Make config match the code at tuneavideo port for diffusers

unet/config.json

@@ -1,6 +1,6 @@
 {
   "_class_name": "UNet3DConditionModel",
-  "_diffusers_version": "0.11.1",
+  "_diffusers_version": "0.17.0.dev0",
   "act_fn": "silu",
   "attention_head_dim": 8,
   "block_out_channels": [
@@ -10,7 +10,6 @@
     1280
   ],
   "center_input_sample": false,
-  "class_embed_type": null,
   "cross_attention_dim": 768,
   "down_block_types": [
     "CrossAttnDownBlock3D",
@@ -19,16 +18,11 @@
     "DownBlock3D"
   ],
   "downsample_padding": 1,
-  "dual_cross_attention": false,
-  "flip_sin_to_cos": true,
-  "freq_shift": 0,
   "in_channels": 4,
   "layers_per_block": 2,
   "mid_block_scale_factor": 1,
-  "mid_block_type": "UNetMidBlock3DCrossAttn",
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
-  "num_class_embeds": null,
   "only_cross_attention": false,
   "out_channels": 4,
   "resnet_time_scale_shift": "default",
@@ -41,7 +35,6 @@
   ],
   "upcast_attention": false,
   "use_linear_projection": false,
-  "conv_type": "inflated_conv_3d",
   "use_temporal_transformer": false, 
   "use_temporal_conv": false,
   "sub_blocks_type": "3d"