File size: 16,120 Bytes
4479d6a
 
e1d8f96
 
 
 
 
 
 
4479d6a
ea857b2
e1d8f96
c3aee77
ea857b2
e1d8f96
 
 
 
 
ea857b2
e1d8f96
 
 
 
 
 
 
 
 
 
c3aee77
 
 
e1d8f96
 
 
 
 
 
 
 
 
 
 
c3aee77
e1d8f96
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c3aee77
 
e1d8f96
 
 
 
 
 
 
 
c3aee77
 
e1d8f96
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c3aee77
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
---
license: creativeml-openrail-m
language:
- en
tags:
- di.ffusion.ai
- stable-diffusion
- LyCORIS
- LoRA
---

# Model Card for di.FFUSION.ai Text Encoder - SD 2.1 LyCORIS
![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/zcw9AUCSbanb61xe6pIUc.png)


<!-- Provide a quick summary of what the model is/does. [Optional] -->
di.FFUSION.ai-tXe-FXAA
Trained on &#34;121361&#34; images.

- **DOWNLOAD:** https://huggingface.co/FFusion/FFUSION.ai-Text-Encoder-LyCORIS-SD-2.1/blob/main/di.FFUSION.ai-tXe-FXAA.safetensors

Enhance your model&#39;s quality and sharpness using your own pre-trained Unet.


The text encoder (without UNET) is wrapped in LyCORIS. Optimizer: torch.optim.adamw.AdamW(weight_decay=0.01, betas=(0.9, 0.99))

Network dimension/rank: 768.0 Alpha: 768.0 Module: lycoris.kohya {&#39;conv_dim&#39;: &#39;256&#39;, &#39;conv_alpha&#39;: &#39;256&#39;, &#39;algo&#39;: &#39;loha&#39;}

Large size due to Lyco CONV 256


![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/Ig1IOYZAyUrhpWIhdC6U-.png)
![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/66eAHPc501sbQx35-B0Oo.png)

This is a heavy experimental version we used to test even with sloppy captions (quick WD tags and terrible clip), yet the results were satisfying.

Note: This is not the text encoder used in the official FFUSION AI model.


# SAMPLES

**Available also at https://civitai.com/models/83622**


![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/agjJ--YR_k_Pbn8tOMsqr.png)





For a1111
Install https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris

Download di.FFUSION.ai-tXe-FXAA to /models/Lycoris

Option1:

Insert <lyco:di.FFUSION.ai-tXe-FXAA:1.0> to prompt
No need to split Unet and Text Enc as its only TX encoder there.

You can go up to 2x weights

Option2: If you need it always ON (ex run a batch from txt file) then you can go to settings / Quicksettings list


![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/N6M4-9eIkvi3nn3koh1fA.png)


add sd_lyco



restart and you should have a drop-down now 🤟 🥃


![image.png](https://cdn-uploads.huggingface.co/production/uploads/6380cf05f496d57325c12194/e8ROXaN8jIaT9lu7tNRjD.png)


#  Table of Contents

- [Model Card for di.FFUSION.ai Text Encoder - SD 2.1 LyCORIS](#model-card-for--model_id-)
- [Table of Contents](#table-of-contents)
- [Table of Contents](#table-of-contents-1)
- [Model Details](#model-details)
  - [Model Description](#model-description)
- [Uses](#uses)
  - [Direct Use](#direct-use)
  - [Downstream Use [Optional]](#downstream-use-optional)
  - [Out-of-Scope Use](#out-of-scope-use)
- [Bias, Risks, and Limitations](#bias-risks-and-limitations)
  - [Recommendations](#recommendations)
- [Training Details](#training-details)
  - [Training Data](#training-data)
  - [Training Procedure](#training-procedure)
    - [Preprocessing](#preprocessing)
    - [Speeds, Sizes, Times](#speeds-sizes-times)
- [Evaluation](#evaluation)
  - [Testing Data, Factors & Metrics](#testing-data-factors--metrics)
    - [Testing Data](#testing-data)
    - [Factors](#factors)
    - [Metrics](#metrics)
  - [Results](#results)
- [Model Examination](#model-examination)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications [optional]](#technical-specifications-optional)
  - [Model Architecture and Objective](#model-architecture-and-objective)
  - [Compute Infrastructure](#compute-infrastructure)
    - [Hardware](#hardware)
    - [Software](#software)
- [Citation](#citation)
- [Glossary [optional]](#glossary-optional)
- [More Information [optional]](#more-information-optional)
- [Model Card Authors [optional]](#model-card-authors-optional)
- [Model Card Contact](#model-card-contact)
- [How to Get Started with the Model](#how-to-get-started-with-the-model)


