KYU

Pre-training model dataset questions in SO-VITS-SVC 5.0

1

I want to know the dataset used for pre-training models in SO-VITS-SVC 5.0. Did you use the dataset specified in the repository?

During training, an issue was discovered where f0_loss was constantly zero. Any help would be appreciated.

I'm wondering if the code in that GitHub repository also performs well at cloning songs. Thanks.

Can you provide the training code for that model? ___ import torch from naturalspeech2_pytorch import Trainer, EncodecWrapper, Model, NaturalSpeech2, SpeechPromptEncoder codec = EncodecWrapper() def main(): model = Model( dim =...

Can you also share the inference code? Thank you.

Please let me know the format of the data to train the model and a detailed guide to train the model. Thank you.

Trained Lora how-to questions

1

https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/train/stable_diffusion_xl/train_sdxl_lora.py ___ hello. After training with the code here, can I use the Lora as output in any code from that repository? I'm a coding beginner. Please tell me how...

stable diffusion sdxl train error

2

(test) C:\DiffSynth-Studio\examples\train\stable_diffusion_xl>python train_sdxl_lora.py --pretrained_path models/stable_diffusion_xl/autismmixSDXL_autismmixConfetti.safetensors --dataset_path C:\DiffSynth-Studio\data\test --output_path ./models --max_epochs 1 --steps_per_epoch 500 --height 1024 --width 1024 --center_crop --precision "32" --learning_rate 1e-4 --lora_rank 4 --lora_alpha 4 --use_gradient_checkpointing 2024-11-01 14:08:11.631651: I...

I was wondering how the datasets used in that github repository are formatted and should be placed in a folder.