Add controlnet model

.gitignore

@@ -5,8 +5,8 @@ __pycache__/
 
 
 # Weights artefacts
-*.pth
-*.safetensors
+#*.pth
+#*.safetensors
 
 # C extensions
 *.so

README.md

@@ -21,7 +21,7 @@ a handsome man relaxing in a chair, shirt widely unbuttoned, eyes closed, close
 # TODO
 
 [x] изучить как обернуть в huggingface app
-[ ] запушить модель
+[ ] запушить модель (https://huggingface.co/docs/hub/repositories-getting-started)
 [ ] Пример как можно сделать апку: https://huggingface.co/spaces/wildoctopus/cloth-segmentation/tree/main
 [ ] Написать UI с RGB цветами (чтобы можно было одежду по шаблону а не изображению генерить)
 [ ] Добавить seed в параметры 

config.py

@@ -12,11 +12,9 @@ class PipelineConfig:
     """Fashion Controlnet Pipeline Config."""
 
     base_model_path: str = 'stabilityai/stable-diffusion-xl-base-1.0'
-    # /pub/home/korostelev/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b'
 
-    controlnet_path: str = r"C:\Users\dragynir\.cache\huggingface\hub\fashion_controlnet"
-
-    # https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
     vae_path: str = 'madebyollin/sdxl-vae-fp16-fix'
 
+    controlnet_path: str = os.path.join(weights_path, 'controlnet')
+
     segmentation_model_path: str = os.path.join(weights_path, 'cloth_segm.pth')

weights/controlnet/config.json

@@ -0,0 +1,57 @@
+{
+  "_class_name": "ControlNetModel",
+  "_diffusers_version": "0.25.0.dev0",
+  "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
+  ],
+  "class_embed_type": null,
+  "conditioning_channels": 3,
+  "conditioning_embedding_out_channels": [
+    16,
+    32,
+    96,
+    256
+  ],
+  "controlnet_conditioning_channel_order": "rgb",
+  "cross_attention_dim": 2048,
+  "down_block_types": [
+    "DownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D"
+  ],
+  "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": "UNetMidBlock2DCrossAttn",
+  "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
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
+}

weights/controlnet/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f85f8ade2174d2757de3b00d302309d99e0b7995dc45f610ad9bf3d9aac95470
+size 5004167864