# Model Details

## Model Description

<!-- Provide a longer summary of what this model is/does. -->
di.FFUSION.ai-tXe-FXAA
Trained on &#34;121361&#34; images.


Enhance your model&#39;s quality and sharpness using your own pre-trained Unet.

The text encoder (without UNET) is wrapped in LyCORIS. Optimizer: torch.optim.adamw.AdamW(weight_decay=0.01, betas=(0.9, 0.99))

Network dimension/rank: 768.0 Alpha: 768.0 Module: lycoris.kohya {&#39;conv_dim&#39;: &#39;256&#39;, &#39;conv_alpha&#39;: &#39;256&#39;, &#39;algo&#39;: &#39;loha&#39;}

Large size due to Lyco CONV 256

This is a heavy experimental version we used to test even with sloppy captions (quick WD tags and terrible clip), yet the results were satisfying.

Note: This is not the text encoder used in the official FFUSION AI model.

- **Developed by:** FFusion.ai
- **Shared by [Optional]:** idle stoev
- **Model type:** Language model
- **Language(s) (NLP):** en
- **License:** creativeml-openrail-m
- **Parent Model:** More information needed
- **Resources for more information:** More information needed



# Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

## Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
<!-- If the user enters content, print that. If not, but they enter a task in the list, use that. If neither, say "more info needed." -->

The text encoder (without UNET) is wrapped in LyCORIS. Optimizer: torch.optim.adamw.AdamW(weight_decay=0.01, betas=(0.9, 0.99))

Network dimension/rank: 768.0 Alpha: 768.0 Module: lycoris.kohya {&#39;conv_dim&#39;: &#39;256&#39;, &#39;conv_alpha&#39;: &#39;256&#39;, &#39;algo&#39;: &#39;loha&#39;}

Large size due to Lyco CONV 256


# Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.


## Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->





# Training Details

## Training Data

<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

Trained on &#34;121361&#34; images.

ss_caption_tag_dropout_rate:  &#34;0.0&#34;,
ss_multires_noise_discount:  &#34;0.3&#34;,
ss_mixed_precision:  &#34;bf16&#34;,
ss_text_encoder_lr:  &#34;1e-07&#34;,
ss_keep_tokens:  &#34;3&#34;,
ss_network_args:  &#34;{&#34;conv_dim&#34;: &#34;256&#34;, &#34;conv_alpha&#34;: &#34;256&#34;, &#34;algo&#34;: &#34;loha&#34;}&#34;,
ss_caption_dropout_rate:  &#34;0.02&#34;,
ss_flip_aug:  &#34;False&#34;,
ss_learning_rate:  &#34;2e-07&#34;,
ss_sd_model_name:  &#34;stabilityai/stable-diffusion-2-1-base&#34;,
ss_max_grad_norm:  &#34;1.0&#34;,
ss_num_epochs:  &#34;2&#34;,
ss_gradient_checkpointing:  &#34;False&#34;,
ss_face_crop_aug_range:  &#34;None&#34;,
ss_epoch:  &#34;2&#34;,
ss_num_train_images:  &#34;121361&#34;,
ss_color_aug:  &#34;False&#34;,
ss_gradient_accumulation_steps:  &#34;1&#34;,
ss_total_batch_size:  &#34;100&#34;,
ss_prior_loss_weight:  &#34;1.0&#34;,
ss_training_comment:  &#34;None&#34;,
ss_network_dim:  &#34;768&#34;,
ss_output_name:  &#34;FusionaMEGA1tX&#34;,
ss_max_bucket_reso:  &#34;1024&#34;,
ss_network_alpha:  &#34;768.0&#34;,
ss_steps:  &#34;2444&#34;,
ss_shuffle_caption:  &#34;True&#34;,
ss_training_finished_at:  &#34;1684158038.0763328&#34;,
ss_min_bucket_reso:  &#34;256&#34;,
ss_noise_offset:  &#34;0.09&#34;,
ss_enable_bucket:  &#34;True&#34;,
ss_batch_size_per_device:  &#34;20&#34;,
ss_max_train_steps:  &#34;2444&#34;,
ss_network_module:  &#34;lycoris.kohya&#34;,


## Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

### Preprocessing

&#34;{&#34;buckets&#34;: {&#34;0&#34;: {&#34;resolution&#34;: [192, 256], &#34;count&#34;: 1}, &#34;1&#34;: {&#34;resolution&#34;: [192, 320], &#34;count&#34;: 1}, &#34;2&#34;: {&#34;resolution&#34;: [256, 384], &#34;count&#34;: 1}, &#34;3&#34;: {&#34;resolution&#34;: [256, 512], &#34;count&#34;: 1}, &#34;4&#34;: {&#34;resolution&#34;: [384, 576], &#34;count&#34;: 2}, &#34;5&#34;: {&#34;resolution&#34;: [384, 640], &#34;count&#34;: 2}, &#34;6&#34;: {&#34;resolution&#34;: [384, 704], &#34;count&#34;: 1}, &#34;7&#34;: {&#34;resolution&#34;: [384, 1088], &#34;count&#34;: 15}, &#34;8&#34;: {&#34;resolution&#34;: [448, 448], &#34;count&#34;: 5}, &#34;9&#34;: {&#34;resolution&#34;: [448, 576], &#34;count&#34;: 1}, &#34;10&#34;: {&#34;resolution&#34;: [448, 640], &#34;count&#34;: 1}, &#34;11&#34;: {&#34;resolution&#34;: [448, 768], &#34;count&#34;: 1}, &#34;12&#34;: {&#34;resolution&#34;: [448, 832], &#34;count&#34;: 1}, &#34;13&#34;: {&#34;resolution&#34;: [448, 1088], &#34;count&#34;: 25}, &#34;14&#34;: {&#34;resolution&#34;: [448, 1216], &#34;count&#34;: 1}, &#34;15&#34;: {&#34;resolution&#34;: [512, 640], &#34;count&#34;: 2}, &#34;16&#34;: {&#34;resolution&#34;: [512, 768], &#34;count&#34;: 10}, &#34;17&#34;: {&#34;resolution&#34;: [512, 832], &#34;count&#34;: 3}, &#34;18&#34;: {&#34;resolution&#34;: [512, 896], &#34;count&#34;: 1525}, &#34;19&#34;: {&#34;resolution&#34;: [512, 960], &#34;count&#34;: 2}, &#34;20&#34;: {&#34;resolution&#34;: [512, 1024], &#34;count&#34;: 665}, &#34;21&#34;: {&#34;resolution&#34;: [512, 1088], &#34;count&#34;: 8}, &#34;22&#34;: {&#34;resolution&#34;: [576, 576], &#34;count&#34;: 5}, &#34;23&#34;: {&#34;resolution&#34;: [576, 768], &#34;count&#34;: 1}, &#34;24&#34;: {&#34;resolution&#34;: [576, 832], &#34;count&#34;: 667}, &#34;25&#34;: {&#34;resolution&#34;: [576, 896], &#34;count&#34;: 9601}, &#34;26&#34;: {&#34;resolution&#34;: [576, 960], &#34;count&#34;: 872}, &#34;27&#34;: {&#34;resolution&#34;: [576, 1024], &#34;count&#34;: 17}, &#34;28&#34;: {&#34;resolution&#34;: [640, 640], &#34;count&#34;: 3}, &#34;29&#34;: {&#34;resolution&#34;: [640, 768], &#34;count&#34;: 7}, &#34;30&#34;: {&#34;resolution&#34;: [640, 832], &#34;count&#34;: 608}, &#34;31&#34;: {&#34;resolution&#34;: [640, 896], &#34;count&#34;: 90}, &#34;32&#34;: {&#34;resolution&#34;: [704, 640], &#34;count&#34;: 1}, &#34;33&#34;: {&#34;resolution&#34;: [704, 704], &#34;count&#34;: 11}, &#34;34&#34;: {&#34;resolution&#34;: [704, 768], &#34;count&#34;: 1}, &#34;35&#34;: {&#34;resolution&#34;: [704, 832], &#34;count&#34;: 1}, &#34;36&#34;: {&#34;resolution&#34;: [768, 640], &#34;count&#34;: 225}, &#34;37&#34;: {&#34;resolution&#34;: [768, 704], &#34;count&#34;: 6}, &#34;38&#34;: {&#34;resolution&#34;: [768, 768], &#34;count&#34;: 74442}, &#34;39&#34;: {&#34;resolution&#34;: [832, 576], &#34;count&#34;: 23784}, &#34;40&#34;: {&#34;resolution&#34;: [832, 640], &#34;count&#34;: 554}, &#34;41&#34;: {&#34;resolution&#34;: [896, 512], &#34;count&#34;: 1235}, &#34;42&#34;: {&#34;resolution&#34;: [896, 576], &#34;count&#34;: 50}, &#34;43&#34;: {&#34;resolution&#34;: [896, 640], &#34;count&#34;: 88}, &#34;44&#34;: {&#34;resolution&#34;: [960, 512], &#34;count&#34;: 165}, &#34;45&#34;: {&#34;resolution&#34;: [960, 576], &#34;count&#34;: 5246}, &#34;46&#34;: {&#34;resolution&#34;: [1024, 448], &#34;count&#34;: 5}, &#34;47&#34;: {&#34;resolution&#34;: [1024, 512], &#34;count&#34;: 1187}, &#34;48&#34;: {&#34;resolution&#34;: [1024, 576], &#34;count&#34;: 40}, &#34;49&#34;: {&#34;resolution&#34;: [1088, 384], &#34;count&#34;: 70}, &#34;50&#34;: {&#34;resolution&#34;: [1088, 448], &#34;count&#34;: 36}, &#34;51&#34;: {&#34;resolution&#34;: [1088, 512], &#34;count&#34;: 3}, &#34;52&#34;: {&#34;resolution&#34;: [1216, 448], &#34;count&#34;: 36}, &#34;53&#34;: {&#34;resolution&#34;: [1344, 320], &#34;count&#34;: 29}, &#34;54&#34;: {&#34;resolution&#34;: [1536, 384], &#34;count&#34;: 1}}, &#34;mean_img_ar_error&#34;: 0.01693107810697896}&#34;,

