haorandai commited on
Commit
b0a1ac5
1 Parent(s): 18d8f5c

End of training

Browse files
README.md CHANGED
@@ -8,6 +8,11 @@ tags:
8
  - text-to-image
9
  - diffusers-training
10
  - diffusers
 
 
 
 
 
11
  inference: true
12
  ---
13
 
 
8
  - text-to-image
9
  - diffusers-training
10
  - diffusers
11
+ - stable-diffusion
12
+ - stable-diffusion-diffusers
13
+ - text-to-image
14
+ - diffusers
15
+ - diffusers-training
16
  inference: true
17
  ---
18
 
feature_extractor/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
logs/text2image-fine-tune/1727395001.1099522/events.out.tfevents.1727395001.datasec.engr.uconn.edu.1408068.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdd1c4ebd0b5f6767ccee858937f90836a77342dedf5bada68ece5e3a9707131
3
+ size 2638
logs/text2image-fine-tune/1727395001.1112392/hparams.yml ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ adam_beta1: 0.9
2
+ adam_beta2: 0.999
3
+ adam_epsilon: 1.0e-08
4
+ adam_weight_decay: 0.01
5
+ allow_tf32: false
6
+ cache_dir: null
7
+ caption_column: text
8
+ center_crop: true
9
+ checkpointing_steps: 500
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 0
12
+ dataset_config_name: null
13
+ dataset_name: haorandai/Orange_Vehicle_Bicycle_lr0.01_e0.02_5_with5constraints
14
+ dream_detail_preservation: 1.0
15
+ dream_training: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ foreach_ema: false
18
+ gradient_accumulation_steps: 4
19
+ gradient_checkpointing: true
20
+ hub_model_id: null
21
+ hub_token: null
22
+ image_column: image
23
+ input_perturbation: 0
24
+ learning_rate: 1.0e-05
25
+ local_rank: -1
26
+ logging_dir: logs
27
+ lr_scheduler: constant
28
+ lr_warmup_steps: 500
29
+ max_grad_norm: 1.0
30
+ max_train_samples: null
31
+ max_train_steps: 200
32
+ mixed_precision: fp16
33
+ noise_offset: 0
34
+ non_ema_revision: null
35
+ num_train_epochs: 67
36
+ offload_ema: false
37
+ output_dir: /mnt/nvme1n1p1/home/biw23007/haoran/finetune_model_color_dir_1e-05_200_withconstraint/haorandai/Orange_Vehicle_Bicycle_lr0.01_e0.02_5_with5constraints/
38
+ prediction_type: null
39
+ pretrained_model_name_or_path: CompVis/stable-diffusion-v1-4
40
+ push_to_hub: true
41
+ random_flip: true
42
+ report_to: tensorboard
43
+ resolution: 512
44
+ resume_from_checkpoint: null
45
+ revision: null
46
+ scale_lr: false
47
+ seed: null
48
+ snr_gamma: null
49
+ tracker_project_name: text2image-fine-tune
50
+ train_batch_size: 1
51
+ train_data_dir: null
52
+ use_8bit_adam: false
53
+ use_ema: true
54
+ validation_epochs: 5
55
+ variant: null
logs/text2image-fine-tune/1727416138.8914287/events.out.tfevents.1727416138.datasec.engr.uconn.edu.1454094.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c45610891b388e98c218b848a8d7deb6e678968634d56f7f6abffd4962f3887b
3
+ size 2638
logs/text2image-fine-tune/1727416138.8923857/hparams.yml ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ adam_beta1: 0.9
2
+ adam_beta2: 0.999
3
+ adam_epsilon: 1.0e-08
4
+ adam_weight_decay: 0.01
5
+ allow_tf32: false
6
+ cache_dir: null
7
+ caption_column: text
8
+ center_crop: true
9
+ checkpointing_steps: 500
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 0
12
+ dataset_config_name: null
13
+ dataset_name: haorandai/Orange_Vehicle_Bicycle_lr0.01_e0.02_5_with5constraints
14
+ dream_detail_preservation: 1.0
15
+ dream_training: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ foreach_ema: false
18
+ gradient_accumulation_steps: 4
19
+ gradient_checkpointing: true
20
+ hub_model_id: null
21
+ hub_token: null
22
+ image_column: image
23
+ input_perturbation: 0
24
+ learning_rate: 1.0e-05
25
+ local_rank: -1
26
+ logging_dir: logs
27
+ lr_scheduler: constant
28
+ lr_warmup_steps: 500
29
+ max_grad_norm: 1.0
30
+ max_train_samples: null
31
+ max_train_steps: 200
32
+ mixed_precision: fp16
33
+ noise_offset: 0
34
+ non_ema_revision: null
35
+ num_train_epochs: 67
36
+ offload_ema: false
37
+ output_dir: /mnt/nvme1n1p1/home/biw23007/haoran/finetune_model_color_dir_1e-05_200_withconstraint/haorandai/Orange_Vehicle_Bicycle_lr0.01_e0.02_5_with5constraints/
38
+ prediction_type: null
39
+ pretrained_model_name_or_path: CompVis/stable-diffusion-v1-4
40
+ push_to_hub: true
41
+ random_flip: true
42
+ report_to: tensorboard
43
+ resolution: 512
44
+ resume_from_checkpoint: null
45
+ revision: null
46
+ scale_lr: false
47
+ seed: null
48
+ snr_gamma: null
49
+ tracker_project_name: text2image-fine-tune
50
+ train_batch_size: 1
51
+ train_data_dir: null
52
+ use_8bit_adam: false
53
+ use_ema: true
54
+ validation_epochs: 5
55
+ variant: null
logs/text2image-fine-tune/events.out.tfevents.1727395001.datasec.engr.uconn.edu.1408068.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13ab60a448eb2e7b3a8f8ba8702a13ee2f77b13529e52dbe669bedd48bcac7bf
3
+ size 9761
logs/text2image-fine-tune/events.out.tfevents.1727416138.datasec.engr.uconn.edu.1454094.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:191478f317122e42dcc5fe0f9785001f8835aaa7304e3f45ea07a3df73682d12
3
+ size 9761
model_index.json CHANGED
@@ -1,36 +1,32 @@
1
  {
2
- "_class_name": "StableDiffusionXLPipeline",
3
  "_diffusers_version": "0.31.0.dev0",
4
- "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
5
  "feature_extractor": [
6
- null,
7
- null
8
  ],
9
- "force_zeros_for_empty_prompt": true,
10
  "image_encoder": [
11
  null,
12
  null
13
  ],
 
