End of training

README.md

@@ -9,6 +9,12 @@ tags:
 - diffusers
 - controlnet
 - diffusers-training
+- stable-diffusion-xl
+- stable-diffusion-xl-diffusers
+- text-to-image
+- diffusers
+- controlnet
+- diffusers-training
 inference: true
 ---
 

checkpoint-1000/controlnet/config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

checkpoint-1000/controlnet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9aa0310da946922bbe1f437f8d8ba5fe8faf5a4c2de73180ef17ad7e0d78e947
-size 1445157120
+oid sha256:0734204635cfe55040059aa58cea0c732ab50620b61ca44ca8ac5597b371206c
+size 5004167864

checkpoint-1000/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b3b5c904ce497113ccd1ff6dc01fda23ed770393c031781f5d1f2864c573a92
-size 2890518478
+oid sha256:d08d48073aef84329905df4bf609f2dfa9d07ac9360b24ada2b3037d496f6b02
+size 2510393614

checkpoint-1000/random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:902fc90c41bac04feca801e2d338fad3e10833af3a6212b2b9fcbaa9128cd5ad
+oid sha256:196e4fcfd0dc67028502faaa92729774ca4a68325575f7b3553a6484afb49ac3
 size 14408

checkpoint-2000/controlnet/config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

checkpoint-2000/controlnet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6edb8ba340ab1c23d012b360eec63b5c765057c2da6293cd463bbf0fb4dd66f8
-size 1445157120
+oid sha256:022c2c7e62360ea64552fe0a7e9075bf27c95e849b3eef271b67188fb7027f00
+size 5004167864

checkpoint-2000/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c5a3e062d915355b7d070304dfbf7b769f8aba5803e206ce41372765467f2a1
-size 2890518478
+oid sha256:915d2aaccf8beff3d8274ebc0529c858e40ab86373e5ec24ee60e0cdd75044ba
+size 2510393614

checkpoint-2000/random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:859f55f7a36db49767fdb103144102f47b2b087ecfa2d56c316891fb47d4c008
-size 14344
+oid sha256:b74b7db71964a1040fda20a42d565a3dd95b465980105cae6faf3860b55158b8
+size 14408

checkpoint-3000/controlnet/config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

checkpoint-3000/controlnet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71169c4001d76b9c0324874a5fff03c013c45d1d9c5ad410ac0f8cf27ceee5b2
-size 1445157120
+oid sha256:c01d53f8299ada75eea82872941c4b0d936788a2598d3f9e09b5a6b0f9586e22
+size 5004167864

checkpoint-3000/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdf05e5ec234decbac48a2d33b73d31245e0f2771ea20226ba493caefe91d0a9
-size 2890518478
+oid sha256:5227e59e6e9688b1619cef0381488200a24052c39b9334cd95c87d73fbca884e
+size 2510393614

checkpoint-3000/random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d47dab87ea4581a9837164914f8b6409a9cb2851d2314fe0a1480921b0fd3946
+oid sha256:30db01d6364583d05e85e8326c408d47d042b5b3f1c56142e1735750a03017a0
 size 14408

checkpoint-4000/controlnet/config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

checkpoint-4000/controlnet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ca8ca1ecd17ca9c41058247fa69761e1224c4ace2b9d7a74fd213f4590e5ce6
-size 1445157120
+oid sha256:c61efd37ff8816d8de1db8cd76cf3a92b0b8542f8b81368cdbcac10b26ddca3c
+size 5004167864

checkpoint-4000/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63c703e3f30d1caba9709ea660659c74d24818737b81248ce973166ab6894711
-size 2890518478
+oid sha256:11f1019905254b71867d7b7160f0bb0bd8bc80daa971ec709c7c7d4e047e9372
+size 2510393614

checkpoint-4000/random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8339a87c15d5479a891291b77672b772078d1f580acef23c5dbbc0b9bab54c1
-size 14344
+oid sha256:d31cd70c2ba15a4cf4655b2bd65670965992b23da3e91e09a7257c900537bcfd
+size 14408

checkpoint-5000/controlnet/config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

checkpoint-5000/controlnet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e452225e153c779250f0f72565d4853d123933307a01b8e801582c1cc0da1d9d
-size 1445157120
+oid sha256:0124f2e083b565fd6f304bf709528f7b4d4ad94d71734384bc9b073d2ff266d0
+size 5004167864

checkpoint-5000/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8cd0790e3e4d65f80df27df50f1fbe2711274ef8cb9ea56e6b2c18d6b1a1fce0
-size 2890518478
+oid sha256:ed1591a442222e0390b55fd07ddcac5c93be31c180649d36921031845303aa1c
+size 2510393614

checkpoint-5000/random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71543d0c595b08536e9a5829f7ab2aeea94955f6662cbb466020880500f53052
+oid sha256:3d9880959925c08f7cff7f53206eda7aedf0ee3cacfdee243598699a2f876dd0
 size 14408

config.json

@@ -1,15 +1,18 @@
 {
   "_class_name": "ControlNetModel",
-  "_diffusers_version": "0.32.0.dev0",
+  "_diffusers_version": "0.35.0.dev0",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "addition_time_embed_dim": 256,
+  "attention_head_dim": [
+    5,
+    10,
+    20
+  ],
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "class_embed_type": null,
@@ -21,12 +24,11 @@
     256
   ],
   "controlnet_conditioning_channel_order": "rgb",
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 2048,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "encoder_hid_dim": null,
@@ -43,9 +45,13 @@
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_time_scale_shift": "default",
-  "transformer_layers_per_block": 1,
-  "upcast_attention": false,
-  "use_linear_projection": false
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
+  "upcast_attention": null,
+  "use_linear_projection": true
 }

diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e452225e153c779250f0f72565d4853d123933307a01b8e801582c1cc0da1d9d
-size 1445157120
+oid sha256:0124f2e083b565fd6f304bf709528f7b4d4ad94d71734384bc9b073d2ff266d0
+size 5004167864