End of training

README.md

@@ -10,11 +10,11 @@ tags:
 - diffusers-training
 - stable-diffusion-xl
 - stable-diffusion-xl-diffusers
-- stable-diffusion-xl
-- stable-diffusion-xl-diffusers
+- stable-diffusion
+- stable-diffusion-diffusers
 - text-to-image
-- diffusers-training
 - diffusers
+- diffusers-training
 inference: true
 ---
 

logs/text2image-fine-tune/1728308018.007925/events.out.tfevents.1728308018.datasec.engr.uconn.edu.5056.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4773490919c0a2aa253454d6511102a44970c5e9dd845a4364fb23609995696f
+size 2645

logs/text2image-fine-tune/1728308018.0096645/hparams.yml

@@ -0,0 +1,55 @@
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+adam_weight_decay: 0.01
+allow_tf32: false
+cache_dir: null
+caption_column: text
+center_crop: true
+checkpointing_steps: 500
+checkpoints_total_limit: null
+dataloader_num_workers: 0
+dataset_config_name: null
+dataset_name: haorandai/Vehicle_Bicycle_Epsilon0.1_20Samples_With20Constraints
+dream_detail_preservation: 1.0
+dream_training: false
+enable_xformers_memory_efficient_attention: false
+foreach_ema: false
+gradient_accumulation_steps: 4
+gradient_checkpointing: true
+hub_model_id: null
+hub_token: null
+image_column: image
+input_perturbation: 0
+learning_rate: 1.0e-05
+local_rank: -1
+logging_dir: logs
+lr_scheduler: constant
+lr_warmup_steps: 500
+max_grad_norm: 1.0
+max_train_samples: null
+max_train_steps: 200
+mixed_precision: fp16
+noise_offset: 0
+non_ema_revision: null
+num_train_epochs: 20
+offload_ema: false
+output_dir: /mnt/nvme1n1p1/home/biw23007/haoran/different_uf_finetune_model_dir_1e-05_200_withconstraint/haorandai/Vehicle_Bicycle_Epsilon0.1_20Samples_With20Constraints/
+prediction_type: null
+pretrained_model_name_or_path: CompVis/stable-diffusion-v1-4
+push_to_hub: true
+random_flip: true
+report_to: tensorboard
+resolution: 224
+resume_from_checkpoint: null
+revision: null
+scale_lr: false
+seed: null
+snr_gamma: null
+tracker_project_name: text2image-fine-tune
+train_batch_size: 1
+train_data_dir: null
+use_8bit_adam: false
+use_ema: true
+validation_epochs: 5
+variant: null

logs/text2image-fine-tune/events.out.tfevents.1728308018.datasec.engr.uconn.edu.5056.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c207d8554e642aeeee5e39e0a2e2826556c5d6d826286e3a1a1ad3f30cf0df
+size 9761

model_index.json

@@ -1,36 +1,32 @@
 {
-  "_class_name": "StableDiffusionXLPipeline",
+  "_class_name": "StableDiffusionPipeline",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
+  "_name_or_path": "CompVis/stable-diffusion-v1-4",
   "feature_extractor": [
-    null,
-    null
+    "transformers",
+    "CLIPImageProcessor"
   ],
-  "force_zeros_for_empty_prompt": true,
   "image_encoder": [
     null,
     null
   ],
+  "requires_safety_checker": true,
+  "safety_checker": [
+    "stable_diffusion",
+    "StableDiffusionSafetyChecker"
+  ],
   "scheduler": [
     "diffusers",
-    "EulerDiscreteScheduler"
+    "PNDMScheduler"
   ],
   "text_encoder": [
     "transformers",
     "CLIPTextModel"
   ],
-  "text_encoder_2": [
-    "transformers",
-    "CLIPTextModelWithProjection"
-  ],
   "tokenizer": [
     "transformers",
     "CLIPTokenizer"
   ],
-  "tokenizer_2": [
-    "transformers",
-    "CLIPTokenizer"
-  ],
   "unet": [
     "diffusers",
     "UNet2DConditionModel"

