ahbpp commited on
Commit
79eb124
·
verified ·
1 Parent(s): 70afd3f

End of training

Browse files
Files changed (45) hide show
  1. README.md +48 -0
  2. checkpoint-1000/model.safetensors +3 -0
  3. checkpoint-1000/optimizer.bin +3 -0
  4. checkpoint-1000/pytorch_lora_weights.safetensors +3 -0
  5. checkpoint-1000/random_states_0.pkl +3 -0
  6. checkpoint-1000/scaler.pt +3 -0
  7. checkpoint-1000/scheduler.bin +3 -0
  8. checkpoint-2000/model.safetensors +3 -0
  9. checkpoint-2000/optimizer.bin +3 -0
  10. checkpoint-2000/pytorch_lora_weights.safetensors +3 -0
  11. checkpoint-2000/random_states_0.pkl +3 -0
  12. checkpoint-2000/scaler.pt +3 -0
  13. checkpoint-2000/scheduler.bin +3 -0
  14. image_0.png +0 -0
  15. image_1.png +0 -0
  16. image_2.png +0 -0
  17. image_3.png +0 -0
  18. logs/text2image-fine-tune-derm/1746347039.4979656/events.out.tfevents.1746347039.instance-20250419-040903.111674.1 +3 -0
  19. logs/text2image-fine-tune-derm/1746347039.4994361/hparams.yml +52 -0
  20. logs/text2image-fine-tune-derm/events.out.tfevents.1746347039.instance-20250419-040903.111674.0 +3 -0
  21. logs/text2image-fine-tune-sdxl/1746384950.9689298/events.out.tfevents.1746384950.instance-20250419-040903.301021.1 +3 -0
  22. logs/text2image-fine-tune-sdxl/1746384950.9704707/hparams.yml +54 -0
  23. logs/text2image-fine-tune-sdxl/events.out.tfevents.1746384950.instance-20250419-040903.301021.0 +3 -0
  24. logs/text2image-fine-tune/1746331545.6744382/events.out.tfevents.1746331545.instance-20250419-040903.43135.1 +3 -0
  25. logs/text2image-fine-tune/1746331545.6759384/hparams.yml +52 -0
  26. logs/text2image-fine-tune/1746346359.7851052/events.out.tfevents.1746346359.instance-20250419-040903.109020.1 +3 -0
  27. logs/text2image-fine-tune/1746346359.786598/hparams.yml +52 -0
  28. logs/text2image-fine-tune/1746380173.2502353/events.out.tfevents.1746380173.instance-20250419-040903.259892.1 +3 -0
  29. logs/text2image-fine-tune/1746380173.2518716/hparams.yml +54 -0
  30. logs/text2image-fine-tune/1746380279.3313491/events.out.tfevents.1746380279.instance-20250419-040903.263699.1 +3 -0
  31. logs/text2image-fine-tune/1746380279.3329923/hparams.yml +54 -0
  32. logs/text2image-fine-tune/1746380623.5504863/events.out.tfevents.1746380623.instance-20250419-040903.265802.1 +3 -0
  33. logs/text2image-fine-tune/1746380623.552036/hparams.yml +54 -0
  34. logs/text2image-fine-tune/1746380716.9023485/events.out.tfevents.1746380716.instance-20250419-040903.266398.1 +3 -0
  35. logs/text2image-fine-tune/1746380716.903856/hparams.yml +54 -0
  36. logs/text2image-fine-tune/1746383221.49296/events.out.tfevents.1746383221.instance-20250419-040903.286415.1 +3 -0
  37. logs/text2image-fine-tune/1746383221.4944603/hparams.yml +54 -0
  38. logs/text2image-fine-tune/events.out.tfevents.1746331545.instance-20250419-040903.43135.0 +3 -0
  39. logs/text2image-fine-tune/events.out.tfevents.1746346359.instance-20250419-040903.109020.0 +3 -0
  40. logs/text2image-fine-tune/events.out.tfevents.1746380173.instance-20250419-040903.259892.0 +3 -0
  41. logs/text2image-fine-tune/events.out.tfevents.1746380279.instance-20250419-040903.263699.0 +3 -0
  42. logs/text2image-fine-tune/events.out.tfevents.1746380623.instance-20250419-040903.265802.0 +3 -0
  43. logs/text2image-fine-tune/events.out.tfevents.1746380716.instance-20250419-040903.266398.0 +3 -0
  44. logs/text2image-fine-tune/events.out.tfevents.1746383221.instance-20250419-040903.286415.0 +3 -0
  45. pytorch_lora_weights.safetensors +3 -0
README.md ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: stabilityai/stable-diffusion-xl-base-1.0
