200 steps trained @ 150 batch size on 1080p and portrait images

model_index.json

@@ -23,8 +23,8 @@
     "CLIPTokenizer"
   ],
   "unet": [
-    "diffusers",
-    "UNet2DConditionModel"
+    "torch",
+    "OptimizedModule"
   ],
   "vae": [
     "diffusers",

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/notebooks/datasets/models/checkpoint-17700/text_encoder",
+  "_name_or_path": "/notebooks/datasets/models/checkpoint-200/text_encoder",
   "architectures": [
     "CLIPTextModel"
   ],

text_encoder/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90ec9eca4b371f76e15544909ba2c0cd19c8a67532d7461d4c7e3a1536af1dd1
+oid sha256:012bcb2fc34c59837cf9f13d28da60f45ddebb1ee0f0455e08d04a4735016045
 size 1361597016

unet/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.17.0.dev0",
-  "_name_or_path": "/root/.cache/huggingface/hub/models--ptx0--pseudo-real/snapshots/f0feef8d707854092d85c0eb2f88cf813e4b5958/unet",
+  "_name_or_path": "/notebooks/datasets/models/checkpoint-200/unet",
   "act_fn": "silu",
   "addition_embed_type": null,
   "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:d31d02d2b42cdc5038b1f56d34eba6718487f30f411e3b8040edeaf3dd6a58fd
+oid sha256:2c926ff1a9b73131b39d1f5bf07178c7305ca788858d58d6bdebcbb2dc977d68
 size 3463726504

vae/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "AutoencoderKL",
   "_diffusers_version": "0.17.0.dev0",
-  "_name_or_path": "/root/.cache/huggingface/hub/models--ptx0--pseudo-real/snapshots/f0feef8d707854092d85c0eb2f88cf813e4b5958/vae",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--ptx0--pseudo-real-beta/snapshots/774dcaa1ca29956cd673b7bd6e220a40e860a419/vae",
   "act_fn": "silu",
   "block_out_channels": [
     128,