | closed | false | null | [] | null | 6 | 2023-10-22T21:42:32 | 2023-10-25T10:44:25 | 2023-10-24T10:06:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
- `transformers` version: 4.34.0
- Platform: Linux-6.5.6-arch2-1-x86_64-with-glibc2.35
- Python version: 3.10.12
- Huggingface_hub version: 0.17.3
- Safetensors version: 0.4.0
- Accelerate version: 0.23.0
- Accelerate config: not found
- PyTorch version (GPU?): 2.1.0+cu121 (True)
- Tensorflo... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1045/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1045/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1044 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1044/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1044/comments | https://api.github.com/repos/huggingface/peft/issues/1044/events | https://github.com/huggingface/peft/issues/1044 | 1,956,047,529 | I_kwDOIf9iDM50luqp | 1,044 | Peft is not working with stabilityai/stablelm-3b-4e1t | {
"login": "apcameron",
"id": 37645737,
"node_id": "MDQ6VXNlcjM3NjQ1NzM3",
"avatar_url": "https://avatars.githubusercontent.com/u/37645737?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/apcameron",
"html_url": "https://github.com/apcameron",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | null | 3 | 2023-10-22T20:37:42 | 2023-10-23T12:30:25 | 2023-10-23T12:30:25 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
pip install git+https://github.com/huggingface/peft.git
Finethining fails with
-------
<pre> File "/home/andrew/Downloads/finetune/env/lib/python3.11/site-packages/trl/trainer/sft_trainer.py", line 143, in __init__
model = get_peft_model(model, peft_config)
^^^^^^^^... | {
"login": "BenjaminBossan",
"id": 6229650,
"node_id": "MDQ6VXNlcjYyMjk2NTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6229650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BenjaminBossan",
"html_url": "https://github.com/BenjaminBossan",
"followers_url": "https://api.gith... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1044/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1044/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1043 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1043/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1043/comments | https://api.github.com/repos/huggingface/peft/issues/1043/events | https://github.com/huggingface/peft/issues/1043 | 1,955,970,786 | I_kwDOIf9iDM50lb7i | 1,043 | Merge and unload changes inference a lot on quantized llama 2 | {
"login": "Lanzelot-Moll",
"id": 76201970,
"node_id": "MDQ6VXNlcjc2MjAxOTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/76201970?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Lanzelot-Moll",
"html_url": "https://github.com/Lanzelot-Moll",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | null | 9 | 2023-10-22T16:45:27 | 2023-11-15T17:30:23 | 2023-11-15T17:30:22 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
I fine-tuned a quantized LLama 2 7b chat model using peft and it worked great. The only problem is the inference is reaaaally slow.
So I read the docs and found the merge_and_unload. Once I apply this the inference doubles in speed, but the results worsen a lot using the same parameters for genera... | {
"login": "Lanzelot-Moll",
"id": 76201970,
"node_id": "MDQ6VXNlcjc2MjAxOTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/76201970?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Lanzelot-Moll",
"html_url": "https://github.com/Lanzelot-Moll",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1043/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1043/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1042 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1042/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1042/comments | https://api.github.com/repos/huggingface/peft/issues/1042/events | https://github.com/huggingface/peft/issues/1042 | 1,955,338,957 | I_kwDOIf9iDM50jBrN | 1,042 | Load PeftModel in another PretrainedModel | {
"login": "sqzhang-lazy",
"id": 68679419,
"node_id": "MDQ6VXNlcjY4Njc5NDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/68679419?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sqzhang-lazy",
"html_url": "https://github.com/sqzhang-lazy",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | null | 10 | 2023-10-21T05:51:22 | 2023-12-18T11:35:17 | 2023-12-18T11:35:16 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### System Info
torch 2.0.1
transformers 4.34.0
peft 0.5.0
### Who can help?
_No response_
### Information
- [ ] The official example scripts
- [ ] My own modified scripts
### Tasks
- [ ] An officially supported task in the `examples... | {
"login": "younesbelkada",
"id": 49240599,
"node_id": "MDQ6VXNlcjQ5MjQwNTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/younesbelkada",
"html_url": "https://github.com/younesbelkada",
"followers_url": "https://api.githu... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1042/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1042/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1040 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1040/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1040/comments | https://api.github.com/repos/huggingface/peft/issues/1040/events | https://github.com/huggingface/peft/issues/1040 | 1,953,950,028 | I_kwDOIf9iDM50dulM | 1,040 | Multiple Adapters merging question | {
"login": "AlbertoZerbinati",
"id": 44242851,
"node_id": "MDQ6VXNlcjQ0MjQyODUx",
"avatar_url": "https://avatars.githubusercontent.com/u/44242851?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AlbertoZerbinati",
"html_url": "https://github.com/AlbertoZerbinati",
"followers_url": "https://... | [] | closed | false | null | [] | null | 5 | 2023-10-20T10:02:31 | 2024-01-19T01:26:33 | 2023-11-20T15:48:00 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
I'm experimenting with merging multiple different Lora adapters into a single model.
This is the code I'm using:
```python
# load the model
model = AutoModelForCausalLM.from_pretrained(
"model_path",
device_map="auto",
torch_dtype=torch.float16,
)
# apply and merge adapter ... | {
"login": "AlbertoZerbinati",
"id": 44242851,
"node_id": "MDQ6VXNlcjQ0MjQyODUx",
"avatar_url": "https://avatars.githubusercontent.com/u/44242851?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AlbertoZerbinati",
"html_url": "https://github.com/AlbertoZerbinati",
"followers_url": "https://... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1040/reactions",
"total_count": 5,
"+1": 5,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1040/timeline | null | completed | false |
https://api.github.com/repos/huggingface/peft/issues/1038 | https://api.github.com/repos/huggingface/peft | https://api.github.com/repos/huggingface/peft/issues/1038/labels{/name} | https://api.github.com/repos/huggingface/peft/issues/1038/comments | https://api.github.com/repos/huggingface/peft/issues/1038/events | https://github.com/huggingface/peft/issues/1038 | 1,953,604,233 | I_kwDOIf9iDM50caKJ | 1,038 | Add VeRA (Vector Based Random Matrix Adaption) | {
"login": "vvvm23",
"id": 44398246,
"node_id": "MDQ6VXNlcjQ0Mzk4MjQ2",
"avatar_url": "https://avatars.githubusercontent.com/u/44398246?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vvvm23",
"html_url": "https://github.com/vvvm23",
"followers_url": "https://api.github.com/users/vvvm23/fo... | [
{
"id": 5192583244,
"node_id": "LA_kwDOIf9iDM8AAAABNYCITA",
"url": "https://api.github.com/repos/huggingface/peft/labels/PRs%20welcome%20to%20address%20this",
"name": "PRs welcome to address this",
"color": "0052cc",
"default": false,
"description": "contributions are welcome from commun... | closed | false | null | [] | null | 4 | 2023-10-20T06:12:14 | 2024-02-01T15:04:21 | 2024-02-01T15:04:21 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | ### Feature request
A recent paper by Qualcomm AI Research proposed a new parameter efficient finetuning method called VeRA. This uses two projection matrices (as in LoRA) that are randomly initialised, frozen, and shared between all layers. Two scaling vectors are created and are trained. This reduces the number of t... | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url"... | {
"url": "https://api.github.com/repos/huggingface/peft/issues/1038/reactions",
"total_count": 5,
"+1": 5,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/peft/issues/1038/timeline | null | completed | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.