3
+ library_name: diffusers
4
+ license: creativeml-openrail-m
5
+ inference: true
6
+ tags:
7
+ - stable-diffusion-xl
8
+ - stable-diffusion-xl-diffusers
9
+ - text-to-image
10
+ - diffusers
11
+ - diffusers-training
12
+ - lora
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the training script had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+
19
+ # LoRA text2image fine-tuning - ahbpp/sdxl-lora-finetune-ham10000-v2
20
+
21
+ These are LoRA adaption weights for stabilityai/stable-diffusion-xl-base-1.0. The weights were fine-tuned on the BoooomNing/ham10000 dataset. You can find some example images in the following.
22
+
23
+ ![img_0](./image_0.png)
24
+ ![img_1](./image_1.png)
25
+ ![img_2](./image_2.png)
26
+ ![img_3](./image_3.png)
27
+
28
+
29
+ LoRA for the text encoder was enabled: False.
30
+
31
+ Special VAE used for training: None.
32
+
33
+
34
+ ## Intended uses & limitations
35
+
36
+ #### How to use
37
+
38
+ ```python
39
+ # TODO: add an example code snippet for running this diffusion pipeline
40
+ ```
41
+
42
+ #### Limitations and bias
43
+
44
+ [TODO: provide examples of latent issues and potential remediations]
45
+
46
+ ## Training details
47
+
48
+ [TODO: describe the data used to train the model]
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84d8cb35ab7add502e93809b5785039b2a9e800bbd9f1dd96898af6b47892fb4
3
+ size 1722353544
checkpoint-1000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea6f558b75ac568742d9f450466569a83c1f4b0b25ee1135057a1aa14846245b
3
+ size 47393347
checkpoint-1000/pytorch_lora_weights.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:101460a69a04740a2c4ec5d9d945257b56dcc2774829968d2f43cbc9fff8a45a
3
+ size 23390424
checkpoint-1000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ed1e373001af5dbbf2219cf1487d1b06baeb56500ed9e42c6bf764c92db115e
3
+ size 14885
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
3
+ size 1383
checkpoint-1000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:379ec43386a084c335149b37e3948d98a6c19f4b19e372af6fd0e61e21ca8b4d
3
+ size 1401
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c49ac22a89f66d14b4db161d205bf88e040853bd8e36f32d1f1d49e82c3b272b
3
+ size 1722353544
checkpoint-2000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d22d4c0e0b53501f861050d3d3b76407c88673b15b2ec744cc6c0376560410e
3
+ size 6592523
checkpoint-2000/pytorch_lora_weights.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6db8df6afa3346b8c88e4e89ba4353911d7517d5543c6607d241a15c7d18fa98
3
+ size 3226184
checkpoint-2000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65989d11a0e33b9af69595464b916a2638a8965f93dfcee239a0744fd4b1d502
3
+ size 14885
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
3
+ size 1383
checkpoint-2000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7ac7eb719e4a06f6d00c05bb31a0c341feb0b546cd53ba4e41f13f29c41f6ec
3
+ size 1401
image_0.png ADDED
image_1.png ADDED
image_2.png ADDED
image_3.png ADDED
logs/text2image-fine-tune-derm/1746347039.4979656/events.out.tfevents.1746347039.instance-20250419-040903.111674.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37c622d35dbdf66936f37afce8a99e7cc60a679f356e8d08d95acfc3e65fdaee
3
+ size 2508
logs/text2image-fine-tune-derm/1746347039.4994361/hparams.yml ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ enable_xformers_memory_efficient_attention: false
15
+ gradient_accumulation_steps: 1
16
+ gradient_checkpointing: false
17
+ hub_model_id: ham10000-lora
18
+ hub_token: null
19
+ image_column: image
20
+ image_interpolation_mode: lanczos
21
+ learning_rate: 5.0e-05
22
+ local_rank: -1