 
 
 
 
14
  "scheduler": [
15
  "diffusers",
16
- "EulerDiscreteScheduler"
17
  ],
18
  "text_encoder": [
19
  "transformers",
20
  "CLIPTextModel"
21
  ],
22
- "text_encoder_2": [
23
- "transformers",
24
- "CLIPTextModelWithProjection"
25
- ],
26
  "tokenizer": [
27
  "transformers",
28
  "CLIPTokenizer"
29
  ],
30
- "tokenizer_2": [
31
- "transformers",
32
- "CLIPTokenizer"
33
- ],
34
  "unet": [
35
  "diffusers",
36
  "UNet2DConditionModel"
 
1
  {
2
+ "_class_name": "StableDiffusionPipeline",
3
  "_diffusers_version": "0.31.0.dev0",
4
+ "_name_or_path": "CompVis/stable-diffusion-v1-4",
5
  "feature_extractor": [
6
+ "transformers",
7
+ "CLIPImageProcessor"
8
  ],
 
9
  "image_encoder": [
10
  null,
11
  null
12
  ],
13
+ "requires_safety_checker": true,
14
+ "safety_checker": [
15
+ "stable_diffusion",
16
+ "StableDiffusionSafetyChecker"
17
+ ],
18
  "scheduler": [
19
  "diffusers",
20
+ "PNDMScheduler"
21
  ],
22
  "text_encoder": [
23
  "transformers",
24
  "CLIPTextModel"
25
  ],
 
 
 
 
26
  "tokenizer": [
27
  "transformers",
28
  "CLIPTokenizer"
29
  ],
 
 
 