scheduler/scheduler_config.json

@@ -1,23 +1,15 @@
 {
-  "_class_name": "EulerDiscreteScheduler",
+  "_class_name": "PNDMScheduler",
   "_diffusers_version": "0.31.0.dev0",
   "beta_end": 0.012,
   "beta_schedule": "scaled_linear",
   "beta_start": 0.00085,
   "clip_sample": false,
-  "final_sigmas_type": "zero",
-  "interpolation_type": "linear",
   "num_train_timesteps": 1000,
   "prediction_type": "epsilon",
-  "rescale_betas_zero_snr": false,
-  "sample_max_value": 1.0,
   "set_alpha_to_one": false,
-  "sigma_max": null,
-  "sigma_min": null,
   "skip_prk_steps": true,
   "steps_offset": 1,
   "timestep_spacing": "leading",
-  "timestep_type": "discrete",
-  "trained_betas": null,
-  "use_karras_sigmas": false
+  "trained_betas": null
 }

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/home/biw23007/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/text_encoder",
+  "_name_or_path": "CompVis/stable-diffusion-v1-4",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -18,7 +18,7 @@
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
   "pad_token_id": 1,
-  "projection_dim": 768,
+  "projection_dim": 512,
   "torch_dtype": "float16",
   "transformers_version": "4.44.2",
   "vocab_size": 49408

unet/config.json

@@ -1,20 +1,17 @@
 {
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
+  "_name_or_path": "CompVis/stable-diffusion-v1-4",
   "act_fn": "silu",
-  "addition_embed_type": "text_time",
+  "addition_embed_type": null,
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": 256,
-  "attention_head_dim": [
-    5,
-    10,
-    20
-  ],
+  "addition_time_embed_dim": null,
+  "attention_head_dim": 8,
   "attention_type": "default",
   "block_out_channels": [
     320,
     640,
+    1280,
     1280
   ],
   "center_input_sample": false,
@@ -22,12 +19,13 @@
   "class_embeddings_concat": false,
   "conv_in_kernel": 3,
   "conv_out_kernel": 3,
-  "cross_attention_dim": 2048,
+  "cross_attention_dim": 768,
   "cross_attention_norm": null,
   "down_block_types": [
-    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D"
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
   ],
   "downsample_padding": 1,
   "dropout": 0.0,
@@ -47,27 +45,24 @@
   "num_class_embeds": null,
   "only_cross_attention": false,
   "out_channels": 4,
-  "projection_class_embeddings_input_dim": 2816,
+  "projection_class_embeddings_input_dim": null,
   "resnet_out_scale_factor": 1.0,
   "resnet_skip_time_act": false,
   "resnet_time_scale_shift": "default",
   "reverse_transformer_layers_per_block": null,
-  "sample_size": 128,
+  "sample_size": 64,
   "time_cond_proj_dim": null,
   "time_embedding_act_fn": null,
   "time_embedding_dim": null,
   "time_embedding_type": "positional",
   "timestep_post_act": null,
-  "transformer_layers_per_block": [
-    1,
-    2,
-    10
-  ],
+  "transformer_layers_per_block": 1,
   "up_block_types": [
+    "UpBlock2D",
     "CrossAttnUpBlock2D",
     "CrossAttnUpBlock2D",
-    "UpBlock2D"
+    "CrossAttnUpBlock2D"
   ],
-  "upcast_attention": null,
-  "use_linear_projection": true
+  "upcast_attention": false,
+  "use_linear_projection": false
 }

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44c71a1d806b334fe4d68f0a3ab32d30de0c7adf11dfb40a34b8543f668c4892
+oid sha256:43484359db0988fec15e9d375bce4e8b7e1ee96051ce32dcaedf864493bc24c5
 size 3438167536

vae/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "AutoencoderKL",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
+  "_name_or_path": "CompVis/stable-diffusion-v1-4",
   "act_fn": "silu",
   "block_out_channels": [
     128,
@@ -24,8 +24,8 @@
   "mid_block_add_attention": true,
   "norm_num_groups": 32,
   "out_channels": 3,
-  "sample_size": 1024,
-  "scaling_factor": 0.13025,
+  "sample_size": 512,
+  "scaling_factor": 0.18215,
   "shift_factor": null,
   "up_block_types": [
     "UpDecoderBlock2D",

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcb60880a46b63dea58e9bc591abe15f8350bde47b405f9c38f4be70c6161e68
+oid sha256:4fbcf0ebe55a0984f5a5e00d8c4521d52359af7229bb4d81890039d2aa16dd7c
 size 167335342