Data

Tools

Indexes

Applications

Experiments

Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / kohya-ss/sd-scripts issues and pull requests

#2153 - Fix alphas cumprod after add_noise for DDIMScheduler

Pull Request - State: open - Opened by rockerBOO 16 days ago

#2088 - Update workflows to read-all instead of write-all

Pull Request - State: closed - Opened by rockerBOO 2 months ago - 1 comment

#2087 - Wrong training steps when images are in jxl format

Issue - State: closed - Opened by Hirmuolio 3 months ago - 2 comments

#2086 - Add English docs with Japanese details

Pull Request - State: closed - Opened by kohya-ss 3 months ago
Labels: codex

#2084 - Learning rate not adjusting with adaptive optimizer

Issue - State: open - Opened by VoltLoop 3 months ago - 5 comments

#2083 - [REQUEST] Support masked loss training in loras

Issue - State: closed - Opened by Tophness 3 months ago

#2082 - FLUX training all blocks

Issue - State: open - Opened by PluginBOXone 3 months ago - 1 comment

#2081 - Chroma Support ?

Issue - State: open - Opened by wzgrx 3 months ago - 2 comments

#2080 - Add Partitioned VAE

Pull Request - State: open - Opened by rockerBOO 3 months ago - 1 comment

#2078 - RuntimeError: "eye" not implemented for 'BFloat16'

Issue - State: open - Opened by FMsunyh 3 months ago - 4 comments

#2077 - How does validation loss interact with repeat?

Issue - State: open - Opened by ToothPaste73 3 months ago

#2076 - Kohya not seeing training images

Issue - State: closed - Opened by WTW84 3 months ago - 2 comments

#2074 - Deepspeed readme

Pull Request - State: closed - Opened by kohya-ss 3 months ago

#2073 - Fix mean grad norms

Pull Request - State: closed - Opened by rockerBOO 3 months ago - 1 comment

#2072 - Add pythonpath to pytest.ini

Pull Request - State: closed - Opened by rockerBOO 3 months ago - 2 comments

#2071 - GitHub Actions pytest failed

Issue - State: closed - Opened by kohya-ss 3 months ago - 1 comment

#2069 - Implement Ai Toolkits Version of DOP

Issue - State: closed - Opened by RougeXAi 3 months ago - 1 comment

#2067 - Wandb not logging steps if no `--logging_dir` set

Issue - State: open - Opened by SharkWipf 3 months ago - 3 comments

#2066 - Quick fix flux sampling scales

Pull Request - State: closed - Opened by kohya-ss 3 months ago

#2065 - Create FUNDING.yml

Pull Request - State: closed - Opened by kohya-ss 3 months ago

#2064 - Add CFG for sampling in training with FLUX.1

Pull Request - State: closed - Opened by kohya-ss 3 months ago

#2056 - Added --blocks_to_swap_while_sampling, may allow faster sample image generation

Pull Request - State: open - Opened by araleza 3 months ago - 3 comments

#2054 - WD14 tagging model download path issue

Issue - State: closed - Opened by DavidJaeck 3 months ago - 3 comments

#2047 - Add AID Activation by Interval-wise Dropout

Pull Request - State: open - Opened by rockerBOO 4 months ago - 5 comments

#2045 - Add Flash, cuDNN, Efficient attention for Flux

Pull Request - State: open - Opened by rockerBOO 4 months ago - 3 comments

#2042 - experimental: Documentation Updates with Gemini

Pull Request - State: open - Opened by kohya-ss 4 months ago - 3 comments

#2040 - Training support for HiDream-I1?

Issue - State: open - Opened by nitinh12 4 months ago - 14 comments

#2038 - TOML questions - validation loss and DoRA

Issue - State: open - Opened by Kahldris 4 months ago - 2 comments

#2037 - Add wavelet loss for networks

Pull Request - State: open - Opened by rockerBOO 4 months ago - 5 comments

#2036 - Progress2

Pull Request - State: closed - Opened by imon1-pocketfm 4 months ago

#2035 - progress interceptor

Pull Request - State: closed - Opened by imon1-pocketfm 4 months ago - 1 comment

#2034 - flux inference

Issue - State: open - Opened by aminaxashfaq 4 months ago - 1 comment

#2032 - Add parse blocks to allow setting values per block

Pull Request - State: open - Opened by rockerBOO 4 months ago

#2031 - FLUX ControlNet inference script

Issue - State: open - Opened by mfclabber 4 months ago - 1 comment

#2029 - `--console_log_file` significantly increases VRAM usage 🤔

Issue - State: closed - Opened by SharkWipf 4 months ago - 15 comments

#2028 - Fix resize PR link

Pull Request - State: closed - Opened by rockerBOO 4 months ago - 1 comment

#2025 - failed to train controlnet with flux.1-dev model

Issue - State: closed - Opened by Johnson-yue 4 months ago - 2 comments

#2024 - how to build dataset.toml for training controlnet with flux-dev?

