Linaqruf/kohya-trainer issues and pull requests

#106 - Errors in 4.4, Finetuning notebook does not run

Issue - State: closed - Opened by Taikakim over 1 year ago - 4 comments

#104 - issue with fast kohya trainer colab

Issue - State: closed - Opened by x-legion over 1 year ago - 1 comment

#103 - "Syntax Error" while running "Start Training"

Issue - State: closed - Opened by HyzQuadriceps over 1 year ago

#102 - Load StableDiffusion checkpoint error

Issue - State: closed - Opened by cerarslan over 1 year ago - 1 comment

#101 - what's wrong?

Issue - State: closed - Opened by foxya27 over 1 year ago - 1 comment

#100 - Is there a paperspace port?

Issue - State: open - Opened by ca3gamedev over 1 year ago - 1 comment

#99 - What I'm doing wrong?

Issue - State: closed - Opened by Gvaroo over 1 year ago - 2 comments

#98 - Update fast-kohya-trainer.ipynb

Pull Request - State: closed - Opened by freecoderai01 over 1 year ago

#97 - Update fast-kohya-trainer.ipynb

Pull Request - State: closed - Opened by freecoderai01 over 1 year ago

#96 - Kohya LoRA Dreambooth error: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 14.00 MiB

Issue - State: closed - Opened by YousufSSyed over 1 year ago - 6 comments

#95 - Kohya Trainer V13

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#94 - Fresh install errors

Issue - State: closed - Opened by DrGunnarMallon over 1 year ago - 2 comments

#93 - error when training even tho it installed everything correctly

Issue - State: closed - Opened by 2blackbar over 1 year ago - 2 comments

#92 - Textual inversion

Issue - State: closed - Opened by Rudy34160 over 1 year ago - 2 comments

#91 - Could this work without an Nvidia GPU? Would it work on my M1 Mac?

Issue - State: closed - Opened by YousufSSyed over 1 year ago - 2 comments

#90 - add support for lion optimizer

Issue - State: closed - Opened by x-legion over 1 year ago - 4 comments

#89 - [Request/question] Extract Lora from Checkpoint

Issue - State: open - Opened by Omenizer over 1 year ago - 2 comments

#88 - Cannot run the inference cell

Issue - State: closed - Opened by IcelandicIcecream over 1 year ago - 1 comment

#87 - Problem when running training cell

Issue - State: closed - Opened by JJJ-art111 over 1 year ago - 2 comments

#85 - FATAL: this function is for sm80, but was built for sm750

Issue - State: closed - Opened by MushroomFleet over 1 year ago - 2 comments

#84 - Diffusion with offset noise

Pull Request - State: closed - Opened by younyokel over 1 year ago

#83 - No such file or directory

Issue - State: closed - Opened by kolabearafk over 1 year ago - 4 comments

#82 - EPoch number does not make sense

Issue - State: closed - Opened by remybonnav over 1 year ago - 2 comments

#81 - learning rate

Issue - State: closed - Opened by remybonnav over 1 year ago - 4 comments

#80 - Error running the Finetuner Lora model in WebUI

Issue - State: closed - Opened by younyokel over 1 year ago - 3 comments

#79 - I didn't get what is the 'Native Training', is that the way same with ldm?

Issue - State: closed - Opened by Alfred-Duncan over 1 year ago - 1 comment

#78 - Huge Changes Incoming

Issue - State: closed - Opened by Linaqruf over 1 year ago - 20 comments
Labels: documentation, enhancement

#77 - [Request] Add Safebooru to booruscraper

Issue - State: closed - Opened by AnANTKA over 1 year ago - 1 comment

#76 - Could this trick be implemented ?

Issue - State: closed - Opened by Njbx over 1 year ago - 10 comments

#75 - Finetuning sd-v2-depth models?

