Trained for 0 epochs and 6500 steps.

Trained with datasets ['text-embeds-pixart-filter', 'photo-concept-bucket', 'moviecollection', 'experimental', 'ethnic', 'sports', 'architecture', 'shutterstock', 'cinemamix-1mp', 'nsfw-1024', 'anatomy', 'bg20k-1024', 'yoga', 'photo-aesthetics', 'text-1mp', 'movieposters', 'normalnudes', 'pixel-art', 'signs', 'midjourney-v6-520k-raw', 'sfwbooru', 'nijijourney-v6-520k-raw', 'dalle3']
Learning rate 1e-06, batch size 24, and 1 gradient accumulation steps.
Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False
Using 'linspace' timestep spacing.
Base model: ptx0/pixart-900m-1024-ft-large
VAE: madebyollin/sdxl-vae-fp16-fix

.gitattributes

@@ -226,3 +226,4 @@ training_state-photo-concept-bucket.json filter=lfs diff=lfs merge=lfs -text
 assets/image_173_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_190_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_86_2.png filter=lfs diff=lfs merge=lfs -text
+training_state-dalle3.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -77,7 +77,7 @@ You may reuse the base model text encoder for inference.
 ## Training settings
 
 - Training epochs: 0
-- Training steps: 6000
+- Training steps: 6500
 - Learning rate: 1e-06
 - Effective batch size: 192
   - Micro-batch size: 24
@@ -167,7 +167,7 @@ You may reuse the base model text encoder for inference.
 ### anatomy
 - Repeats: 5
 - Total number of images: ~15168
-- Total number of aspect buckets: 3
+- Total number of aspect buckets: 2
 - Resolution: 1.0 megapixels
 - Cropped: True
 - Crop style: random
@@ -247,7 +247,7 @@ You may reuse the base model text encoder for inference.
 ### sfwbooru
 - Repeats: 0
 - Total number of images: ~423552
-- Total number of aspect buckets: 25
+- Total number of aspect buckets: 22
 - Resolution: 1.0 megapixels
 - Cropped: False
 - Crop style: None
@@ -255,7 +255,7 @@ You may reuse the base model text encoder for inference.
 ### nijijourney-v6-520k-raw
 - Repeats: 0
 - Total number of images: ~670976
-- Total number of aspect buckets: 9
+- Total number of aspect buckets: 8
 - Resolution: 1.0 megapixels
 - Cropped: False
 - Crop style: None

optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:628f076f19a3e7d4a4802a5550c04db94e8e406729a29176d4e340749e82039f
+oid sha256:bcff4c17312d745b136c8190f9727d3fc2c2edd25ee230c282d74cec68b3e4e1
 size 5451415117

random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed8c07b13bbaf98c98e6f43a5e88e51be1853d173e9127e879ba372ee4eed54b
+oid sha256:02d96097f72f30354fa85e73245707b2e4359c7a4eb6cee9b2ff3a53a7722036
 size 16036

scheduler.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa90d1596151fc982c9df01c65a1b9accaa596352fb73dd2d76f53176687cc0d
+oid sha256:2857172e458093a6b00728a13ba73b13ad03ca2bba97709c461c686e18e69d52
 size 1000

training_state-photo-concept-bucket.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cfc8fd93b0098850c4aeb5709e40838c5588c8b2e2bf75ccfc0e2552a2af1c01
-size 6317385
+oid sha256:f81049a4be735eff10f1cf3f2c65602186f3f40a32a7744c5b8e5cecb67f1b3c
+size 6454827

training_state.json

@@ -1 +1 @@
-{"global_step": 6000, "epoch_step": 1, "epoch": 1, "exhausted_backends": ["signs", "nijijourney", "pixel-art", "propagandaposters", "sports", "gay", "experimental", "yoga", "ethnic", "cinemamix-1mp", "architecture", "mj-60", "normalnudes", "bookcovers", "celebrities", "movieposters", "shutterstock", "nsfw-1024", "photo-aesthetics", "moviecollection", "bg20k-1024"], "repeats": {"bookcovers": 0, "signs": 0, "normalnudes": 0, "nijijourney": 0, "movieposters": 0, "celebrities": 0, "pixel-art": 0, "propagandaposters": 0, "sports": 0, "moviecollection": 0, "gay": 0, "experimental": 0, "yoga": 0, "ethnic": 0, "cinemamix-1mp": 0, "architecture": 0, "mj-60": 0, "text-1mp": 13, "shutterstock": 0, "nsfw-1024": 0, "photo-aesthetics": 0, "anatomy": 5, "bg20k-1024": 0}}
+{"global_step": 6500, "epoch_step": 1, "epoch": 1, "exhausted_backends": ["signs", "nijijourney", "pixel-art", "propagandaposters", "sports", "gay", "experimental", "yoga", "ethnic", "cinemamix-1mp", "architecture", "mj-60", "normalnudes", "bookcovers", "celebrities", "movieposters", "shutterstock", "nsfw-1024", "photo-aesthetics", "moviecollection", "bg20k-1024"], "repeats": {"bookcovers": 0, "signs": 0, "normalnudes": 0, "nijijourney": 0, "movieposters": 0, "celebrities": 0, "pixel-art": 0, "propagandaposters": 0, "sports": 0, "moviecollection": 0, "gay": 0, "experimental": 0, "yoga": 0, "ethnic": 0, "cinemamix-1mp": 0, "architecture": 0, "mj-60": 0, "text-1mp": 14, "shutterstock": 0, "nsfw-1024": 0, "photo-aesthetics": 0, "anatomy": 5, "bg20k-1024": 0}}

transformer/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adfcfa6afa55315cd426029676bda667d7bf44d2a2ef92eeb58f27437eee187e
+oid sha256:34b5f18a661013502c57468cb3ad67ed59955081630f044bf5dfd6b566007d41
 size 1816969728