PAIR
/

Text-to-Video

How should these parameters be set?

#5
by yue1123 - opened

num_frame_embeddings_range: 16
start_frame_training: 16
start_frame_ctrl: 16
load_trained_base_model_and_resampler_from_ckpt: ''
load_trained_controlnet_from_ckpt: ''
unet_params:
class_path: t2v_enhanced.model.pl_module_params_controlnet.UNetParams
init_args:
conditioning_embedding_out_channels:
- 32
- 96
- 256
- 512
ckpt_spatial_layers: ''

Sign up or log in to comment