23
+ logging_dir: logs
24
+ lr_scheduler: cosine
25
+ lr_warmup_steps: 0
26
+ max_grad_norm: 1.0
27
+ max_train_samples: null
28
+ max_train_steps: 2090
29
+ mixed_precision: fp16
30
+ noise_offset: 0
31
+ num_train_epochs: 10
32
+ num_validation_images: 4
33
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
34
+ prediction_type: null
35
+ pretrained_model_name_or_path: stable-diffusion-v1-5/stable-diffusion-v1-5
36
+ push_to_hub: true
37
+ random_flip: true
38
+ rank: 4
39
+ report_to: tensorboard
40
+ resolution: 512
41
+ resume_from_checkpoint: null
42
+ revision: null
43
+ scale_lr: false
44
+ seed: 1444
45
+ snr_gamma: null
46
+ train_batch_size: 48
47
+ train_data_dir: null
48
+ use_8bit_adam: false
49
+ validation_epochs: 1
50
+ validation_prompt: High-resolution dermatoscopic image of Melanoma on the chest of
51
+ a 70-year-old male patient
52
+ variant: null
logs/text2image-fine-tune-derm/events.out.tfevents.1746347039.instance-20250419-040903.111674.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae8b7ecf21f82c8ecb268fc2383c345e3c4e1fd538249ac31fd948d85c6064bc
3
+ size 13022183
logs/text2image-fine-tune-sdxl/1746384950.9689298/events.out.tfevents.1746384950.instance-20250419-040903.301021.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e368d7b9319cd061c4c7e657f9268a0853c66a444cbdb65ec79c358dd1acf2d
3
+ size 2567
logs/text2image-fine-tune-sdxl/1746384950.9704707/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: true
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000-v2
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 1565
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 5
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 32
49
+ train_data_dir: null
50
+ train_text_encoder: false
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune-sdxl/events.out.tfevents.1746384950.instance-20250419-040903.301021.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a5f19be0465ed61748f319b3cd64f303e0ea8ebd31c6f721caaaf964e132f82
3
+ size 150702
logs/text2image-fine-tune/1746331545.6744382/events.out.tfevents.1746331545.instance-20250419-040903.43135.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f3af370d3c53799bc524d1150a492344b6c975fa8745ac0b6511c5398b277d0
3
+ size 2508
logs/text2image-fine-tune/1746331545.6759384/hparams.yml ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ enable_xformers_memory_efficient_attention: false
15
+ gradient_accumulation_steps: 1
16
+ gradient_checkpointing: false
17
+ hub_model_id: ham10000-lora
18
+ hub_token: null
19
+ image_column: image
20
+ image_interpolation_mode: lanczos
21
+ learning_rate: 0.0002
22
+ local_rank: -1
23
+ logging_dir: logs
24
+ lr_scheduler: cosine
25
+ lr_warmup_steps: 0
26
+ max_grad_norm: 1.0
27
+ max_train_samples: null
28
+ max_train_steps: 1045
29
+ mixed_precision: fp16
30
+ noise_offset: 0
31
+ num_train_epochs: 5
32
+ num_validation_images: 4
33
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
34
+ prediction_type: null
35
+ pretrained_model_name_or_path: stable-diffusion-v1-5/stable-diffusion-v1-5
36
+ push_to_hub: true
37
+ random_flip: true
38
+ rank: 4
39
+ report_to: tensorboard
40
+ resolution: 512
41
+ resume_from_checkpoint: null
42
+ revision: null
43
+ scale_lr: false
44
+ seed: 1444
45
+ snr_gamma: null
46
+ train_batch_size: 48
47
+ train_data_dir: null
48
+ use_8bit_adam: false
49
+ validation_epochs: 1
50
+ validation_prompt: High-resolution dermatoscopic image of Melanoma on the chest of
51
+ a 70-year-old male patient
52
+ variant: null
