update motion model, musev and musev_referencenet

motion/musev/unet/config.json

@@ -0,0 +1,63 @@
+{
+  "_class_name": "UNet3DConditionModel",
+  "_diffusers_version": "0.21.0.dev0",
+  "_name_or_path": "/cfs-datasets/users/anchorxia/experiments/musev/outputs/20231018_AnivV0_tfm=TT_2SelfAttn_reference-only_noise=vf/checkpoints/checkpoint-16400/",
+  "act_fn": "silu",
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 768,
+  "down_block_types": [
+    "CrossAttnDownBlock3D",
+    "CrossAttnDownBlock3D",
+    "CrossAttnDownBlock3D",
+    "DownBlock3D"
+  ],
+  "downsample_padding": 1,
+  "dual_cross_attention": false,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "keep_vision_condtion": false,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock3DCrossAttn",
+  "need_adain_temporal_cond": true,
+  "need_spatial_position_emb": false,
+  "need_t2i_ip_adapter": true,
+  "need_transformer_in": true,
+  "need_zero_vis_cond_temb": true,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_2d_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 64,
+  "t2i_ip_adapter_attn_processor": "NonParamReferenceIPXFormersAttnProcessor",
+  "temporal_conv_block": "TemporalConvLayer",
+  "temporal_transformer": "TransformerTemporalModel",
+  "time_cond_proj_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "up_block_types": [
+    "UpBlock3D",
+    "CrossAttnUpBlock3D",
+    "CrossAttnUpBlock3D",
+    "CrossAttnUpBlock3D"
+  ],
+  "upcast_attention": false,
+  "use_anivv1_cfg": false,
+  "use_linear_projection": false
+}

motion/musev/unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb9579635bb9aa892919ed29bb4163b13c8e76833b628bd30899f8cb854b1b24
+size 5680558644

motion/musev_referencenet/ip_adapter_image_proj.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d590237340010d0e744470650ffa6ae441eeb64dca685d10f30492feb63cca91
+size 12603123

motion/musev_referencenet/referencenet/config.json

@@ -0,0 +1,71 @@
+{
+  "_class_name": "ReferenceNet2D",
+  "_diffusers_version": "0.24.0.dev0",
+  "_name_or_path": "/cfs-datasets/users/anchorxia/proj/musev/outputs/20231219_m=AnivV2_visnoise=1_referencenet_downblock_ipadapter/checkpoints/checkpoint-50000",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": null,
+  "attention_head_dim": 8,
+  "attention_type": "default",
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 1280,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dropout": 0.0,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "layers_per_block": 2,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "need_block_embs": true,
+  "need_self_attn_block_embs": false,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_attention_heads": null,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "remove_unused_layer": true,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "reverse_transformer_layers_per_block": null,
+  "sample_size": null,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "transformer_layers_per_block": 1,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": false
+}

motion/musev_referencenet/referencenet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c0b063bd509a8461e095565ca754f50d0e322998c4036c4191839388cc52205
+size 1394979885

motion/musev_referencenet/unet/config.json

@@ -0,0 +1,68 @@
+{
+  "_class_name": "UNet3DConditionModel",
+  "_diffusers_version": "0.24.0.dev0",
+  "_name_or_path": "/cfs-datasets/users/anchorxia/proj/musev/outputs/20231219_m=AnivV2_visnoise=1_referencenet_downblock_ipadapter/checkpoints/checkpoint-50000",
+  "act_fn": "silu",
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 768,
+  "down_block_types": [
+    "CrossAttnDownBlock3D",
+    "CrossAttnDownBlock3D",
+    "CrossAttnDownBlock3D",
+    "DownBlock3D"
+  ],
+  "downsample_padding": 1,
+  "dual_cross_attention": false,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "ip_adapter_cross_attn": true,
+  "keep_vision_condtion": true,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock3DCrossAttn",
+  "need_adain_temporal_cond": true,
+  "need_refer_emb": true,
+  "need_spatial_position_emb": false,
+  "need_t2i_ip_adapter": true,
+  "need_transformer_in": false,
+  "need_zero_vis_cond_temb": true,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "norm_spatial_length": true,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_2d_skip_time_act": true,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 64,
+  "spatial_max_length": 2048,
+  "t2i_crossattn_ip_adapter_attn_processor": "T2IReferencenetIPAdapterXFormersAttnProcessor",
+  "t2i_ip_adapter_attn_processor": "NonParamReferenceIPXFormersAttnProcessor",
+  "temporal_conv_block": "TemporalConvLayer",
+  "temporal_transformer": "TransformerTemporalModel",
+  "time_cond_proj_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "up_block_types": [
+    "UpBlock3D",
+    "CrossAttnUpBlock3D",
+    "CrossAttnUpBlock3D",
+    "CrossAttnUpBlock3D"
+  ],
+  "upcast_attention": false,
+  "use_anivv1_cfg": true,
+  "use_linear_projection": false
+}

motion/musev_referencenet/unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36807f2ff465566e9d8ba99c56f7c93380386099fbb8b6251a80406d4b089699
+size 5930137448