Upload folder using huggingface_hub (#1)

- 581b72b24425d2df7e32f91bf48800162bf91521a861ddbfa2df2d81bbdc5aa8 (23f09c93f09c39da8a91ecd941ad294ed3a30097)
- 30987f24140cba6faa309d9dbd91071d198e6c018e47ed48653f599317348c4b (c017f2901ae977dd0a4b098752183c2bba0fadab)
- cb968c552ed8186acd379e191576e8c82530fcad8f8ced56561caacc364abfa2 (05228789cf4b6d9339ac54d9c90b8293bc0c4e83)
- 88b91466e8d86e5c39d2850cfee8022c91ab9763ea75e7d5cf5f8c00dfc4dd53 (8e0b27d672443875cef4b85b0e12256cf62527e6)
- a9f9de5ee24967b9fe63d2db7bd05d6cdba96673d3355d231fbb5d539af60e06 (ab21145637e0f216cf057a2ee417e95dd70222c0)
- bc52680ad2acf357b036e503500464b8b13c4b4c60c14f2eff643e95737b3466 (38c45151725c5e0a6d99e7fe0115a0ac9ad376a4)
- aad62f1db1cdf5ff8b789bae2e85edf78e97081a49b1754e8f04c30399fb0bbb (6df0ecddd50bdaa1cb856768c08289fd0710391a)

exsswimsuit-06-21_23-26-35-save-420-30-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61c8239e05c077afd1702e37a0a376f497909c7c116760d3ada74d548ebee7be
+size 6938084280

exsswimsuit-06-21_23-26-35-save-420-30-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine

exsswimsuit-06-21_23-37-39-save-840-60-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e32c98811820b3ee5161ee88ebdfe32fb72822089614e1299659f4ec2b00e526
+size 6938084280

exsswimsuit-06-21_23-37-39-save-840-60-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine

exsswimsuit-06-21_23-48-42-save-1260-90-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38dd2c1ed1f2b8378d0a74018c76fb6d11c5dbaf2affadae77a9583ff1785f94
+size 6938084280

exsswimsuit-06-21_23-48-42-save-1260-90-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine

exsswimsuit-06-21_23-59-43-save-1680-120-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8187b2746617d0a4ac59b2ab2d425c1c9c9a5dec66b368782a655d6bd5e5ded0
+size 6938084280

exsswimsuit-06-21_23-59-43-save-1680-120-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine

exsswimsuit-06-22_00-10-40-save-2100-150-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:634785ab3dc1ec7baab100bfc81e91f48cae5d10e15907d7f583d4f0a08506d4
+size 6938084280

exsswimsuit-06-22_00-10-40-save-2100-150-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine

exsswimsuit-06-22_00-21-52-save-2520-180-0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb6d608dd5be29690ab4b6c92a45d4ca6d9cfcff654422177ad4281d32c111f2
+size 6938084280

exsswimsuit-06-22_00-21-52-save-2520-180-0.yaml

@@ -0,0 +1,100 @@
+model:
+  params:
+    conditioner_config:
+      params:
+        emb_models:
+        - input_key: txt
+          is_trainable: false
+          params:
+            layer: hidden
+            layer_idx: 11
+          target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
+        - input_key: txt
+          is_trainable: false
+          params:
+            always_return_pooled: true
+            arch: ViT-bigG-14
+            freeze: true
+            layer: penultimate
+            legacy: false
+            version: laion2b_s39b_b160k
+          target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
+        - input_key: original_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: crop_coords_top_left
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+        - input_key: target_size_as_tuple
+          is_trainable: false
+          params:
+            outdim: 256
+          target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
+      target: sgm.modules.GeneralConditioner
+    denoiser_config:
+      params:
+        discretization_config:
+          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
+        num_idx: 1000
+        scaling_config:
+          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
+        weighting_config:
+          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
+      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser
+    disable_first_stage_autocast: true
+    first_stage_config:
+      params:
+        ddconfig:
+          attn_resolutions: []
+          attn_type: vanilla-xformers
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          double_z: true
+          dropout: 0.0
+          in_channels: 3
+          num_res_blocks: 2
+          out_ch: 3
+          resolution: 256
+          z_channels: 4
+        embed_dim: 4
+        lossconfig:
+          target: torch.nn.Identity
+        monitor: val/rec_loss
+      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
+    network_config:
+      params:
+        adm_in_channels: 2816
+        attention_resolutions:
+        - 4
+        - 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        context_dim: 2048
+        in_channels: 4
+        legacy: false
+        model_channels: 320
+        num_classes: sequential
+        num_head_channels: 64
+        num_res_blocks: 2
+        out_channels: 4
+        spatial_transformer_attn_type: softmax-xformers
+        transformer_depth:
+        - 1
+        - 2
+        - 10
+        use_checkpoint: true
+        use_linear_in_transformer: true
+        use_spatial_transformer: true
+      target: sgm.modules.diffusionmodules.openaimodel.UNetModel
+    scale_factor: 0.13025
+  target: sgm.models.diffusion.DiffusionEngine