|
Download README.md from someonecantcode/FlowMatching: direct link, hf CLI and curl.
- Browser
- Download file 1.38 kB
-
https://huggingface.co/someonecantcode/FlowMatching/resolve/main/README.md
- Command line
-
hf download hf://someonecantcode/FlowMatching/README.md
-
curl -L -o README.md https://huggingface.co/someonecantcode/FlowMatching/resolve/main/README.md
1.38 kB
metadata
license: gpl-3.0
datasets:
- stanford-vision-lab/gpic
language:
- en
pipeline_tag: image-text-to-image
tags:
- VAE
- mmDiT
Weights for mmDiT and VAE from Diffusion Transformer.
Weights structure:
flow_match_checkpoint = {
'model': raw_model.state_dict(),
'optimizer': optimizer.state_dict(),
'ema': ema.state_dict,
'step': step
}
vae_checkpoint = vae.state_dict(),
Hyperparameters used for model.
@dataclass
class TrainConfig:
lr: float = 1e-4
batch_size: int = 8
total_steps: int = 10000 # 50k steps is ideal.
lr_cooldown_steps: int = 20000
starting_step: int = 0
logging_interval: int = 100
grad_accum_steps: int = 2 # not really needed
save_dir: str = "./saved_models/"
ckpt: str = "flowmatch-ckpt.pt"
@dataclass
class VAEConfig:
kl_beta: float = 5e-2
fft_weight: float = 0.05
lpips_weight: float = 0.5
latent_channels: tuple[int, ...] = (128, 256, 512, 512)
z_channels: int = 32
n_heads: int = 8
resnet_blocks: int = 2
@dataclass
class mmDiTConfig:
d_model: int = 768
n_heads: int = 32
n_blocks: int = 16
patch_size: int = 2
color_channels: int = 32 # z_channels
height: int = 32 # latent
width: int = 32 # latent
t_channels: int = 64