Trained for 1 epochs and 1500 steps.

Trained with datasets ['text-embeds-pixart-nofilter', 'photo-concept-bucket']
Learning rate 1e-06, batch size 32, and 2 gradient accumulation steps.
Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False
Using 'trailing' timestep spacing.
Base model: PixArt-alpha/PixArt-Sigma-XL-2-1024-MS
VAE: madebyollin/sdxl-vae-fp16-fix

README.md

@@ -1005,8 +1005,8 @@ You may reuse the base model text encoder for inference.
 
 ## Training settings
 
-- Training epochs: 0
-- Training steps: 1000
+- Training epochs: 1
+- Training steps: 1500
 - Learning rate: 1e-06
 - Effective batch size: 512
   - Micro-batch size: 32

optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02fa923032d676df03eb42a629fb0ecf89af1b6fe23a2196535b113578ee42dc
+oid sha256:3947b35c7d6d913df5b1f8c9f1a9caaa3fa2b0ade38c9a713243df4f1f1d8faf
 size 3665677155

random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6fb16d8520f0b105d7d5a67a095b509755464afb3ae42da3a73e2dfcbc96c3e5
+oid sha256:8e7c018d11efa4f2032d7af2fb39f9c1f434fccb3aa9f0b8d9ed8b21ff861acd
 size 16100

scheduler.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0d51a7f7b0addc7f7990d433ab69df71d0909e735408de9c655863c506f4b88
+oid sha256:1c82973153dea5cf9a471058fa34947c8e4abee4df03cb78fc273cd87224bacd
 size 1000

training_state.json

@@ -1 +1 @@
-{"global_step": 1000, "epoch_step": 1000, "epoch": 1, "exhausted_backends": [], "repeats": {}}
+{"global_step": 1500, "epoch_step": 408, "epoch": 2, "exhausted_backends": [], "repeats": {"photo-concept-bucket": 0}}

transformer/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2fc6dbac615318a314888e6f2e6ade28d52fb1fb8c8233773d22465628e6d19
+oid sha256:34eb047cb2af055a214e8c5252e250829eb961411be0e6d6beb7ccf3a966079f
 size 1221780352