anchor commited on
Commit
05830e5
1 Parent(s): 7cb68a5

add new model, train unet motion and ipadapter, fix referencenet, t2i

Browse files
motion/musev_referencenet_pose/ip_adapter_image_proj.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f82b521bcf7831558ce2cc97721e84dbf36b793c2ec9fae16d63a8b2f0f7508
3
+ size 12603123
motion/musev_referencenet_pose/unet/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet3DConditionModel",
3
+ "_diffusers_version": "0.24.0.dev0",
4
+ "_name_or_path": "/cfs-datasets/public_models/fantasticmix_v10",
5
+ "act_fn": "silu",
6
+ "attention_head_dim": 8,
7
+ "block_out_channels": [
8
+ 320,
9
+ 640,
10
+ 1280,
11
+ 1280
12
+ ],
13
+ "center_input_sample": false,
14
+ "class_embed_type": null,
15
+ "class_embeddings_concat": false,
16
+ "conv_in_kernel": 3,
17
+ "conv_out_kernel": 3,
18
+ "cross_attention_dim": 768,
19
+ "down_block_types": [
20
+ "CrossAttnDownBlock3D",
21
+ "CrossAttnDownBlock3D",
22
+ "CrossAttnDownBlock3D",
23
+ "DownBlock3D"
24
+ ],
25
+ "downsample_padding": 1,
26
+ "dual_cross_attention": false,
27
+ "flip_sin_to_cos": true,
28
+ "freq_shift": 0,
29
+ "in_channels": 4,
30
+ "ip_adapter_cross_attn": true,
31
+ "keep_vision_condtion": true,
32
+ "layers_per_block": 2,
33
+ "mid_block_scale_factor": 1,
34
+ "mid_block_type": "UNetMidBlock3DCrossAttn",
35
+ "need_adain_temporal_cond": true,
36
+ "need_refer_emb": true,
37
+ "need_spatial_position_emb": false,
38
+ "need_t2i_facein": false,
39
+ "need_t2i_ip_adapter": true,
40
+ "need_t2i_ip_adapter_face": false,
41
+ "need_transformer_in": false,
42
+ "need_vis_cond_mask": false,
43
+ "need_zero_vis_cond_temb": true,
44
+ "norm_eps": 1e-05,
45
+ "norm_num_groups": 32,
46
+ "norm_spatial_length": true,
47
+ "num_class_embeds": null,
48
+ "only_cross_attention": false,
49
+ "out_channels": 4,
50
+ "projection_class_embeddings_input_dim": null,
51
+ "resnet_2d_skip_time_act": true,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 64,
54
+ "spatial_max_length": 2048,
55
+ "t2i_crossattn_ip_adapter_attn_processor": "T2IReferencenetIPAdapterXFormersAttnProcessor",
56
+ "t2i_ip_adapter_attn_processor": "NonParamReferenceIPXFormersAttnProcessor",
57
+ "temporal_conv_block": "TemporalConvLayer",
58
+ "temporal_transformer": "TransformerTemporalModel",
59
+ "time_cond_proj_dim": null,
60
+ "time_embedding_type": "positional",
61
+ "timestep_post_act": null,
62
+ "up_block_types": [
63
+ "UpBlock3D",
64
+ "CrossAttnUpBlock3D",
65
+ "CrossAttnUpBlock3D",
66
+ "CrossAttnUpBlock3D"
67
+ ],
68
+ "upcast_attention": false,
69
+ "use_anivv1_cfg": true,
70
+ "use_linear_projection": false
71
+ }
motion/musev_referencenet_pose/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69dd5cd26d9ff46b6c5e8879ed9b82aa7606810755cb2a9e5b3ed785242d0ebc
3
+ size 5930137448