logs/text2image-fine-tune/1746346359.7851052/events.out.tfevents.1746346359.instance-20250419-040903.109020.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8aecd6e611ad93bb635921bad2523f202b499876f967c184e5d29c0ce20822f1
3
+ size 2508
logs/text2image-fine-tune/1746346359.786598/hparams.yml ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ enable_xformers_memory_efficient_attention: false
15
+ gradient_accumulation_steps: 1
16
+ gradient_checkpointing: false
17
+ hub_model_id: ham10000-lora
18
+ hub_token: null
19
+ image_column: image
20
+ image_interpolation_mode: lanczos
21
+ learning_rate: 5.0e-05
22
+ local_rank: -1
23
+ logging_dir: logs
24
+ lr_scheduler: cosine
25
+ lr_warmup_steps: 0
26
+ max_grad_norm: 1.0
27
+ max_train_samples: null
28
+ max_train_steps: 2090
29
+ mixed_precision: fp16
30
+ noise_offset: 0
31
+ num_train_epochs: 10
32
+ num_validation_images: 4
33
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
34
+ prediction_type: null
35
+ pretrained_model_name_or_path: stable-diffusion-v1-5/stable-diffusion-v1-5
36
+ push_to_hub: true
37
+ random_flip: true
38
+ rank: 4
39
+ report_to: tensorboard
40
+ resolution: 512
41
+ resume_from_checkpoint: null
42
+ revision: null
43
+ scale_lr: false
44
+ seed: 1444
45
+ snr_gamma: null
46
+ train_batch_size: 48
47
+ train_data_dir: null
48
+ use_8bit_adam: false
49
+ validation_epochs: 1
50
+ validation_prompt: High-resolution dermatoscopic image of Melanoma on the chest of
51
+ a 70-year-old male patient
52
+ variant: null
logs/text2image-fine-tune/1746380173.2502353/events.out.tfevents.1746380173.instance-20250419-040903.259892.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89fa45587c002605cb61e45059b69100883e2b91bbb763c961816a101799c147
3
+ size 2564
logs/text2image-fine-tune/1746380173.2518716/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 418
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 2
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 48
49
+ train_data_dir: null
50
+ train_text_encoder: false
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune/1746380279.3313491/events.out.tfevents.1746380279.instance-20250419-040903.263699.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:021ded1060dbe096e0661a7a3e05779ee59d5484e431279925547135488e9cca
3
+ size 2564
logs/text2image-fine-tune/1746380279.3329923/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 5008
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 2
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 4
49
+ train_data_dir: null
50
+ train_text_encoder: false
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune/1746380623.5504863/events.out.tfevents.1746380623.instance-20250419-040903.265802.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a1717c1e5da5a72f5bcaae82d40d1d7969859f658765e4d63c7764b7beb2cf
3
+ size 2564
logs/text2image-fine-tune/1746380623.552036/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: true
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 1252
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 2
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 16
49
+ train_data_dir: null
50
+ train_text_encoder: true
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune/1746380716.9023485/events.out.tfevents.1746380716.instance-20250419-040903.266398.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0621b5f250416873b4c95d9e85962d3ec483b1eee9189a4490c7b3cee924cf1e
3
+ size 2564
logs/text2image-fine-tune/1746380716.903856/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: true
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 939
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 3
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 32
