khanon commited on
Commit
4d16852
1 Parent(s): fa1113b

adds new Junko LoRA

Browse files
README.md CHANGED
@@ -36,6 +36,11 @@ Here you will find the various LoRAs I've trained, typically of Blue Archive cha
36
  ### Izuna
37
  [Available on old Mega.co.nz repository.](https://mega.nz/folder/SqYwQTRI#GN2SmGTBsV6S4q-L-V4VeA)
38
 
 
 
 
 
 
39
  ### Koharu
40
  [Shimoe Koharu / 下江コハル / 시모에 코하루 / 下江小春](https://huggingface.co/khanon/lora-training/blob/main/koharu/README.md)
41
 
 
36
  ### Izuna
37
  [Available on old Mega.co.nz repository.](https://mega.nz/folder/SqYwQTRI#GN2SmGTBsV6S4q-L-V4VeA)
38
 
39
+ ### Junko
40
+ [Akashi Junko / 赤司ジュンコ / 아카시 준코 / 赤司淳子](https://huggingface.co/khanon/lora-training/blob/main/junko/README.md)
41
+
42
+ [![Junko](junko/chara-junko-v1c.png)](https://huggingface.co/khanon/lora-training/blob/main/junko/README.md)
43
+
44
  ### Koharu
45
  [Shimoe Koharu / 下江コハル / 시모에 코하루 / 下江小春](https://huggingface.co/khanon/lora-training/blob/main/koharu/README.md)
46
 
junko/README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Akashi Junko (Blue Archive)
2
+ 赤司ジュンコ (ブルーアーカイブ) / 아카시 준코 (블루 아카이브) / 赤司淳子 (碧蓝档案)
3
+
4
+ [**Download here.**](https://huggingface.co/khanon/lora-training/blob/main/junko/chara-junko-v1c.safetensors)
5
+
6
+ ## Table of Contents
7
+ - [Preview](#preview)
8
+ - [Usage](#usage)
9
+ - [Training](#training)
10
+ - [Revisions](#revisions)
11
+
12
+ ## Preview
13
+ ![Junko portrait](chara-junko-v1c.png)
14
+ ![Junko preview 1](example-001-v1c.png)
15
+ ![Junko preview 2](example-002-v1c.png)
16
+ ![Junko preview 3](example-003-v1c.png)
17
+
18
+ ## Usage
19
+ Use any or all of the following tags to summon Junko: `junko, slit pupils, demon horns, halo, twintails, hair ribbon, pointy ears, demon wings`
20
+ - You can also use `low wings` if the wings appear too high.
21
+
22
+ For her normal outfit: `military uniform, short sleeves, black shirt, plaid skirt, red necktie, thigh strap, black boots`
23
+
24
+ For her New Year alt: `japanese clothes, yellow kimono, black hakama skirt, black boots, kinchaku`
25
+
26
+ For her normal expression: `closed mouth, smile, :3`
27
+ - You may need to prefix the colon with a backslash character.
28
+
29
+ For her hangry expression: `open mouth, wavy mouth, skin fang, (tearing up, crying with eyes open:0.5)`
30
+ - The AI is very aggressive about drawing tears/crying. You may need to reduce the emphasis.
31
+
32
+ ## Training
33
+ *Exact parameters are provided in the accompanying JSON files.*
34
+ - Trained on a set of 140 images.
35
+ - 131 normal images (9 repeats)
36
+ - 9 "multiple views" images (6 repeats)
37
+ - These were reduced because the AI was generating too many "multiple views" images.
38
+ - 3 batch size, 4 epochs
39
+ - `(131 * 9 + 9 * 6) / 3 * 4` = 1644 steps
40
+ - 0.0727 loss
41
+ - Initially tagged with WD1.4 swin-v2 model. Tags pruned/edited for consistency.
42
+ - `constant_with_warmup` scheduler
43
+ - 1.5e-5 text encoder LR
44
+ - 1.5e-4 unet LR
45
+ - 1e-5 optimizer LR
46
+ - Used network_dimension 128 (same as usual) / network alpha 128 (default)
47
+ - Resized to 32 after training
48
+ - Training resolution 832x832.
49
+ - Trained without VAE.
50
+ - [Dataset can be found on the mega.co.nz repository.](https://mega.nz/folder/SnQDnCRD#ruLvChZGf2vQtWJN87Qs0Q)
51
+
52
+ ## Revisions
53
+ - v1c (2023-02-15)
54
+ - Initial release.
junko/chara-junko-v1c.png ADDED

