Trained for 0 epochs and 5600 steps.

Trained with datasets ['text-embeds-sdxl', 'photocb-clip-embeds', 'celebrities', 'movieposters', 'normalnudes', 'propagandaposters', 'guys', 'pixel-art', 'signs', 'moviecollection', 'bookcovers', 'nijijourney', 'experimental', 'ethnic', 'sports', 'gay', 'architecture', 'shutterstock', 'cinemamix-1mp', 'nsfw-1024', 'anatomy', 'bg20k-1024', 'yoga', 'photo-aesthetics', 'text-1mp', 'photo-concept-bucket']
Learning rate 1e-06, batch size 8, and 4 gradient accumulation steps.
Used DDPM noise scheduler for training with v_prediction prediction type and rescaled_betas_zero_snr=True
Using 'trailing' timestep spacing.
Base model: ptx0/terminus-xl-velocity-v1
VAE: madebyollin/sdxl-vae-fp16-fix

README.md

@@ -44,7 +44,7 @@ You may reuse the base model text encoder for inference.
 ## Training settings
 
 - Training epochs: 0
-- Training steps: 5500
+- Training steps: 5600
 - Learning rate: 1e-06
 - Effective batch size: 32
   - Micro-batch size: 8
@@ -67,7 +67,7 @@ You may reuse the base model text encoder for inference.
 - Crop style: random
 - Crop aspect: random
 ### movieposters
-- Repeats: 5
+- Repeats: 25
 - Total number of images: 1728
 - Total number of aspect buckets: 3
 - Resolution: 1.0 megapixels
@@ -107,7 +107,7 @@ You may reuse the base model text encoder for inference.
 - Crop style: random
 - Crop aspect: random
 ### signs
-- Repeats: 5
+- Repeats: 25
 - Total number of images: 352
 - Total number of aspect buckets: 3
 - Resolution: 1.0 megapixels
@@ -213,7 +213,7 @@ You may reuse the base model text encoder for inference.
 ### bg20k-1024
 - Repeats: 0
 - Total number of images: 89250
-- Total number of aspect buckets: 2
+- Total number of aspect buckets: 3
 - Resolution: 1.0 megapixels
 - Cropped: True
 - Crop style: random

optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:400ae98197b63d1e231d68434967a989d6d26f523e58ffc47f40178410d81d04
+oid sha256:39aac7d3ef92c6c46baa04d9c50dbcd04f96fd15bb210fd5c0e9adf8a3f598b0
 size 15406336826

random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e814f76188faeaa10fb633fa5d29e0ecf1a55e43150ca2ddeb469f5b6b45630
-size 14408
+oid sha256:1bdafbf7bec3f162698104cd2d02020cd492ce580bf7e3d197dc5b98d0e74800
+size 14280

scheduler.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bc146e60a062a6786bb3606b7e6b2452ab72964e6a44acff076d8b2d7cd2a19
+oid sha256:b6813e50b115c99d61ffcf61ce231df3f75282c5a4ea0ce645961bd0ef935048
 size 1000

training_state-bg20k-1024.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2568ea7b7106c16029870672f1498279218d409dc03c398c2fc6039c1bbf7e1c
-size 16046983
+oid sha256:91ec1218433718ad538871edf3103079633bf2c65e11a0c0ff0c22caa96dd196
+size 16135315

training_state-photo-concept-bucket.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83a8cdd237f1e9980f06c80c2cc6986088b61d2fdda36a6f1cea93c66c8e9090
-size 15449978
+oid sha256:ec5a0c8b7efa20fc10ac0c09cbfa7752e3b395f3405ff5a7474474308eac8f3a
+size 15466180

training_state.json

@@ -1 +1 @@
-{"global_step": 5500, "epoch_step": 166, "epoch": 1, "exhausted_backends": ["guys", "signs", "nijijourney", "propagandaposters", "bookcovers", "pixel-art", "normalnudes", "celebrities", "sports", "movieposters", "moviecollection", "gay", "ethnic", "experimental", "yoga", "architecture", "cinemamix-1mp"], "repeats": {"guys": 0, "signs": 0, "nijijourney": 0, "propagandaposters": 0, "bookcovers": 0, "pixel-art": 0, "normalnudes": 0, "celebrities": 0, "sports": 0, "movieposters": 0, "moviecollection": 0, "gay": 0, "ethnic": 0, "experimental": 0, "yoga": 0, "architecture": 0, "cinemamix-1mp": 0}}
+{"global_step": 5600, "epoch_step": 141, "epoch": 1, "exhausted_backends": ["guys", "signs", "nijijourney", "propagandaposters", "bookcovers", "pixel-art", "normalnudes", "celebrities", "sports", "movieposters", "moviecollection", "gay", "ethnic", "experimental", "yoga", "architecture", "cinemamix-1mp"], "repeats": {"guys": 0, "signs": 0, "nijijourney": 0, "propagandaposters": 0, "bookcovers": 0, "pixel-art": 0, "normalnudes": 0, "celebrities": 0, "sports": 0, "movieposters": 0, "moviecollection": 0, "gay": 0, "ethnic": 0, "experimental": 0, "yoga": 0, "architecture": 0, "cinemamix-1mp": 0}}

unet/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.27.2",
-  "_name_or_path": "/notebooks/datasets/models/checkpoint-5400",
+  "_name_or_path": "/notebooks/datasets/models/checkpoint-5500",
   "act_fn": "silu",
   "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8acc6fbba00bc0971dfdb3bae5de90add1e4907c9057182ad2e54c9123195f3
+oid sha256:f18f1033e3bf14393b5d6f0ce51bb4a6eda072ca906902beeff95fb3f50985fb
 size 5135151440