 
30
  "unet": [
31
  "diffusers",
32
  "UNet2DConditionModel"
safety_checker/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/biw23007/.cache/huggingface/hub/models--CompVis--stable-diffusion-v1-4/snapshots/133a221b8aa7292a167afc5127cb63fb5005638b/safety_checker",
3
+ "architectures": [
4
+ "StableDiffusionSafetyChecker"
5
+ ],
6
+ "initializer_factor": 1.0,
7
+ "logit_scale_init_value": 2.6592,
8
+ "model_type": "clip",
9
+ "projection_dim": 768,
10
+ "text_config": {
11
+ "dropout": 0.0,
12
+ "hidden_size": 768,
13
+ "intermediate_size": 3072,
14
+ "model_type": "clip_text_model",
15
+ "num_attention_heads": 12
16
+ },
17
+ "torch_dtype": "float32",
18
+ "transformers_version": "4.44.2",
19
+ "vision_config": {
20
+ "dropout": 0.0,
21
+ "hidden_size": 1024,
22
+ "intermediate_size": 4096,
23
+ "model_type": "clip_vision_model",
24
+ "num_attention_heads": 16,
25
+ "num_hidden_layers": 24,
26
+ "patch_size": 14
27
+ }
28
+ }
safety_checker/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb351a5ded815c3ff744968ad9c6b218d071b9d313d04f35e813b84b4c0ffde8
3
+ size 1215979664
scheduler/scheduler_config.json CHANGED
@@ -1,23 +1,15 @@
1
  {
2
- "_class_name": "EulerDiscreteScheduler",
3
  "_diffusers_version": "0.31.0.dev0",
4
  "beta_end": 0.012,
5
  "beta_schedule": "scaled_linear",
6
  "beta_start": 0.00085,
7
  "clip_sample": false,
8
- "final_sigmas_type": "zero",
9
- "interpolation_type": "linear",
10
  "num_train_timesteps": 1000,
11
  "prediction_type": "epsilon",
12
- "rescale_betas_zero_snr": false,
13
- "sample_max_value": 1.0,
14
  "set_alpha_to_one": false,
15
- "sigma_max": null,
16
- "sigma_min": null,
17
  "skip_prk_steps": true,
18
  "steps_offset": 1,
19
  "timestep_spacing": "leading",
20
- "timestep_type": "discrete",
21
- "trained_betas": null,
22
- "use_karras_sigmas": false
23
  }
 
1
  {
2
+ "_class_name": "PNDMScheduler",
3
  "_diffusers_version": "0.31.0.dev0",
4
  "beta_end": 0.012,
5
  "beta_schedule": "scaled_linear",
6
  "beta_start": 0.00085,
7
  "clip_sample": false,
 
 
8
  "num_train_timesteps": 1000,
9
  "prediction_type": "epsilon",
 
 
10
  "set_alpha_to_one": false,
 
 
11
  "skip_prk_steps": true,
12
  "steps_offset": 1,
13
  "timestep_spacing": "leading",
14
+ "trained_betas": null
 
 
15
  }
text_encoder/config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/home/biw23007/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/text_encoder",
3
  "architectures": [
4
  "CLIPTextModel"
5
  ],
@@ -18,7 +18,7 @@
18
  "num_attention_heads": 12,
19
  "num_hidden_layers": 12,
20
  "pad_token_id": 1,
21
- "projection_dim": 768,
22
  "torch_dtype": "float16",
23
  "transformers_version": "4.44.2",
24
  "vocab_size": 49408
 
1
  {
2
+ "_name_or_path": "CompVis/stable-diffusion-v1-4",
3
  "architectures": [
4
  "CLIPTextModel"
5
  ],
 
18
  "num_attention_heads": 12,
19
  "num_hidden_layers": 12,
20
  "pad_token_id": 1,
21
+ "projection_dim": 512,
22
  "torch_dtype": "float16",
23
  "transformers_version": "4.44.2",
24
  "vocab_size": 49408
unet/config.json CHANGED
@@ -1,20 +1,17 @@
1
  {
2
  "_class_name": "UNet2DConditionModel",
3
  "_diffusers_version": "0.31.0.dev0",
4
- "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
5
  "act_fn": "silu",
6
- "addition_embed_type": "text_time",
7
  "addition_embed_type_num_heads": 64,
8
- "addition_time_embed_dim": 256,
9
- "attention_head_dim": [
10
- 5,
11
- 10,
12
- 20
13
- ],
14
  "attention_type": "default",
15
  "block_out_channels": [
16
  320,
17
  640,
 
18
  1280
19
  ],
20
  "center_input_sample": false,
@@ -22,12 +19,13 @@
22
  "class_embeddings_concat": false,
23
  "conv_in_kernel": 3,
24
  "conv_out_kernel": 3,
25
- "cross_attention_dim": 2048,
26
  "cross_attention_norm": null,
27
  "down_block_types": [
28
- "DownBlock2D",
29
  "CrossAttnDownBlock2D",
30
- "CrossAttnDownBlock2D"
 