Issue - State: open - Opened by Johnson-yue 4 months ago - 1 comment

#2023 - TypeError: FineTuningDataset

Issue - State: closed - Opened by rena0408 4 months ago - 3 comments

#2022 - Fix size parameter types and improve resize_image interpolation

Pull Request - State: closed - Opened by LexSong 4 months ago - 4 comments

#2021 - Slow iteration/seconds on flux dev lora training

Issue - State: open - Opened by asinghan2603 4 months ago - 5 comments

#2019 - Add gradient noise scale logging

Pull Request - State: open - Opened by rockerBOO 4 months ago - 1 comment

#2018 - Resize interpolation small fix

Pull Request - State: closed - Opened by kohya-ss 4 months ago - 2 comments

#2016 - Wavelet loss function?

Issue - State: open - Opened by nitinh12 4 months ago - 2 comments

#2015 - Using --vae_batch_size to set batch size for dynamic latent generation

Pull Request - State: closed - Opened by DKnight54 4 months ago - 1 comment

#2014 - Exception in ASGI application

Issue - State: open - Opened by llm8047 4 months ago - 1 comment

#2012 - How to use custom learning scheduler?

Issue - State: open - Opened by MehmetcanTozlu 4 months ago - 4 comments

#2011 - Custom Optimizer

Issue - State: open - Opened by FidanVural 4 months ago - 9 comments

#2009 - EMA for Flux Dev

Issue - State: open - Opened by CelesteHargreeves 4 months ago

#2008 - Update sdxl_original_unet.py for apple silicon

Pull Request - State: open - Opened by cf 4 months ago

#2007 - Support more checkpoint files for flux

Pull Request - State: open - Opened by rockerBOO 4 months ago - 1 comment

#2005 - full dreambooth for cogview4？

Issue - State: open - Opened by godwenbin 4 months ago

#2004 - Could you merge flux-sample-cfg to sd3 branch?

Issue - State: closed - Opened by Ice-YY 4 months ago - 1 comment

#2003 - Fix missing parameters in model conversion script

Pull Request - State: open - Opened by laolongboy 4 months ago

#2002 - Add URAE Ultra-Resolution Adaptation with Ease support

Pull Request - State: open - Opened by rockerBOO 4 months ago

#2001 - Add initialization URAE, PiSSA for flux

Pull Request - State: open - Opened by rockerBOO 4 months ago - 8 comments

#2000 - Add flexibility to block swapping for Flex model

Pull Request - State: open - Opened by rockerBOO 4 months ago

#1999 - resume step calculation is flawed

Issue - State: open - Opened by Deathawaits4 4 months ago - 3 comments

#1998 - Adaptive LR for each block (SDXL)

Issue - State: open - Opened by iqddd 4 months ago

#1997 - full dreambooth for flux?

Issue - State: open - Opened by bryanhpchiang 4 months ago - 1 comment

#1996 - Error when running convert_diffusers_to_flux

Issue - State: open - Opened by laolongboy 4 months ago - 2 comments

#1993 - blurred background

Issue - State: closed - Opened by debedebe 4 months ago

#1992 - Add IP noise gamma for Flux

Pull Request - State: closed - Opened by rockerBOO 5 months ago - 14 comments

#1991 - Support alpha cumulative product using shifted sigmas for Flux

Pull Request - State: open - Opened by rockerBOO 5 months ago - 1 comment

#1990 - Explaination of --model_prediction_type

Issue - State: open - Opened by sipie800 5 months ago

#1988 - block swap with no_grad context

Pull Request - State: open - Opened by kohya-ss 5 months ago - 1 comment

#1987 - Support for new T5 tokenizer

Issue - State: open - Opened by Kaoru8 5 months ago - 9 comments

#1985 - Support pytorch_optimizer

Pull Request - State: open - Opened by gesen2egee 5 months ago - 1 comment

#1983 - illustriousXL 1.0 train -> black images

Issue - State: closed - Opened by kalle07 5 months ago - 2 comments

#1982 - can't run flux_minimal_inference.py

Issue - State: open - Opened by bryanhpchiang 5 months ago

#1977 - Why does the loss fluctuate sharply when repeats=1?

Issue - State: open - Opened by gesen2egee 5 months ago - 1 comment

#1976 - Error in swap_weight_devices_no_cuda When Running on Intel CPU

Issue - State: open - Opened by gal-dahan 5 months ago - 1 comment

#1974 - Add LoRA-GGPO for Flux

Pull Request - State: closed - Opened by rockerBOO 5 months ago - 19 comments

#1972 - 解決済み

Issue - State: open - Opened by ayukawa05 5 months ago - 2 comments

#1970 - train_single_block_indices Parameter Appears Ineffective

Issue - State: open - Opened by Brackarnwar 5 months ago - 1 comment

#1968 - Why are model names hardcoded in analyze_checkpoint_state?

Issue - State: open - Opened by CaledoniaProject 5 months ago - 2 comments