import torch from torch import autocast from diffusers import StableDiffusionPipeline pipe = StableDiffusionPipeline.from_pretrained( 'hakurei/waifu-diffusion', torch_dtype=torch.float32 ).to('cuda') prompt = "1girl, aqua eyes, without clothes, blonde hair, open mouth, earrings, beach background, hat, hoop earrings, jewelry, looking at viewer, without shirt, short hair, simple background, solo, upper body, without shirt" with autocast("cuda"): image = pipe(prompt, guidance_scale=6)["sample"][0] image.save("test.png")

.gitattributes

@@ -33,11 +33,3 @@ safety_checker/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
 text_encoder/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
 unet/diffusion_pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
 vae/diffusion_pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
-text_encoder/model.safetensors filter=lfs diff=lfs merge=lfs -text
-unet/diffusion_pytorch_model.safetensors filter=lfs diff=lfs merge=lfs -text
-vae/diffusion_pytorch_model.safetensors filter=lfs diff=lfs merge=lfs -text
-safety_checker/model.safetensors filter=lfs diff=lfs merge=lfs -text
-unet/diffusion_pytorch_model.fp16.safetensors filter=lfs diff=lfs merge=lfs -text
-text_encoder/model.fp16.safetensors filter=lfs diff=lfs merge=lfs -text
-vae/diffusion_pytorch_model.fp16.safetensors filter=lfs diff=lfs merge=lfs -text
-safety_checker/model.fp16.safetensors filter=lfs diff=lfs merge=lfs -text

README.md

@@ -5,19 +5,17 @@ tags:
 - stable-diffusion
 - text-to-image
 license: creativeml-openrail-m
-inference: true
+inference: false
 
 ---
 
-# waifu-diffusion v1.4 - Diffusion for Weebs
+# waifu-diffusion v1.3 - Diffusion for Weebs
 
 waifu-diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through fine-tuning.
 
-![image](https://user-images.githubusercontent.com/26317155/210155933-db3a5f1a-1ec3-4777-915c-6deff2841ce9.png)
+<img src=https://i.imgur.com/Y5Tmw1S.png width=75% height=75%>
 
-<sub>masterpiece, best quality, 1girl, green hair, sweater, looking at viewer, upper body, beanie, outdoors, watercolor, night, turtleneck</sub>
-
-[Original Weights](https://huggingface.co/hakurei/waifu-diffusion-v1-4)
+[Original Weights](https://huggingface.co/hakurei/waifu-diffusion-v1-3)
 
 # Gradio & Colab
 
@@ -64,9 +62,9 @@ image.save("test.png")
 
 ## Team Members and Acknowledgements
 
-This project would not have been possible without the incredible work by Stability AI and Novel AI.
+This project would not have been possible without the incredible work by the [CompVis Researchers](https://ommer-lab.com/).
 
-- [Haru](https://github.com/harubaru)
+- [Anthony Mercurio](https://github.com/harubaru)
 - [Salt](https://github.com/sALTaccount/)
 - [Sta @ Bit192](https://twitter.com/naclbbr)
 

feature_extractor/preprocessor_config.json

@@ -1,12 +1,8 @@
 {
-  "crop_size": {
-    "height": 224,
-    "width": 224
-  },
+  "crop_size": 224,
   "do_center_crop": true,
   "do_convert_rgb": true,
   "do_normalize": true,
-  "do_rescale": true,
   "do_resize": true,
   "feature_extractor_type": "CLIPFeatureExtractor",
   "image_mean": [
@@ -14,15 +10,11 @@
     0.4578275,
     0.40821073
   ],
-  "image_processor_type": "CLIPImageProcessor",
   "image_std": [
     0.26862954,
     0.26130258,
     0.27577711
   ],
   "resample": 3,
-  "rescale_factor": 0.00392156862745098,
-  "size": {
-    "shortest_edge": 224
-  }
+  "size": 224
 }

model_index.json

@@ -1,18 +1,17 @@
 {
   "_class_name": "StableDiffusionPipeline",
-  "_diffusers_version": "0.10.2",
+  "_diffusers_version": "0.4.1",
   "feature_extractor": [
     "transformers",
-    "CLIPImageProcessor"
+    "CLIPFeatureExtractor"
   ],
-  "requires_safety_checker": true,
   "safety_checker": [
     "stable_diffusion",
     "StableDiffusionSafetyChecker"
   ],
   "scheduler": [
     "diffusers",
-    "PNDMScheduler"
+    "LMSDiscreteScheduler"
   ],
   "text_encoder": [
     "transformers",

