Update configs

diffusion_net/config.json

@@ -2,7 +2,6 @@
   "_center_input_sample": false,
   "_class_name": "UNet3DConditionModel",
   "_diffusers_version": "0.31.0",
-  "_landmark_net": false,
   "_out_channels": 4,
   "act_fn": "silu",
   "addition_embed_type": null,
@@ -32,7 +31,6 @@
   "dropout": 0.0,
   "dual_cross_attention": false,
   "emo_drop_rate": 0.05,
-  "emotion_signal": true,
   "encoder_hid_dim": null,
   "encoder_hid_dim_type": null,
   "flip_sin_to_cos": true,
@@ -42,7 +40,6 @@
   "mid_block_only_cross_attention": null,
   "mid_block_scale_factor": 1,
   "mid_block_type": "UNetMidBlock3DCrossAttn",
-  "motion_module_decoder_only": false,
   "motion_module_kwargs": {
     "attention_block_types": [
       "Temporal_Self",
@@ -52,17 +49,14 @@
     "num_transformer_block": 1,
     "temporal_attention_dim_div": 1,
     "temporal_position_encoding": true,
-    "temporal_position_encoding_max_len": 32,
-    "use_linear_attn": true
+    "temporal_position_encoding_max_len": 32
   },
-  "motion_module_mid_block": true,
   "motion_module_resolutions": [
     1,
     2,
     4,
     8
   ],
-  "motion_module_type": "MemoryLinearAttn",
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
   "num_attention_heads": null,
@@ -73,24 +67,12 @@
   "resnet_time_scale_shift": "default",
   "reverse_transformer_layers_per_block": null,
   "sample_size": 64,
-  "stack_enable_blocks_depth": [
-    0,
-    1,
-    2,
-    3
-  ],
-  "stack_enable_blocks_name": [
-    "up",
-    "down",
-    "mid"
-  ],
   "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,
-  "two_branches_atten": true,
   "unet_use_cross_frame_attention": false,
   "unet_use_temporal_attention": false,
   "up_block_types": [
@@ -100,10 +82,6 @@
     "CrossAttnUpBlock3D"
   ],
   "upcast_attention": false,
-  "use_audio_module": true,
-  "use_face_masks": false,
   "use_inflated_groupnorm": true,
-  "use_linear_projection": false,
-  "use_motion_module": true,
-  "use_past_frames": true
+  "use_linear_projection": false
 }

reference_net/config.json

@@ -2,7 +2,6 @@
   "_center_input_sample": false,
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.31.0",
-  "_landmark_net": false,
   "_out_channels": 4,
   "act_fn": "silu",
   "addition_embed_type": null,