Git LFS Details

  • SHA256: f90f0ed47f2a6b4eeddfce122bb8ef9ba31ce936f5b7e22b367654bb1a9d3e40
  • Pointer size: 131 Bytes
  • Size of remote file: 442 kB
junko/chara-junko-v1c.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:179e0cf8bdd52ce09a4c71aef584eb8e39926b6732642981fdf358f613eab165
3
+ size 37879476
junko/example-001-v1c.png ADDED

Git LFS Details

  • SHA256: 0ad07e2b7fa00e6b0d93737e72496894f1de0586d23473c4bca3aabd3ab2d09e
  • Pointer size: 132 Bytes
  • Size of remote file: 1.96 MB
junko/example-002-v1.png ADDED

Git LFS Details

  • SHA256: 21c24b784d87c7603cab3b37f10769ea6e90d46dcd4db9bfd451db578488b4fe
  • Pointer size: 132 Bytes
  • Size of remote file: 2.06 MB
junko/example-002-v1c.png ADDED

Git LFS Details

  • SHA256: 730c5198004ee15e1b84457ce7c4f423533aee90f8e1cac10aacbb05491ba420
  • Pointer size: 132 Bytes
  • Size of remote file: 1.77 MB
junko/example-003-v1c.png ADDED

Git LFS Details

  • SHA256: 8b019453e8186cba7d4fc3db55162b5e0454c06a7ed5da1ff72f7c4f12364ac5
  • Pointer size: 132 Bytes
  • Size of remote file: 2 MB
junko/lora_chara_junko_v1_143i9r.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "pretrained_model_name_or_path": "G:/sd/repo/models/Stable-diffusion/nai-animefull-final-pruned.safetensors",
3
+ "v2": false,
4
+ "v_parameterization": false,
5
+ "logging_dir": "",
6
+ "train_data_dir": "G:/sd/training/datasets/junko/dataset",
7
+ "reg_data_dir": "",
8
+ "output_dir": "G:/sd/lora/trained/chara/junko",
9
+ "max_resolution": "832,832",
10
+ "learning_rate": "1e-5",
11
+ "lr_scheduler": "constant_with_warmup",
12
+ "lr_warmup": "5",
13
+ "train_batch_size": 3,
14
+ "epoch": "4",
15
+ "save_every_n_epochs": "",
16
+ "mixed_precision": "fp16",
17
+ "save_precision": "fp16",
18
+ "seed": "31337",
19
+ "num_cpu_threads_per_process": 32,
20
+ "cache_latents": true,
21
+ "caption_extension": ".txt",
22
+ "enable_bucket": true,
23
+ "gradient_checkpointing": false,
24
+ "full_fp16": false,
25
+ "no_token_padding": false,
26
+ "stop_text_encoder_training": 0,
27
+ "use_8bit_adam": true,
28
+ "xformers": true,
29
+ "save_model_as": "safetensors",
30
+ "shuffle_caption": true,
31
+ "save_state": false,
32
+ "resume": "",
33
+ "prior_loss_weight": 1.0,
34
+ "text_encoder_lr": "1.5e-5",
35
+ "unet_lr": "1.5e-4",
36
+ "network_dim": 128,
37
+ "lora_network_weights": "",
38
+ "color_aug": false,
39
+ "flip_aug": false,
40
+ "clip_skip": 2,
41
+ "gradient_accumulation_steps": 1.0,
42
+ "mem_eff_attn": false,
43
+ "output_name": "chara-junko-v1",
44
+ "model_list": "",
45
+ "max_token_length": "150",
46
+ "max_train_epochs": "",
47
+ "max_data_loader_n_workers": "",
48
+ "network_alpha": 128,
49
+ "training_comment": "Character: `junko, slit pupils, demon horns, halo, twintails, hair ribbon, pointy ears, demon wings`\nStandard outfit: `military uniform, black shirt, plaid skirt, red necktie, thigh strap, black boots`\nNew Year alt: `japanese clothes, yellow kimono, black hakama, black boots, kinchaku`\n\nNormal expression: `smile, closed mouth, skin fang, :3, v-shaped eyebrows`\nHangry expression: `tearing up, open mouth, crying with eyes open`\n\n(143 images * 9 repeats) / 3 batch size * 4 epochs = 1716 steps",