safety_checker/config.json

@@ -1,5 +1,5 @@
 {
-  "_commit_hash": "cb41f3a270d63d454d385fc2e4f571c487c253c5",
+  "_commit_hash": null,
   "_name_or_path": "CompVis/stable-diffusion-safety-checker",
   "architectures": [
     "StableDiffusionSafetyChecker"
@@ -14,7 +14,6 @@
     "architectures": null,
     "attention_dropout": 0.0,
     "bad_words_ids": null,
-    "begin_suppress_tokens": null,
     "bos_token_id": 0,
     "chunk_size_feed_forward": 0,
     "cross_attention_hidden_size": null,
@@ -62,14 +61,12 @@
     "pad_token_id": 1,
     "prefix": null,
     "problem_type": null,
-    "projection_dim": 512,
     "pruned_heads": {},
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,
     "return_dict": true,
     "return_dict_in_generate": false,
     "sep_token_id": null,
-    "suppress_tokens": null,
     "task_specific_params": null,
     "temperature": 1.0,
     "tf_legacy_loss": false,
@@ -80,7 +77,7 @@
     "top_p": 1.0,
     "torch_dtype": null,
     "torchscript": false,
-    "transformers_version": "4.25.1",
+    "transformers_version": "4.22.2",
     "typical_p": 1.0,
     "use_bfloat16": false,
     "vocab_size": 49408
@@ -99,7 +96,6 @@
     "architectures": null,
     "attention_dropout": 0.0,
     "bad_words_ids": null,
-    "begin_suppress_tokens": null,
     "bos_token_id": null,
     "chunk_size_feed_forward": 0,
     "cross_attention_hidden_size": null,
@@ -149,14 +145,12 @@
     "patch_size": 14,
     "prefix": null,
     "problem_type": null,
-    "projection_dim": 512,
     "pruned_heads": {},
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,
     "return_dict": true,
     "return_dict_in_generate": false,
     "sep_token_id": null,
-    "suppress_tokens": null,
     "task_specific_params": null,
     "temperature": 1.0,
     "tf_legacy_loss": false,
@@ -167,7 +161,7 @@
     "top_p": 1.0,
     "torch_dtype": null,
     "torchscript": false,
-    "transformers_version": "4.25.1",
+    "transformers_version": "4.22.2",
     "typical_p": 1.0,
     "use_bfloat16": false
   },

safety_checker/model.fp16.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:08902f19b1cfebd7c989f152fc0507bef6898c706a91d666509383122324b511
-size 608018440

safety_checker/model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9d6a233ff6fd5ccb9f76fd99618d73369c52dd3d8222376384d0e601911089e8
-size 1215981830

safety_checker/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16d28f2b37109f222cdc33620fdd262102ac32112be0352a7f77e9614b35a394
-size 1216064769
+oid sha256:193490b58ef62739077262e833bf091c66c29488058681ac25cf7df3d8190974
+size 1216061799

safety_checker/pytorch_model.fp16.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:22ba87205445ad5def13e54919b038dcfb7321ec1c3f4b12487d4fba6036125f
-size 608103564

scheduler/scheduler_config.json

@@ -1,14 +1,9 @@
 {
-  "_class_name": "PNDMScheduler",
-  "_diffusers_version": "0.10.2",
+  "_class_name": "LMSDiscreteScheduler",
+  "_diffusers_version": "0.4.1",
   "beta_end": 0.012,
   "beta_schedule": "scaled_linear",
   "beta_start": 0.00085,
-  "clip_sample": false,
   "num_train_timesteps": 1000,
-  "prediction_type": "epsilon",
-  "set_alpha_to_one": false,
-  "skip_prk_steps": true,
-  "steps_offset": 1,
   "trained_betas": null
 }

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/mnt/sd-finetune-data/finetunes/step_57000",
+  "_name_or_path": "openai/clip-vit-large-patch14",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -7,19 +7,19 @@
   "bos_token_id": 0,
   "dropout": 0.0,
   "eos_token_id": 2,
-  "hidden_act": "gelu",
-  "hidden_size": 1024,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 768,
   "initializer_factor": 1.0,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 77,
   "model_type": "clip_text_model",
-  "num_attention_heads": 16,
-  "num_hidden_layers": 23,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
   "pad_token_id": 1,
-  "projection_dim": 512,
+  "projection_dim": 768,
   "torch_dtype": "float32",
-  "transformers_version": "4.25.1",
+  "transformers_version": "4.22.2",
   "vocab_size": 49408
 }

text_encoder/model.fp16.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:22bc8e104d064b678ef7d2d2b217d4a8c9bfb79fb35792417cdf228e70adc7fb
-size 680821096

text_encoder/model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:661a5d7f8e19fce696aa9d932ab97b546b4d4a2a2d87238a17761bef2704269f
-size 1361597016

text_encoder/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:040fc6498aa3cdbb926dc2d01c3d6629521e5f085d901d5e8d8c2b0e0aa2b1ce
-size 1361679905
+oid sha256:770a47a9ffdcfda0b05506a7888ed714d06131d60267e6cf52765d61cf59fd67
+size 492305335

text_encoder/pytorch_model.fp16.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ba47a6c751cce5b6c4d5c79c8cd63bab63bfce3539e7805a70b8dca9d1f2f151
-size 680900852

tokenizer/special_tokens_map.json

@@ -13,7 +13,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "!",
+  "pad_token": "<|endoftext|>",
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer/tokenizer_config.json

