Update unet/config.json

unet/config.json

@@ -1,7 +1,6 @@
 {
   "_class_name": "UNet2DConditionModel",
-  "_diffusers_version": "0.33.1",
-  "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
+  "_diffusers_version": "0.19.0.dev0",
   "act_fn": "silu",
   "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
@@ -11,7 +10,6 @@
     10,
     20
   ],
-  "attention_type": "default",
   "block_out_channels": [
     320,
     640,
@@ -30,7 +28,6 @@
     "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
-  "dropout": 0.0,
   "dual_cross_attention": false,
   "encoder_hid_dim": null,
   "encoder_hid_dim_type": null,
@@ -40,7 +37,7 @@
   "layers_per_block": 2,
   "mid_block_only_cross_attention": null,
   "mid_block_scale_factor": 1,
-  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "mid_block_type": "UNetMidBlock2D",
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
   "num_attention_heads": null,
@@ -51,7 +48,6 @@
   "resnet_out_scale_factor": 1.0,
   "resnet_skip_time_act": false,
   "resnet_time_scale_shift": "default",
-  "reverse_transformer_layers_per_block": null,
   "sample_size": 128,
   "time_cond_proj_dim": null,
   "time_embedding_act_fn": null,
@@ -59,9 +55,14 @@
   "time_embedding_type": "positional",
   "timestep_post_act": null,
   "transformer_layers_per_block": [
-    1,
-    2,
-    10
+    [1],
+    [2,2],
+    [4,4]
+  ],
+  "reverse_transformer_layers_per_block": [
+      [4,4,10],
+      [2,1,1],
+      1
   ],
   "up_block_types": [
     "CrossAttnUpBlock2D",