knock4869
/

base

Model card Files Files and versions Community

knock4869 commited on May 11, 2023

Commit

9f13d60

1 Parent(s): 7af2a15

synchronized 2023-05-07

Browse files

This view is limited to 50 files because it contains too many changes. See raw diff

Files changed (50) hide show

checkpoints-configs/animefull-final-pruned.yaml +70 -0
checkpoints-configs/animefull-latest.yaml +69 -0
checkpoints-configs/animefull-prevgood.yaml +69 -0
checkpoints-configs/animesfw-final-pruned.yaml +70 -0
checkpoints-configs/animesfw-latest.yaml +69 -0
checkpoints-configs/animesfw-prevgood.yaml +69 -0
checkpoints-configs/wd-1-4-anime_e1.yaml +67 -0
checkpoints-configs/wd-1-5-beta2-aesthetic-fp16.yaml +68 -0
checkpoints-configs/wd-1-5-beta2-aesthetic-fp32.yaml +68 -0
checkpoints-configs/wd-1-5-beta2-fp16.yaml +68 -0
checkpoints-configs/wd-1-5-beta2-fp32.yaml +68 -0
checkpoints-unclip/illuminatiDiffusionV1_v11-unclip-h-fp16.safetensors +3 -0
checkpoints-unclip/sd21-unclip-h.ckpt +3 -0
checkpoints-unclip/sd21-unclip-l.ckpt +3 -0
checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-h-fp16.safetensors +3 -0
checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-h-fp32.safetensors +3 -0
checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-l-fp16.safetensors +3 -0
checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-l-fp32.safetensors +3 -0
checkpoints-unclip/wd-1-5-beta2-unclip-h-fp16.safetensors +3 -0
checkpoints-unclip/wd-1-5-beta2-unclip-h-fp32.safetensors +3 -0
checkpoints/512-depth-ema.ckpt +3 -0
checkpoints/512-inpainting-ema.ckpt +3 -0
checkpoints/anime700k-64bs-0.1ucg-penultimate-1epoch-clip-ema-continue-76000-ema-pruned.ckpt +3 -0
checkpoints/anime700k-64bs-0.1ucg-penultimate-1epoch-clip-ema-continue-76000.pt +3 -0
checkpoints/animefull-final-pruned.ckpt +3 -0
checkpoints/animefull-latest.ckpt +3 -0
checkpoints/animefull-prevgood.ckpt +3 -0
checkpoints/animesfw-final-pruned.ckpt +3 -0
checkpoints/animesfw-latest.ckpt +3 -0
checkpoints/animesfw-prevgood.ckpt +3 -0
checkpoints/sd-v1-5-inpainting.ckpt +3 -0
checkpoints/v1-5-pruned-emaonly.ckpt +3 -0
checkpoints/v1-5-pruned.ckpt +3 -0
checkpoints/v2-1_512-ema-pruned.ckpt +3 -0
checkpoints/v2-1_512-nonema-pruned.ckpt +3 -0
checkpoints/v2-1_768-ema-pruned.ckpt +3 -0
checkpoints/v2-1_768-nonema-pruned.ckpt +3 -0
checkpoints/wd-1-5-beta2-aesthetic-fp16.safetensors +3 -0
checkpoints/wd-1-5-beta2-aesthetic-fp32.safetensors +3 -0
checkpoints/wd-1-5-beta2-fp16.safetensors +3 -0
checkpoints/wd-1-5-beta2-fp32.safetensors +3 -0
checkpoints/wd15-beta2-frosty-aesthetic-fp32.safetensors +3 -0
clip_vision/clip-vit-base-16.bin +3 -0
clip_vision/clip-vit-base-32.bin +3 -0
clip_vision/clip-vit-large-14-336.bin +3 -0
clip_vision/clip-vit-large-14.bin +3 -0
controlnet-annotator/150_16_swin_l_oneformer_coco_100ep.pth +3 -0
controlnet-annotator/250_16_swin_l_oneformer_ade20k_160k.pth +3 -0
controlnet-annotator/ControlNetHED.pth +3 -0
controlnet-annotator/RealESRGAN_x4plus.pth +3 -0

checkpoints-configs/animefull-final-pruned.yaml ADDED Viewed

	@@ -0,0 +1,70 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/animefull-latest.yaml ADDED Viewed

