SDXL checks

brushnet/brushnet.py

@@ -482,9 +482,9 @@ class BrushNetModel(ModelMixin, ConfigMixin):
             conditioning_channels=conditioning_channels,
             flip_sin_to_cos=unet.config.flip_sin_to_cos,
             freq_shift=unet.config.freq_shift,
-            down_block_types=['DownBlock2D','DownBlock2D','DownBlock2D','DownBlock2D'],
+            down_block_types=["DownBlock2D" for block_name in unet.config.down_block_types],
             mid_block_type='MidBlock2D',
-            up_block_types=['UpBlock2D','UpBlock2D','UpBlock2D','UpBlock2D'],
+            up_block_types=["UpBlock2D" for block_name in unet.config.down_block_types],
             only_cross_attention=unet.config.only_cross_attention,
             block_out_channels=unet.config.block_out_channels,
             layers_per_block=unet.config.layers_per_block,

brushnet/brushnet_xl.json

@@ -0,0 +1,63 @@
+{
+  "_class_name": "BrushNetModel",
+  "_diffusers_version": "0.27.0.dev0",
+  "_name_or_path": "runs/logs/brushnetsdxl_randommask/checkpoint-80000",
+  "act_fn": "silu",
+  "addition_embed_type": "text_time",
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
+  "block_out_channels": [
+    320,
+    640,
+    1280
+  ],
+  "brushnet_conditioning_channel_order": "rgb",
+  "class_embed_type": null,
+  "conditioning_channels": 5,
+  "conditioning_embedding_out_channels": [
+    16,
+    32,
+    96,
+    256
+  ],
+  "cross_attention_dim": 2048,
+  "down_block_types": [
+    "DownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "global_pool_conditions": false,
+  "in_channels": 4,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "MidBlock2D",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_attention_heads": null,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "projection_class_embeddings_input_dim": 2816,
+  "resnet_time_scale_shift": "default",
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "up_block_types": [
+    "UpBlock2D",
+    "UpBlock2D",
+    "UpBlock2D"
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
+}