49
+ train_data_dir: null
50
+ train_text_encoder: true
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune/1746383221.49296/events.out.tfevents.1746383221.instance-20250419-040903.286415.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30a86ce90ddeca5fbf1e889e83a505a9dd19f7a6219b58d076a025513c4a9eaa
3
+ size 2567
logs/text2image-fine-tune/1746383221.4944603/hparams.yml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: true
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ center_crop: true
9
+ checkpointing_steps: 1000
10
+ checkpoints_total_limit: null
11
+ dataloader_num_workers: 8
12
+ dataset_config_name: null
13
+ dataset_name: BoooomNing/ham10000
14
+ debug_loss: false
15
+ enable_npu_flash_attention: false
16
+ enable_xformers_memory_efficient_attention: false
17
+ gradient_accumulation_steps: 1
18
+ gradient_checkpointing: false
19
+ hub_model_id: sdxl-lora-finetune-ham10000-v2
20
+ hub_token: null
21
+ image_column: image
22
+ learning_rate: 0.0001
23
+ local_rank: -1
24
+ logging_dir: logs
25
+ lr_scheduler: cosine
26
+ lr_warmup_steps: 0
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 3130
30
+ mixed_precision: fp16
31
+ noise_offset: 0
32
+ num_train_epochs: 10
33
+ num_validation_images: 4
34
+ output_dir: /home/jupyter/aleksei_cache_dir/sd_runs/ham10000
35
+ prediction_type: null
36
+ pretrained_model_name_or_path: stabilityai/stable-diffusion-xl-base-1.0
37
+ pretrained_vae_model_name_or_path: null
38
+ push_to_hub: true
39
+ random_flip: true
40
+ rank: 4
41
+ report_to: tensorboard
42
+ resolution: 512
43
+ resume_from_checkpoint: null
44
+ revision: null
45
+ scale_lr: false
46
+ seed: 1444
47
+ snr_gamma: null
48
+ train_batch_size: 32
49
+ train_data_dir: null
50
+ train_text_encoder: false
51
+ use_8bit_adam: false
52
+ validation_epochs: 1
53
+ validation_prompt: A dermatoscopic image of Melanoma
54
+ variant: null
logs/text2image-fine-tune/events.out.tfevents.1746331545.instance-20250419-040903.43135.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1884c75e368b3c359cae075a6764b9af9d1f1df645c5ba1fb7d5e6f3233fd33e
3
+ size 5831978
logs/text2image-fine-tune/events.out.tfevents.1746346359.instance-20250419-040903.109020.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d5a06ddcf9177a4b2daefc4d85d9a35cc98ff952017803ebee958a417d2b88d
3
+ size 1072329
logs/text2image-fine-tune/events.out.tfevents.1746380173.instance-20250419-040903.259892.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:397f708f98dbc62518245ff04d9b13e5875f263b5115deeb0b7cd78491a98447
3
+ size 88
logs/text2image-fine-tune/events.out.tfevents.1746380279.instance-20250419-040903.263699.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:007cb0345257f87be2212bf24de0a6c6e0620445b61b6d2b3fe5c27fb52d3ef3
3
+ size 3592
logs/text2image-fine-tune/events.out.tfevents.1746380623.instance-20250419-040903.265802.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8f76cd69a13cdf23438fe83d16ab6e4dd93d62d1213cf92d4701f6edeb1691f
3
+ size 1096
logs/text2image-fine-tune/events.out.tfevents.1746380716.instance-20250419-040903.266398.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:720a30e6c45022a4a9b04d6e3cc93b5ed7abd80b64f3c73f26a8fabdf484fecd
3
+ size 95340
logs/text2image-fine-tune/events.out.tfevents.1746383221.instance-20250419-040903.286415.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc3c385e6b21ba9146dcafb739bc07807a20bc88b7e9345350fb7fde575220b6
3
+ size 68502
pytorch_lora_weights.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03c5e2795cb941ccd550cc9345bc40b1c8c5c528adca8a901a207d814bc42ed
3
+ size 23390424