 
31
  ],
32
  "downsample_padding": 1,
33
  "dropout": 0.0,
@@ -47,27 +45,24 @@
47
  "num_class_embeds": null,
48
  "only_cross_attention": false,
49
  "out_channels": 4,
50
- "projection_class_embeddings_input_dim": 2816,
51
  "resnet_out_scale_factor": 1.0,
52
  "resnet_skip_time_act": false,
53
  "resnet_time_scale_shift": "default",
54
  "reverse_transformer_layers_per_block": null,
55
- "sample_size": 128,
56
  "time_cond_proj_dim": null,
57
  "time_embedding_act_fn": null,
58
  "time_embedding_dim": null,
59
  "time_embedding_type": "positional",
60
  "timestep_post_act": null,
61
- "transformer_layers_per_block": [
62
- 1,
63
- 2,
64
- 10
65
- ],
66
  "up_block_types": [
 
67
  "CrossAttnUpBlock2D",
68
  "CrossAttnUpBlock2D",
69
- "UpBlock2D"
70
  ],
71
- "upcast_attention": null,
72
- "use_linear_projection": true
73
  }
 
1
  {
2
  "_class_name": "UNet2DConditionModel",
3
  "_diffusers_version": "0.31.0.dev0",
4
+ "_name_or_path": "CompVis/stable-diffusion-v1-4",
5
  "act_fn": "silu",
6
+ "addition_embed_type": null,
7
  "addition_embed_type_num_heads": 64,
8
+ "addition_time_embed_dim": null,
9
+ "attention_head_dim": 8,
 
 
 
 
10
  "attention_type": "default",
11
  "block_out_channels": [
12
  320,
13
  640,
14
+ 1280,
15
  1280
16
  ],
17
  "center_input_sample": false,
 
19
  "class_embeddings_concat": false,
20
  "conv_in_kernel": 3,
21
  "conv_out_kernel": 3,
22
+ "cross_attention_dim": 768,
23
  "cross_attention_norm": null,
24
  "down_block_types": [
 
25
  "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "DownBlock2D"
29
  ],
30
  "downsample_padding": 1,
31
  "dropout": 0.0,
 
45
  "num_class_embeds": null,
46
  "only_cross_attention": false,
47
  "out_channels": 4,
48
+ "projection_class_embeddings_input_dim": null,
49
  "resnet_out_scale_factor": 1.0,
50
  "resnet_skip_time_act": false,
51
  "resnet_time_scale_shift": "default",
52
  "reverse_transformer_layers_per_block": null,
53
+ "sample_size": 64,
54
  "time_cond_proj_dim": null,
55
  "time_embedding_act_fn": null,
56
  "time_embedding_dim": null,
57
  "time_embedding_type": "positional",
58
  "timestep_post_act": null,
59
+ "transformer_layers_per_block": 1,
 
 
 
 
60
  "up_block_types": [
61
+ "UpBlock2D",
62
  "CrossAttnUpBlock2D",
63
  "CrossAttnUpBlock2D",
64
+ "CrossAttnUpBlock2D"
65
  ],
66
+ "upcast_attention": false,
67
+ "use_linear_projection": false
68
  }
unet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0db57cf437926706215a44c67d3fef8d3feb4b63c63f70a5af29abab54ee2631
3
+ size 3438167536
vae/config.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.31.0.dev0",
4
- "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
5
  "act_fn": "silu",
6
  "block_out_channels": [
7
  128,
@@ -24,8 +24,8 @@
24
  "mid_block_add_attention": true,
25
  "norm_num_groups": 32,
26
  "out_channels": 3,
27
- "sample_size": 1024,
28
- "scaling_factor": 0.13025,
29
  "shift_factor": null,
30
  "up_block_types": [
31
  "UpDecoderBlock2D",
 
1
  {
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.31.0.dev0",
4
+ "_name_or_path": "CompVis/stable-diffusion-v1-4",
5
  "act_fn": "silu",
6
  "block_out_channels": [
7
  128,
 
24
  "mid_block_add_attention": true,
25
  "norm_num_groups": 32,
26
  "out_channels": 3,
27
+ "sample_size": 512,
28
+ "scaling_factor": 0.18215,
29
  "shift_factor": null,
30
  "up_block_types": [
31
  "UpDecoderBlock2D",
vae/diffusion_pytorch_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bcb60880a46b63dea58e9bc591abe15f8350bde47b405f9c38f4be70c6161e68
3
  size 167335342
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fbcf0ebe55a0984f5a5e00d8c4521d52359af7229bb4d81890039d2aa16dd7c
3
  size 167335342