Issue - State: closed - Opened by 0x1355 over 1 year ago - 2 comments

#74 - Caption using Coca instead of BLIP

Issue - State: open - Opened by DanielShemesh over 1 year ago - 2 comments
Labels: enhancement

#73 - [Feature] Add support for the script to extract a custom Dreambooth model into a LoRA embedding

Issue - State: open - Opened by flesler over 1 year ago - 1 comment
Labels: enhancement

#72 - No such file or directory even though the file exists

Issue - State: closed - Opened by AnANTKA over 1 year ago - 2 comments
Labels: help wanted

#71 - Auto captioning for images

Issue - State: closed - Opened by Gyro0o over 1 year ago - 4 comments
Labels: enhancement

#70 - Error Lora Dreambooth

Issue - State: closed - Opened by cdefghijkl over 1 year ago - 4 comments

#69 - Mixup between `remove_words` and `undesired_words` during GIT Captioning

Issue - State: closed - Opened by MasterScrat over 1 year ago - 1 comment
Labels: bug

#68 - Assertion Error when trying to train (Kohya trainer not lora)

Issue - State: closed - Opened by AnANTKA over 1 year ago - 1 comment

#67 - Why no more BLIP?

Issue - State: closed - Opened by MasterScrat over 1 year ago - 9 comments
Labels: good first issue

#66 - Runtime error: size mismatch?

Issue - State: closed - Opened by mhuynh654 over 1 year ago - 3 comments

#65 - feature req: add separate weight for unet and text encoder in the inference cell.

Issue - State: closed - Opened by x-legion over 1 year ago - 1 comment
Labels: wontfix

#63 - Image does not have a caption even though I used waifu diffusion auto tagger

Issue - State: closed - Opened by AnANTKA over 1 year ago - 4 comments

#62 - Xformers for A100

Issue - State: closed - Opened by Wawaworker over 1 year ago - 2 comments

#61 - please explain native vs dreambooth

Issue - State: closed - Opened by maxixa over 1 year ago - 1 comment

#59 - Question: What is the difference between lora dreambooth vs lora fine-tuning?

Issue - State: closed - Opened by YukiSakuma over 1 year ago - 7 comments

#58 - To do: Delete image upscaler cell

Issue - State: closed - Opened by Linaqruf over 1 year ago

#57 - How to resume LORA training ?

Issue - State: closed - Opened by gitaiuser20 over 1 year ago - 19 comments
Labels: good first issue

#56 - update kohya trainer v12

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#55 - Running into an error on my first attempted use

Issue - State: closed - Opened by Wawaworker over 1 year ago - 4 comments

#54 - Why is there no "--keep_tokens" field?

Issue - State: closed - Opened by noriwaru over 1 year ago - 4 comments

#53 - Kohya colab not working properly?

Issue - State: closed - Opened by Atoli over 1 year ago - 8 comments

#52 - Need help! learning_rate

Issue - State: closed - Opened by soyusora over 1 year ago - 3 comments

#49 - Why has the "--learning_rate=" setting disappeared from Kohya LoRA Dreambooth?

Issue - State: closed - Opened by noriwaru over 1 year ago - 3 comments

#48 - Custom Tag not working

Issue - State: closed - Opened by treksis over 1 year ago - 2 comments

#47 - Create JSON file error: list index out of range

Issue - State: closed - Opened by brandnewfakeuser over 1 year ago - 2 comments

#46 - I trained a lora embeding with latest note book,but can't load it in auto1111 additonal network extension.

Issue - State: closed - Opened by eleven-monkey over 1 year ago - 1 comment

#45 - What I'm doing wrong?

Issue - State: closed - Opened by grlfo over 1 year ago - 5 comments

#44 - integrating colab notebook with sd-scripts 0.4.0

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#43 - Colab notebook errors, and warnings and Tensorcore issues.

Issue - State: closed - Opened by DarkAlchy over 1 year ago - 9 comments

#42 - Where is the small LoRA model at not this 2.43GB thing?

Issue - State: closed - Opened by DarkAlchy over 1 year ago - 6 comments

#41 - Error 403 Forbidden when trying to connect to Colab's server

Issue - State: closed - Opened by DarkSolus over 1 year ago - 2 comments

#40 - feat: merging latest code update from kohya

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#39 - (request) adding runtime.unassign() at the end

Issue - State: closed - Opened by treksis over 1 year ago - 1 comment

#38 - Use lr_scheduler

Pull Request - State: closed - Opened by uservar over 1 year ago - 1 comment

#37 - Tensorcore errors and it will not run

Issue - State: closed - Opened by DarkAlchy over 1 year ago - 5 comments

#36 - Is this just for anime?

