Revert "chore: disable safety checker & bump versions"

This reverts commit 1fdb17a069391cde04cf676f1c0a3ade4f2f3d67.

model_index.json

@@ -1,12 +1,33 @@
 {
   "_class_name": "StableDiffusionPipeline",
-  "_diffusers_version": "0.24.0",
-  "feature_extractor": ["transformers", "CLIPFeatureExtractor"],
-  "requires_safety_checker": false,
-  "safety_checker": ["stable_diffusion", "StableDiffusionSafetyChecker"],
-  "scheduler": ["diffusers", "DPMSolverMultistepScheduler"],
-  "text_encoder": ["transformers", "CLIPTextModel"],
-  "tokenizer": ["transformers", "CLIPTokenizer"],
-  "unet": ["diffusers", "UNet2DConditionModel"],
-  "vae": ["diffusers", "AutoencoderKL"]
+  "_diffusers_version": "0.17.0.dev0",
+  "feature_extractor": [
+    "transformers",
+    "CLIPFeatureExtractor"
+  ],
+  "requires_safety_checker": true,
+  "safety_checker": [
+    "stable_diffusion",
+    "StableDiffusionSafetyChecker"
+  ],
+  "scheduler": [
+    "diffusers",
+    "DPMSolverMultistepScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "CLIPTextModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DConditionModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
 }

safety_checker/config.json

@@ -1,7 +1,9 @@
 {
   "_commit_hash": "cb41f3a270d63d454d385fc2e4f571c487c253c5",
   "_name_or_path": "CompVis/stable-diffusion-safety-checker",
-  "architectures": ["StableDiffusionSafetyChecker"],
+  "architectures": [
+    "StableDiffusionSafetyChecker"
+  ],
   "initializer_factor": 1.0,
   "logit_scale_init_value": 2.6592,
   "model_type": "clip",
@@ -42,7 +44,7 @@
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layer_norm_eps": 1e-5,
+    "layer_norm_eps": 1e-05,
     "length_penalty": 1.0,
     "max_length": 20,
     "max_position_embeddings": 77,
@@ -78,7 +80,7 @@
     "top_p": 1.0,
     "torch_dtype": null,
     "torchscript": false,
-    "transformers_version": "4.35.2",
+    "transformers_version": "4.28.1",
     "typical_p": 1.0,
     "use_bfloat16": false,
     "vocab_size": 49408
@@ -122,7 +124,7 @@
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layer_norm_eps": 1e-5,
+    "layer_norm_eps": 1e-05,
     "length_penalty": 1.0,
     "max_length": 20,
     "min_length": 0,
@@ -159,7 +161,7 @@
     "top_p": 1.0,
     "torch_dtype": null,
     "torchscript": false,
-    "transformers_version": "4.35.2",
+    "transformers_version": "4.28.1",
     "typical_p": 1.0,
     "use_bfloat16": false
   }

scheduler/scheduler_config.json

@@ -1,6 +1,6 @@
 {
   "_class_name": "DPMSolverMultistepScheduler",
-  "_diffusers_version": "0.24.0",
+  "_diffusers_version": "0.17.0.dev0",
   "algorithm_type": "dpmsolver++",
   "beta_end": 0.012,
   "beta_schedule": "scaled_linear",

text_encoder/config.json

@@ -1,6 +1,8 @@
 {
   "_name_or_path": "openai/clip-vit-large-patch14",
-  "architectures": ["CLIPTextModel"],
+  "architectures": [
+    "CLIPTextModel"
+  ],
   "attention_dropout": 0.0,
   "bos_token_id": 0,
   "dropout": 0.0,
@@ -10,7 +12,7 @@
   "initializer_factor": 1.0,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "layer_norm_eps": 1e-5,
+  "layer_norm_eps": 1e-05,
   "max_position_embeddings": 77,
   "model_type": "clip_text_model",
   "num_attention_heads": 12,
@@ -18,6 +20,6 @@
   "pad_token_id": 1,
   "projection_dim": 768,
   "torch_dtype": "float16",
-  "transformers_version": "4.35.2",
+  "transformers_version": "4.28.1",
   "vocab_size": 49408
 }

unet/config.json

@@ -1,11 +1,16 @@
 {
   "_class_name": "UNet2DConditionModel",
-  "_diffusers_version": "0.24.0",
+  "_diffusers_version": "0.17.0.dev0",
   "act_fn": "silu",
   "addition_embed_type": null,
   "addition_embed_type_num_heads": 64,
   "attention_head_dim": 8,
-  "block_out_channels": [320, 640, 1280, 1280],
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
   "center_input_sample": false,
   "class_embed_type": null,
   "class_embeddings_concat": false,
@@ -29,7 +34,7 @@
   "mid_block_only_cross_attention": null,
   "mid_block_scale_factor": 1,
   "mid_block_type": "UNetMidBlock2DCrossAttn",
-  "norm_eps": 1e-5,
+  "norm_eps": 1e-05,
   "norm_num_groups": 32,
   "num_class_embeds": null,
   "only_cross_attention": false,

vae/config.json

@@ -1,8 +1,13 @@
 {
   "_class_name": "AutoencoderKL",
-  "_diffusers_version": "0.24.0",
+  "_diffusers_version": "0.17.0.dev0",
   "act_fn": "silu",
-  "block_out_channels": [128, 256, 512, 512],
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
   "down_block_types": [
     "DownEncoderBlock2D",
     "DownEncoderBlock2D",