| { | |
| "_class_name": "UNet3DConditionModel", | |
| "_diffusers_version": "0.17.0.dev0", | |
| "act_fn": "silu", | |
| "attention_head_dim": 8, | |
| "block_out_channels": [ | |
| 320, | |
| 640, | |
| 1280, | |
| 1280 | |
| ], | |
| "center_input_sample": false, | |
| "cross_attention_dim": 768, | |
| "down_block_types": [ | |
| "CrossAttnDownBlockInflated3D", | |
| "CrossAttnDownBlockInflated3D", | |
| "CrossAttnDownBlockInflated3D", | |
| "DownBlockInflated3D" | |
| ], | |
| "downsample_padding": 1, | |
| "in_channels": 4, | |
| "layers_per_block": 2, | |
| "mid_block_scale_factor": 1, | |
| "mid_block_type" : "UNetMidBlockInflated3DCrossAttn", | |
| "norm_eps": 1e-05, | |
| "norm_num_groups": 32, | |
| "only_cross_attention": false, | |
| "out_channels": 4, | |
| "resnet_time_scale_shift": "default", | |
| "sample_size": 64, | |
| "up_block_types": [ | |
| "UpBlockInflated3D", | |
| "CrossAttnUpBlockInflated3D", | |
| "CrossAttnUpBlockInflated3D", | |
| "CrossAttnUpBlockInflated3D" | |
| ], | |
| "upcast_attention": false, | |
| "use_linear_projection": false, | |
| "use_temporal_transformer": false | |
| } | |