Duplicate from gsdf/Replicant

Co-authored-by: gdgsfsfs <gsdf@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,41 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+sample_01.png filter=lfs diff=lfs merge=lfs -text
+sample_02.png filter=lfs diff=lfs merge=lfs -text
+sample_03.png filter=lfs diff=lfs merge=lfs -text
+sample_04.png filter=lfs diff=lfs merge=lfs -text
+sample_05.png filter=lfs diff=lfs merge=lfs -text
+sample_06.png filter=lfs diff=lfs merge=lfs -text
+sample_07.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,59 @@
+---
+license: creativeml-openrail-m
+tags:
+- stable-diffusion
+- stable-diffusion-diffusers
+- text-to-image
+- diffusers
+inference: true
+duplicated_from: gsdf/Replicant
+---
+# Please enable hires. fix when using it. 
+  
+Replicant is built by merging several models with fine-tuning WD1.4 and photorealistic SD2.0 models that works with danbooru tags.I trained 4 models to merge and prepared several LoRa models for tuning.As with SD1.x, merging individually trained models is better quality than training many concepts at once.This model is a workflow test and is not good enough.  WD1.4 seems to vary greatly in quality with/without Hires. fix.In Replicant, the difference in quality is more noticeable because of the detailed drawings.So I recommend enabling Hires.fix for use.
+  
+# Example
+Denoising strength 0.6 is a bit large. I like 0.57 better. 
+The optimal CFG Scale value should also be examined.  
+Hands often multiply. When this happens, increase the value of "extra hands".
+  
+ ![sample1](https://huggingface.co/gsdf/Replicant/resolve/main/sample_01.png)
+((masterpiece, best quality)), 1girl, flower, solo, dress, holding, sky, cloud, hat, outdoors, bangs, bouquet, rose, expressionless, blush, pink hair, flower field, red flower, pink eyes, white dress, looking at viewer, midium hair, holding flower, small breasts, red rose, holding bouquet, sun hat, white headwear, depth of field  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit,(extra arms:1.2), extra hands, fewer digits ,long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 576x384, Denoising strength: 0.6, Hires upscale: 2, Hires upscaler: Latent  
+
+    
+![sample2](https://huggingface.co/gsdf/Replicant/resolve/main/sample_02.png)
+((masterpiece, best quality)), 1girl, skirt, shoes, solo, jacket, holding, alley, sitting, can, sneakers, hood, bag, hoodie, squatting, bangs, shirt, black hair, black skirt, short hair, white jacket, looking away, white footwear, full body, red eyes, long sleeves, open jacket, open clothes, holding can,  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit,(extra arms:1.2), extra legs, extra hands, fewer digits , long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes,drinking  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 576x384, Denoising strength: 0.6, Hires upscale: 2, Hires upscaler: Latent  
+
+    
+![sample3](https://huggingface.co/gsdf/Replicant/resolve/main/sample_03.png)
+((masterpiece, best quality)), 1girl, blood, solo, wings, halo, dress, socks, angel, long hair, shoes, standing, ribbon, long hair, blue eyes, angel wings, blood on clothes, white hair, full body, white wings, black footwear, white dress, feathered wings, white sock, white background, long sleeves, simple background,  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit,(extra arms:1.2), extra legs, extra hands, fewer digits , long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 384x576, Denoising strength: 0.57, Hires upscale: 2, Hires upscaler: Latent  
+
+    
+![sample4](https://huggingface.co/gsdf/Replicant/resolve/main/sample_04.png)
+((masterpiece, best quality)), 1girl, car, solo, shorts, jacket, bangs, sitting, shirt, shoes, hairclip, socks, sneakers, denim, sidelocks, motor vehicle, long hair, ground vehicle,brown hair, looking at viewer, white shirt, black jacket, long sleeves, sports car, vehicle focus, aqua eyes, white socks, blue shorts, open clothes, black footwear, denim shorts, open jacket  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit, (extra arms:1.2), extra hands, fewer digits ,long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 384x576, Denoising strength: 0.6, Hires upscale: 2, Hires upscaler: Latent  
+
+ 
+![sample5](https://huggingface.co/gsdf/Replicant/resolve/main/sample_05.png)
+((masterpiece, best quality)), 1girl, solo, twintails, lollipop, smile, ahoge, hairclip, bow, holding, ribbon, frills, blush, shirt, :d, stuffed toy, pink hair, stuffed animal, red nails, hair ornament, open mouth, looking at viewer, stuffed bunny, nail polish, short sleeves, object hug, puffy sleeves, hair between eyes, upper body, light blue eyes, puffy short sleeves, holding stuffed toy, hair bow, white bow, doll hug, hair ribbon, streaked hair, white shirt  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit, (extra arms:1.2), extra hands, fewer digits ,long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 512x512, Denoising strength: 0.57, Hires upscale: 2, Hires upscaler: Latent  
+
+  
+ ![sample6](https://huggingface.co/gsdf/Replicant/resolve/main/sample_06.png)
+((masterpiece, best quality)), 1girl, solo, tail, barefoot, skirt, sleeping, lying, grass, shirt, outdoors, socks, flower, long hair, on side, animal ears, blonde hair, cat tail, closed eyes, blue skirt, white shirt, cat ears, school uniform, dappled sunlight, short sleeves, bare legs, closed mouth, full body, pleated skirt  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit, (extra arms:1.2), extra hands, fewer digits ,long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 576x384, Denoising strength: 0.6, Hires upscale: 2, Hires upscaler: Latent  
+
+  
+![sample7](https://huggingface.co/gsdf/Replicant/resolve/main/sample_07.png)
+((masterpiece, best quality)), 1girl, car, building, gun, weapon, outdoors, solo, military, day, city, standing, serious, pants, rifle, holding, jacket, motor vehicle, ground vehicle, brown hair, assault rifle, long hair, vehicle focus, holding gun, holding weapon, black footwear, military vehicle, full body, depth of field,  
+Negative prompt: (low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), inaccurate eyes, extra digit, (extra arms:1.2), extra hands, fewer digits ,long body, cropped, jpeg artifacts, signature, watermark, username, blurry, empty eyes  
+Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 10, Size: 576x384, Denoising strength: 0.6, Hires upscale: 2, Hires upscaler: Latent  

Replicant_test01.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87636c68302a3561a02359fc9145640797a02c80e871758815e4df8fb77b7845
+size 5159974609

Replicant_test01.yaml

@@ -0,0 +1,67 @@
+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False # we set this to false because this is an inference only config
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        use_fp16: True
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"