End of training

logs/instruct-pix2pix/1727959192.1853259/events.out.tfevents.1727959192.e248c2e345b9.2582.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a67114b54065fac981d0a38d545643c5e2f037a2914076605ddd2d090f198f76
+size 2437

logs/instruct-pix2pix/1727959192.1870933/hparams.yml

@@ -0,0 +1,51 @@
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+adam_weight_decay: 0.01
+allow_tf32: false
+cache_dir: null
+center_crop: false
+checkpointing_steps: 5000
+checkpoints_total_limit: 1
+conditioning_dropout_prob: 0.05
+dataloader_num_workers: 0
+dataset_config_name: null
+dataset_name: GOFFENEY/nails_diffusion
+edit_prompt_column: instruction
+edited_image_column: ground_truth_image
+enable_xformers_memory_efficient_attention: true
+gradient_accumulation_steps: 10
+gradient_checkpointing: true
+hub_model_id: null
+hub_token: null
+learning_rate: 5.0e-05
+local_rank: -1
+logging_dir: logs
+lr_scheduler: constant
+lr_warmup_steps: 0
+max_grad_norm: 1.0
+max_train_samples: null
+max_train_steps: 15000
+mixed_precision: fp16
+non_ema_revision: null
+num_train_epochs: 7500
+num_validation_images: 4
+original_image_column: input_image
+output_dir: instruct-pix2pix-model
+pretrained_model_name_or_path: stable-diffusion-v1-5/stable-diffusion-v1-5
+push_to_hub: true
+random_flip: true
+report_to: tensorboard
+resolution: 256
+resume_from_checkpoint: null
+revision: null
+scale_lr: false
+seed: 42
+train_batch_size: 1
+train_data_dir: null
+use_8bit_adam: false
+use_ema: false
+val_image_url: null
+validation_epochs: 1
+validation_prompt: null
+variant: null

logs/instruct-pix2pix/1727959728.6075335/events.out.tfevents.1727959728.e248c2e345b9.4941.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b48de8b59f749a2b0cfaea32a09e468e6d297248220362f5b4be4ff4d299af21
+size 2429

logs/instruct-pix2pix/1727959728.6092808/hparams.yml

@@ -0,0 +1,51 @@
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+adam_weight_decay: 0.01
+allow_tf32: false
+cache_dir: null
+center_crop: false
+checkpointing_steps: 5000
+checkpoints_total_limit: 1
+conditioning_dropout_prob: 0.05
+dataloader_num_workers: 0
+dataset_config_name: null
+dataset_name: GOFFENEY/nails_diffusion
+edit_prompt_column: instruction
+edited_image_column: ground_truth_image
+enable_xformers_memory_efficient_attention: true
+gradient_accumulation_steps: 10
+gradient_checkpointing: true
+hub_model_id: null
+hub_token: null
+learning_rate: 5.0e-05
+local_rank: 0
+logging_dir: logs
+lr_scheduler: constant
+lr_warmup_steps: 0
+max_grad_norm: 1.0
+max_train_samples: null
+max_train_steps: 15000
+mixed_precision: fp16
+non_ema_revision: null
+num_train_epochs: 7500
+num_validation_images: 4
+original_image_column: input_image
+output_dir: instruct-pix2pix-model
+pretrained_model_name_or_path: stabilityai/stable-diffusion-2-base
+push_to_hub: true
+random_flip: true
+report_to: tensorboard
+resolution: 256
+resume_from_checkpoint: null
+revision: null
+scale_lr: false
+seed: 42
+train_batch_size: 1
+train_data_dir: null
+use_8bit_adam: false
+use_ema: false
+val_image_url: null
+validation_epochs: 1
+validation_prompt: null
+variant: null

logs/instruct-pix2pix/1727959988.540947/events.out.tfevents.1727959988.e248c2e345b9.6316.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd906a97eb2cb3d04a5f5da647bacacd740e2bb6a32e6042ca67c6ade61acdcf
+size 2429

