Trained for 0 epochs and 2000 steps.

Trained with datasets ['text-embeds-pixart-filter', 'photo-concept-bucket', 'ideogram', '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 'trailing' timestep spacing.
Base model: terminusresearch/pixart-900m-1024-ft-v0.6
VAE: madebyollin/sdxl-vae-fp16-fix

README.md

@@ -47,7 +47,7 @@ You may reuse the base model text encoder for inference.
 ## Training settings
 
 - Training epochs: 0
-- Training steps: 1000
+- Training steps: 2000
 - Learning rate: 1e-06
 - Effective batch size: 192
   - Micro-batch size: 24
@@ -70,6 +70,14 @@ You may reuse the base model text encoder for inference.
 - Cropped: True
 - Crop style: random
 - Crop aspect: square
+### ideogram
+- Repeats: 15
+- Total number of images: ~36096
+- Total number of aspect buckets: 1
+- Resolution: 1.0 megapixels
+- Cropped: True
+- Crop style: random
+- Crop aspect: square
 ### midjourney-v6-520k-raw
 - Repeats: 0
 - Total number of images: ~390912

optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f9f9a3a3f5451c22635b16e8cc7a837edd9ec41c43c63d460e8bb889a7a3472
+oid sha256:dd8f60e8aae208dd92fae9115d03c841fc61241bc8cce99f08e5c10b0cbe1737
 size 5451415117

random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f0edfc2c885f730ef911db10373dce0a3e814e4fdbb2de759c691606ecf21e3
+oid sha256:4cc35cbbfba1d770fe33fbe068175bdf96d2b0925f70d35c452b11e1fe34dee7
 size 16100

scheduler.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:efff19450f55a9358b76f3e5170171942761d1c5b9128683028d0c09b8a24573
+oid sha256:bf997b28d4bfd8d4b58ad8d6128f43c39216511c3e56cc458eab92b813f97246
 size 1000

training_state.json

@@ -1 +1 @@
-{"global_step": 1000, "epoch_step": 1000, "epoch": 1, "exhausted_backends": [], "repeats": {}}
+{"global_step": 2000, "epoch_step": 2000, "epoch": 1, "exhausted_backends": [], "repeats": {}}

transformer/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "PixArtTransformer2DModel",
   "_diffusers_version": "0.30.0.dev0",
-  "_name_or_path": "terminusresearch/pixart-900m-1024-ft-v0.6",
+  "_name_or_path": "/home/ubuntu/training/models/checkpoint-1000",
   "activation_fn": "gelu-approximate",
   "attention_bias": true,
   "attention_head_dim": 72,

transformer/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d877393a75fcacd20413a5c27afdd5bfce4ac8f15411d9236ef4ed7ced00081
+oid sha256:c3919281ef146513cdb0f68a3673f8c0cc69d7e346caae41749fddfab312472b
 size 1816969728