Issue - State: closed - Opened by DarkAlchy over 1 year ago - 2 comments

#35 - Error when start training

Issue - State: closed - Opened by suryacandra over 1 year ago - 4 comments

#34 - FATAL: this function is for sm80, but was built for sm750

Issue - State: closed - Opened by treksis over 1 year ago - 1 comment

#33 - RuntimeError: CUDA error: initialization error

Issue - State: closed - Opened by etherealxx over 1 year ago - 9 comments

#32 - _pickle.UnpicklingError: invalid load key, '\x09'

Issue - State: closed - Opened by TsGrolken over 1 year ago - 1 comment

#31 - Missing dependencies, cannot start training

Issue - State: closed - Opened by hornedpariah over 1 year ago - 5 comments

#30 - Handle tags from booru scrapper

Pull Request - State: closed - Opened by uservar over 1 year ago

#29 - Please help on output parameters and resuming process

Issue - State: closed - Opened by ErwinTATP over 1 year ago - 2 comments

#28 - merge pull the latest kohya-ss/sd-script version

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#27 - Don't work in Colab since yesterday

Issue - State: closed - Opened by Desm0nt over 1 year ago - 4 comments

#26 - [Question] multiple gpu

Issue - State: closed - Opened by dill-shower over 1 year ago

#25 - Add more arguments for Cosine_with_restarts scheduler

Issue - State: closed - Opened by Azurlifrfa over 1 year ago - 2 comments

#24 - Fatal Error using colab when starting training: google colab

Issue - State: closed - Opened by devantang over 1 year ago - 1 comment

#23 - Vast.ai torch 1.13.1+cu116 issue

Issue - State: closed - Opened by Desm0nt over 1 year ago - 1 comment

#22 - How to Resume Training

Issue - State: closed - Opened by KitsunekoFi over 1 year ago - 1 comment

#21 - Batch training dreambooth ?

Issue - State: closed - Opened by Michaelwhite34 over 1 year ago - 2 comments

#20 - LoRA file does not work with A1111 Additional Networks

Issue - State: closed - Opened by MitPitt over 1 year ago - 2 comments

#19 - fix no image with `[` in filename

Pull Request - State: closed - Opened by Kidel over 1 year ago - 1 comment

#18 - problem with training wd 1.4

Issue - State: closed - Opened by Michaelwhite34 over 1 year ago - 3 comments

#17 - Bug: Append Custom Tag not Working

Issue - State: closed - Opened by NekoFi over 1 year ago - 1 comment

#16 - Binning not working

Issue - State: closed - Opened by minermanb over 1 year ago - 2 comments

#15 - Train Model has an error

Issue - State: closed - Opened by Lookatme5555 over 1 year ago - 4 comments

#14 - add gen_img_diffusers to dreambooth

Pull Request - State: closed - Opened by Linaqruf over 1 year ago

#13 - model.ckpt file not found

Issue - State: closed - Opened by DanielShemesh over 1 year ago - 4 comments

#12 - runpod

Pull Request - State: closed - Opened by vibrantrida over 1 year ago

#11 - push v9

Pull Request - State: closed - Opened by Linaqruf almost 2 years ago

#10 - about v2 training

Issue - State: closed - Opened by cameback0123 almost 2 years ago - 4 comments

#9 - Publish Kohya Trainer V8

Pull Request - State: closed - Opened by Linaqruf almost 2 years ago

#8 - About dreambooth reg image

Issue - State: closed - Opened by mio7690 almost 2 years ago - 1 comment

#7 - merge experimental branch with main

Pull Request - State: closed - Opened by Linaqruf almost 2 years ago

#6 - Will this adapt to local server(conda is best),not just colab?

Issue - State: closed - Opened by mio7690 almost 2 years ago - 2 comments

#5 - Run on Runpod / Vast.ai

Issue - State: closed - Opened by andykaufseo almost 2 years ago - 2 comments

#4 - i have a questions for you

Issue - State: closed - Opened by Juan-Cruz-Iturrioz almost 2 years ago - 2 comments

GitHub / Linaqruf/kohya-trainer issues and pull requests