logs/instruct-pix2pix/1727959988.5426655/hparams.yml

@@ -0,0 +1,51 @@
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+adam_weight_decay: 0.01
+allow_tf32: false
+cache_dir: null
+center_crop: false
+checkpointing_steps: 5000
+checkpoints_total_limit: 1
+conditioning_dropout_prob: 0.05
+dataloader_num_workers: 0
+dataset_config_name: null
+dataset_name: GOFFENEY/nails_diffusion
+edit_prompt_column: instruction
+edited_image_column: ground_truth_image
+enable_xformers_memory_efficient_attention: true
+gradient_accumulation_steps: 10
+gradient_checkpointing: true
+hub_model_id: null
+hub_token: null
+learning_rate: 5.0e-05
+local_rank: -1
+logging_dir: logs
+lr_scheduler: constant
+lr_warmup_steps: 0
+max_grad_norm: 1.0
+max_train_samples: null
+max_train_steps: 1500
+mixed_precision: fp16
+non_ema_revision: null
+num_train_epochs: 750
+num_validation_images: 4
+original_image_column: input_image
+output_dir: instruct-pix2pix-model
+pretrained_model_name_or_path: stabilityai/stable-diffusion-2-base
+push_to_hub: true
+random_flip: true
+report_to: tensorboard
+resolution: 256
+resume_from_checkpoint: null
+revision: null
+scale_lr: false
+seed: 42
+train_batch_size: 1
+train_data_dir: null
+use_8bit_adam: false
+use_ema: false
+val_image_url: null
+validation_epochs: 1
+validation_prompt: null
+variant: null

logs/instruct-pix2pix/1727960234.8151736/events.out.tfevents.1727960234.e248c2e345b9.7393.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bd6a5a8c8e5315690d67d18af2bdaedc038f0f001c5e9931ff1e0d90a21ce73
+size 2429

logs/instruct-pix2pix/1727960234.8170383/hparams.yml

@@ -0,0 +1,51 @@
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+adam_weight_decay: 0.01
+allow_tf32: false
+cache_dir: null
+center_crop: false
+checkpointing_steps: 5000
+checkpoints_total_limit: 1
+conditioning_dropout_prob: 0.05
+dataloader_num_workers: 0
+dataset_config_name: null
+dataset_name: GOFFENEY/nails_diffusion
+edit_prompt_column: instruction
+edited_image_column: ground_truth_image
+enable_xformers_memory_efficient_attention: true
+gradient_accumulation_steps: 10
+gradient_checkpointing: true
+hub_model_id: null
+hub_token: null
+learning_rate: 5.0e-05
+local_rank: -1
+logging_dir: logs
+lr_scheduler: constant
+lr_warmup_steps: 0
+max_grad_norm: 1.0
+max_train_samples: null
+max_train_steps: 1500
+mixed_precision: fp16
+non_ema_revision: null
+num_train_epochs: 750
+num_validation_images: 4
+original_image_column: input_image
+output_dir: instruct-pix2pix-model
+pretrained_model_name_or_path: stabilityai/stable-diffusion-2-base
+push_to_hub: true
+random_flip: false
+report_to: tensorboard
+resolution: 256
+resume_from_checkpoint: null
+revision: null
+scale_lr: false
+seed: 42
+train_batch_size: 1
+train_data_dir: null
+use_8bit_adam: false
+use_ema: false
+val_image_url: null
+validation_epochs: 1
+validation_prompt: null
+variant: null

logs/instruct-pix2pix/events.out.tfevents.1727959192.e248c2e345b9.2582.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:291f8c815653b91e73bbe21dae3ce1052954bb58ce0377b5ef380636e127fff8
+size 3496

logs/instruct-pix2pix/events.out.tfevents.1727959728.e248c2e345b9.4941.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d40992291cb5d6876c289a5925f981cbfdb91f33ced3503f962ea4ebb326d2e6
+size 1432

logs/instruct-pix2pix/events.out.tfevents.1727959988.e248c2e345b9.6316.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c14c0ec53f8e34bf5baf52065a368a79ba537e16610fdf73ad810da207f2e998
+size 4072

logs/instruct-pix2pix/events.out.tfevents.1727960234.e248c2e345b9.7393.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba0d14380351b42dd868905a067eb651d7e34fe577ec05127a3fed36e083859c
+size 73461

model_index.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "StableDiffusionInstructPix2PixPipeline",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stable-diffusion-v1-5/stable-diffusion-v1-5",
+  "_name_or_path": "stabilityai/stable-diffusion-2-base",
   "feature_extractor": [
     "transformers",
     "CLIPImageProcessor"
@@ -10,10 +10,10 @@
     null,
     null
   ],
-  "requires_safety_checker": true,
+  "requires_safety_checker": false,
   "safety_checker": [
-    "stable_diffusion",
-    "StableDiffusionSafetyChecker"
+    null,
+    null
   ],
   "scheduler": [
     "diffusers",