50
+ "keep_tokens": 2,
51
+ "lr_scheduler_num_cycles": "",
52
+ "lr_scheduler_power": "",
53
+ "persistent_data_loader_workers": true,
54
+ "bucket_no_upscale": true,
55
+ "random_crop": false,
56
+ "bucket_reso_steps": 64.0,
57
+ "caption_dropout_every_n_epochs": 0.0,
58
+ "caption_dropout_rate": 0
59
+ }
junko/lora_chara_junko_v1c_131i9r-9i6r.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "pretrained_model_name_or_path": "G:/sd/repo/models/Stable-diffusion/nai-animefull-final-pruned.safetensors",
3
+ "v2": false,
4
+ "v_parameterization": false,
5
+ "logging_dir": "",
6
+ "train_data_dir": "G:/sd/training/datasets/junko/dataset",
7
+ "reg_data_dir": "",
8
+ "output_dir": "G:/sd/lora/trained/chara/junko",
9
+ "max_resolution": "768,768",
10
+ "learning_rate": "1e-5",
11
+ "lr_scheduler": "constant_with_warmup",
12
+ "lr_warmup": "5",
13
+ "train_batch_size": 3,
14
+ "epoch": "4",
15
+ "save_every_n_epochs": "",
16
+ "mixed_precision": "fp16",
17
+ "save_precision": "fp16",
18
+ "seed": "31337",
19
+ "num_cpu_threads_per_process": 32,
20
+ "cache_latents": true,
21
+ "caption_extension": ".txt",
22
+ "enable_bucket": true,
23
+ "gradient_checkpointing": false,
24
+ "full_fp16": false,
25
+ "no_token_padding": false,
26
+ "stop_text_encoder_training": 0,
27
+ "use_8bit_adam": true,
28
+ "xformers": true,
29
+ "save_model_as": "safetensors",
30
+ "shuffle_caption": true,
31
+ "save_state": false,
32
+ "resume": "",
33
+ "prior_loss_weight": 1.0,
34
+ "text_encoder_lr": "1.5e-5",
35
+ "unet_lr": "1.5e-4",
36
+ "network_dim": 128,
37
+ "lora_network_weights": "",
38
+ "color_aug": false,
39
+ "flip_aug": false,
40
+ "clip_skip": 2,
41
+ "gradient_accumulation_steps": 1.0,
42
+ "mem_eff_attn": false,
43
+ "output_name": "chara-junko-v1c",
44
+ "model_list": "",
45
+ "max_token_length": "150",
46
+ "max_train_epochs": "",
47
+ "max_data_loader_n_workers": "",
48
+ "network_alpha": 128,
49
+ "training_comment": "Character: `junko, slit pupils, demon horns, halo, twintails, hair ribbon, pointy ears, demon wings`\nStandard outfit: `military uniform, black shirt, plaid skirt, red necktie, thigh strap, black boots`\nNew Year alt: `japanese clothes, yellow kimono, black hakama, black boots, kinchaku`\n\nNormal expression: `smile, closed mouth, skin fang, :3, v-shaped eyebrows`\nHangry expression: `tearing up, open mouth, crying with eyes open`\n\n(131 images * 9 repeats + 9 `multiple_views` images * 6 repeats) / 3 batch size * 4 epochs = 1644 steps",
50
+ "keep_tokens": 2,
51
+ "lr_scheduler_num_cycles": "",
52
+ "lr_scheduler_power": "",
53
+ "persistent_data_loader_workers": true,
54
+ "bucket_no_upscale": true,
55
+ "random_crop": false,
56
+ "bucket_reso_steps": 64.0,
57
+ "caption_dropout_every_n_epochs": 0.0,
58
+ "caption_dropout_rate": 0
59
+ }