@@ -19,7 +19,7 @@
   },
   "errors": "replace",
   "model_max_length": 77,
-  "name_or_path": "/mnt/sd-finetune-data/finetunes/step_57000",
+  "name_or_path": "openai/clip-vit-large-patch14",
   "pad_token": "<|endoftext|>",
   "special_tokens_map_file": "./special_tokens_map.json",
   "tokenizer_class": "CLIPTokenizer",

unet/config.json

@@ -1,14 +1,8 @@
 {
   "_class_name": "UNet2DConditionModel",
-  "_diffusers_version": "0.10.2",
-  "_name_or_path": "/mnt/sd-finetune-data/finetunes/step_57000",
+  "_diffusers_version": "0.4.1",
   "act_fn": "silu",
-  "attention_head_dim": [
-    5,
-    10,
-    20,
-    20
-  ],
+  "attention_head_dim": 8,
   "block_out_channels": [
     320,
     640,
@@ -16,7 +10,7 @@
     1280
   ],
   "center_input_sample": false,
-  "cross_attention_dim": 1024,
+  "cross_attention_dim": 768,
   "down_block_types": [
     "CrossAttnDownBlock2D",
     "CrossAttnDownBlock2D",
@@ -24,7 +18,6 @@
     "DownBlock2D"
   ],
   "downsample_padding": 1,
-  "dual_cross_attention": false,
   "flip_sin_to_cos": true,
   "freq_shift": 0,
   "in_channels": 4,
@@ -32,16 +25,12 @@
   "mid_block_scale_factor": 1,
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
-  "num_class_embeds": null,
-  "only_cross_attention": false,
   "out_channels": 4,
-  "sample_size": 64,
+  "sample_size": 32,
   "up_block_types": [
     "UpBlock2D",
     "CrossAttnUpBlock2D",
     "CrossAttnUpBlock2D",
     "CrossAttnUpBlock2D"
-  ],
-  "upcast_attention": false,
-  "use_linear_projection": true
+  ]
 }

unet/diffusion_pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24d2d0a39a4cd06869c91173d507cb153f272a1a328514f70b7ce9b48cab7e2b
-size 3463934693
+oid sha256:f47e5665f0e85155a5f6f58683b04940c6b132023d584396226bf54419a78831
+size 3438354725

unet/diffusion_pytorch_model.fp16.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:96edda2701914e1e248197bb205305e6aa9cfc776c3372cff9eaf62d4706a3cf
-size 1732107093

unet/diffusion_pytorch_model.fp16.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7b5cdd1c15f025166ded673f898e09e621c9ff828d6508a81e83378a6d0ba8dd
-size 1731904736

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:dda5a15fe85e6ea7fe0e21d06264611246ab60bdbf7001daa1e48028a49cd2e3
-size 3463726500

vae/config.json

@@ -1,7 +1,6 @@
 {
   "_class_name": "AutoencoderKL",
-  "_diffusers_version": "0.10.2",
-  "_name_or_path": "/mnt/sd-finetune-data/base/vae",
+  "_diffusers_version": "0.4.1",
   "act_fn": "silu",
   "block_out_channels": [
     128,
@@ -20,7 +19,7 @@
   "layers_per_block": 2,
   "norm_num_groups": 32,
   "out_channels": 3,
-  "sample_size": 512,
+  "sample_size": 256,
   "up_block_types": [
     "UpDecoderBlock2D",
     "UpDecoderBlock2D",

vae/diffusion_pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e174991e5609bc5c2b3995e3f223fb2c5f0ae3be307fa9591b351d837a08770
-size 334711857
+oid sha256:1b134cded8eb78b184aefb8805b6b572f36fa77b255c483665dda931fa0130c5
+size 334707217

vae/diffusion_pytorch_model.fp16.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d207e4928394a2002bcb7fff829e93bbd2a44bc323e597fdb690d3fc2d064de2
-size 167405651

vae/diffusion_pytorch_model.fp16.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9d8fb415ab4f9782232e7bb82e618e2c0cef0be3593c77a35f5733d8fdd3530f
-size 167335342

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d55443a2d9d4d9decdbe669c51cc6d91eb6a2297477624e2e16a3054f30c2f5a
-size 334643276