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "stable-diffusion-v1-5/stable-diffusion-v1-5",
+  "_name_or_path": "stabilityai/stable-diffusion-2-base",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -7,18 +7,18 @@
   "bos_token_id": 0,
   "dropout": 0.0,
   "eos_token_id": 2,
-  "hidden_act": "quick_gelu",
-  "hidden_size": 768,
+  "hidden_act": "gelu",
+  "hidden_size": 1024,
   "initializer_factor": 1.0,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 77,
   "model_type": "clip_text_model",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 23,
   "pad_token_id": 1,
-  "projection_dim": 768,
+  "projection_dim": 512,
   "torch_dtype": "float16",
   "transformers_version": "4.44.2",
   "vocab_size": 49408

text_encoder/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:660c6f5b1abae9dc498ac2d21e1347d2abdb0cf6c0c0c8576cd796491d9a6cdd
-size 246144152
+oid sha256:bc1827c465450322616f06dea41596eac7d493f4e95904dcb51f0fc745c4e13f
+size 680820392

tokenizer/special_tokens_map.json

@@ -13,7 +13,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<|endoftext|>",
+  "pad_token": "!",
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer/tokenizer_config.json

@@ -1,6 +1,14 @@
 {
   "add_prefix_space": false,
   "added_tokens_decoder": {
+    "0": {
+      "content": "!",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
     "49406": {
       "content": "<|startoftext|>",
       "lstrip": false,
@@ -24,7 +32,7 @@
   "eos_token": "<|endoftext|>",
   "errors": "replace",
   "model_max_length": 77,
-  "pad_token": "<|endoftext|>",
+  "pad_token": "!",
   "tokenizer_class": "CLIPTokenizer",
   "unk_token": "<|endoftext|>"
 }

unet/config.json

@@ -1,12 +1,17 @@
 {
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stable-diffusion-v1-5/stable-diffusion-v1-5",
+  "_name_or_path": "stabilityai/stable-diffusion-2-base",
   "act_fn": "silu",
   "addition_embed_type": null,
   "addition_embed_type_num_heads": 64,
   "addition_time_embed_dim": null,
-  "attention_head_dim": 8,
+  "attention_head_dim": [
+    5,
+    10,
+    20,
+    20
+  ],
   "attention_type": "default",
   "block_out_channels": [
     320,
@@ -19,7 +24,7 @@
   "class_embeddings_concat": false,
   "conv_in_kernel": 3,
   "conv_out_kernel": 3,
-  "cross_attention_dim": 768,
+  "cross_attention_dim": 1024,
   "cross_attention_norm": null,
   "down_block_types": [
     "CrossAttnDownBlock2D",
@@ -64,5 +69,5 @@
     "CrossAttnUpBlock2D"
   ],
   "upcast_attention": false,
-  "use_linear_projection": false
+  "use_linear_projection": true
 }

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ca0cc6cc56ca13760312fa6ca4ead60493084d62205b8b1bd82f7bab647014e
-size 3438213624
+oid sha256:4dcad5896bdc58dd4d4dde49eb414f3445ee0222599ded5f5434f24a809abac1
+size 3463772592

vae/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "AutoencoderKL",
   "_diffusers_version": "0.31.0.dev0",
-  "_name_or_path": "stable-diffusion-v1-5/stable-diffusion-v1-5",
+  "_name_or_path": "stabilityai/stable-diffusion-2-base",
   "act_fn": "silu",
   "block_out_channels": [
     128,

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fbcf0ebe55a0984f5a5e00d8c4521d52359af7229bb4d81890039d2aa16dd7c
+oid sha256:3e4c08995484ee61270175e9e7a072b66a6e4eeb5f0c266667fe1f45b90daf9a
 size 167335342