	@@ -0,0 +1,69 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/animefull-prevgood.yaml ADDED Viewed

	@@ -0,0 +1,69 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/animesfw-final-pruned.yaml ADDED Viewed

	@@ -0,0 +1,70 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/animesfw-latest.yaml ADDED Viewed

	@@ -0,0 +1,69 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/animesfw-prevgood.yaml ADDED Viewed

	@@ -0,0 +1,69 @@

+model:
+  base_learning_rate: 1.0e-04
+  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   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        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_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 512
+          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.FrozenCLIPEmbedder

checkpoints-configs/wd-1-4-anime_e1.yaml ADDED Viewed

	@@ -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"

checkpoints-configs/wd-1-5-beta2-aesthetic-fp16.yaml ADDED Viewed

	@@ -0,0 +1,68 @@

+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    parameterization: "v"
+    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"

checkpoints-configs/wd-1-5-beta2-aesthetic-fp32.yaml ADDED Viewed

	@@ -0,0 +1,68 @@

+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    parameterization: "v"
+    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"

checkpoints-configs/wd-1-5-beta2-fp16.yaml ADDED Viewed

	@@ -0,0 +1,68 @@

+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    parameterization: "v"
+    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"

checkpoints-configs/wd-1-5-beta2-fp32.yaml ADDED Viewed

	@@ -0,0 +1,68 @@

+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    parameterization: "v"
+    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"

checkpoints-unclip/illuminatiDiffusionV1_v11-unclip-h-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f7590644685d2557cb7a34eaa90bb0597d3afa3ed2fb1ca60062a38465c64b0a
+size 3983581602

checkpoints-unclip/sd21-unclip-h.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4124cbcf397ace8c536908f5f877a1f091a41168d1d3d9fd0c84484c1653693b
+size 7967306143

checkpoints-unclip/sd21-unclip-l.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ed12098385701c1850defd3d2bc20905c7c64dad3ccad895a7f13b406aab7602
+size 6161550731

checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-h-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:062603a01ed43b081df0c1b99f82449ba97283135810cca5fae204a32970e027
+size 3983581602

checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-h-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8d22d4afe7c2c406359edf20a2970321ee30d1f0bf2b61d5c04d87de05b6526c
+size 7966941354

checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-l-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:cc1189bc5ffa0f4d7aeb5d17ac672a7f3b85e89b838e2bfa4e12426ee373f5fa
+size 3469908836

checkpoints-unclip/wd-1-5-beta2-aesthetic-unclip-l-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8ce0dede15124052f53422c144a256b3abfc2ef7a4b7918069a0525a516d7033
+size 6161176892

checkpoints-unclip/wd-1-5-beta2-unclip-h-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:73720150fa784a777343df3f49ca7d18b481d2ec042982eb59e9a8eaaaebb987
+size 3983581602

checkpoints-unclip/wd-1-5-beta2-unclip-h-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:efbd5dc1abf48593ddd902aaa7c3dbc2bf31c9ac419b13066398be3aebcd1d41
+size 7966941354

checkpoints/512-depth-ema.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ece79d931a1f3c27fb98a0c0bc6aa7d61168879cd52b4ad7b26e3106f9748ef1
+size 5707592867

checkpoints/512-inpainting-ema.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:2a208a7ded5d42dcb0c0ec908b23c631002091e06afe7e76d16cd11079f8d4e3
+size 5214921607

checkpoints/anime700k-64bs-0.1ucg-penultimate-1epoch-clip-ema-continue-76000-ema-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:939269be42067dc822b206b2fada2947a5c20e896059036705b56d68236bd7ef
+size 2132882528

checkpoints/anime700k-64bs-0.1ucg-penultimate-1epoch-clip-ema-continue-76000.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:828a21b769b3a32b1508f3b47828fc9d7c27c14a0cab64bc42fb785cd24211a6
+size 7703828146

checkpoints/animefull-final-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:89d59c3dde4c56c6d5c41da34cc55ce479d93b4007046980934b14db71bdb2a8
+size 4265377440

checkpoints/animefull-latest.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a7529df02340e5b4c3870c894c1ae84f22ea7b37fd0633e5bacfad9618228032
+size 7703828146

