Repository Issue Activity (beta)

#2420 `torch.cuda.amp` in `__main__` self-test blocks and the IPEX hijack — deprecated since torch 2.3/2.4, unguarded on the main install path

#2417 [Feature Request] Add option to disable cuDNN/MIOpen backend (workaround for AMD RDNA 4 GPU crashes)

#2414 WeightsUnpickler error: Unsupported operand 56

#2405 --tag_replacement doesn't actually replace anything (variable shadowing bug)

#1298 Tutorial for train_controlnet

#2407 [bug] --skip_until_initial_step 使用時、--gradient_accumulation_steps > 1 で再開位置を過剰にスキップして学習が終了する

#2403 Silent `adaln` drop while anima lora training

#2402 LECO: verify_training_args() crashes on missing sample_every_n_epochs/sample_every_n_steps

#2363 [QUESTION] - Anima LoRA training params

#2394 Adding Anima ControlNet support

#2388 [bug] Caption generation fails during prediction

#2392 "Missing keys in state dict" solved when training lora on some anima ft checkpoints

#1304 caption joined by ", " instead of caption_separator when shuffling

#2295 blocks_to_swap parameter is not available on non cuda devices

#2306 Mixed-precision is not available in torch-xpu 2.10

#2383 [Request] add display "discrete_flow_shift", "timestep_sampling"

#2362 Error : cache_latents_to_disk for anima_train (キャッシュ読込エラー)

#2366 sd scriptsのoption console_log_simpleの適用範囲について

#1916 Can not generate full body with face lora model.

#2365 カスタムスケジューラを登録したい／I want to register a custom scheduler

#2364 bf16/float RuntimeError - dtype incompatibility

#2308 Does Anima model support DoRA training?

#2326 npzファイルを画像として開こうとする

#2323 Tag info saved in model metadata ignores keep_tokens_separator

#2266 Significant bottleneck with text encoder outputs and latent caching to disk with .npz vs .safetensors