checkpoints/animefull-prevgood.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f60e935a6266ba9c28f6853938c11f5a068374bb96e43093cb2cb790298614ed
+size 7703828146

checkpoints/animesfw-final-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:22fa233c2dfd7748d534be603345cb9abf994a23244dfdfc1013f4f90322feca
+size 4265377440

checkpoints/animesfw-latest.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:80df01df1bab8952d1c8842d718fc78a575a81c9968d685ca34d725ce0b4c3c5
+size 7703828146

checkpoints/animesfw-prevgood.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e7a10a3217247ca65c9d9d072bdb7e8e6869b6c846527812283c4e423e2f9ac1
+size 7703828146

checkpoints/sd-v1-5-inpainting.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c6bbc15e3224e6973459ba78de4998b80b50112b0ae5b5c67113d56b4e366b19
+size 4265437280

checkpoints/v1-5-pruned-emaonly.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:cc6cb27103417325ff94f52b7a5d2dde45a7515b25c255d8e396c90014281516
+size 4265380512

checkpoints/v1-5-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e1441589a6f3c5a53f5f54d0975a18a7feb7cdf0b0dee276dfc3331ae376a053
+size 7703807346

checkpoints/v2-1_512-ema-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:88ecb782561455673c4b78d05093494b9c539fc6bfc08f3a9a4a0dd7b0b10f36
+size 5214865159

checkpoints/v2-1_512-nonema-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:34dccb9e60d83a4ef963f51e9ff63548323a70024f8679095956d7c403b3ef4a
+size 5214864585

checkpoints/v2-1_768-ema-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ad2a33c361c1f593c4a1fb32ea81afce2b5bb7d1983c6b94793a26a3b54b08a0
+size 5214865159

checkpoints/v2-1_768-nonema-pruned.ckpt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4711ff4dd270067f71d5ea0725e83e4dc7c41f586a588531ee1c3bd549e9119d
+size 5214864585

checkpoints/wd-1-5-beta2-aesthetic-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:96fc8b5de4ea310ac42680dc122485ba3a3cb029913052c9d09a92fea877ce57
+size 2580068692

checkpoints/wd-1-5-beta2-aesthetic-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:52fbdb851b247eec5d9f6e7dab5d3cef28dea28d6d3e99821800b3581faef84b
+size 5159974886

checkpoints/wd-1-5-beta2-fp16.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4bf1855f4d12824b96d9350ad325ac709a488a1b6fb417f0f6f87f85f4b5e1ee
+size 2580068692

checkpoints/wd-1-5-beta2-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:764f93581d80b46011039bb388e899f17f7869fce7e7928b060e9a5574bd8f84
+size 5159974886

checkpoints/wd15-beta2-frosty-aesthetic-fp32.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca00ea7cb1b4d12ab25ffbe16a4c8abb65334ffc196e641f1ea849582d7c622
+size 5159974886

clip_vision/clip-vit-base-16.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ec89c7b09c749a60aae3c9cd910516f24b58214a7df060b48962d14c469cfbf0
+size 598641023

clip_vision/clip-vit-base-32.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a63082132ba4f97a80bea76823f544493bffa8082296d62d71581a4feff1576f
+size 605247071

clip_vision/clip-vit-large-14-336.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c6032c2e0caae3dc2d4fba35535fa6307dbb49df59c7e182b1bc4b3329b81801
+size 1711974081

clip_vision/clip-vit-large-14.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f1a17cdbe0f36fec524f5cafb1c261ea3bbbc13e346e0f74fc9eb0460dedd0d3
+size 1710671599

controlnet-annotator/150_16_swin_l_oneformer_coco_100ep.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0684dfc39720c772b81d43639c3ae1896b5c15aa9ee9a76f4c593b19dfa33855
+size 949602739

controlnet-annotator/250_16_swin_l_oneformer_ade20k_160k.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f7ac095c28ddea4715e854a587eaee24327c624cbbdb17095bc9903c51930b16
+size 949729587

controlnet-annotator/ControlNetHED.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca93762ffd68a29fee1af9d495bf6aab80ae86f08905fb35472a083a4c7a8fa
+size 29444406

controlnet-annotator/RealESRGAN_x4plus.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4fa0d38905f75ac06eb49a7951b426670021be3018265fd191d2125df9d682f1
+size 67040989