### Speeds, Sizes, Times

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

ss_resolution:  &#34;(768, 768)&#34;,
ss_v2:  &#34;True&#34;,
ss_cache_latents:  &#34;False&#34;,
ss_unet_lr:  &#34;2e-07&#34;,
ss_num_reg_images:  &#34;0&#34;,
ss_max_token_length:  &#34;225&#34;,
ss_lr_scheduler:  &#34;linear&#34;,
ss_reg_dataset_dirs:  &#34;{}&#34;,
ss_lr_warmup_steps:  &#34;303&#34;,
ss_num_batches_per_epoch:  &#34;1222&#34;,
ss_lowram:  &#34;False&#34;,
ss_multires_noise_iterations:  &#34;None&#34;,
ss_optimizer:  &#34;torch.optim.adamw.AdamW(weight_decay=0.01,betas=(0.9, 0.99))&#34;,
 
# Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

## Testing Data, Factors & Metrics

### Testing Data

<!-- This should link to a Data Card if possible. -->

More information needed


### Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

More information needed

### Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

More information needed

## Results 

More information needed

# Model Examination

More information needed

# Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).

- **Hardware Type:** 8xA100
- **Hours used:** 64
- **Cloud Provider:** CoreWeave
- **Compute Region:** US Main
- **Carbon Emitted:** 6.72

# Technical Specifications [optional]

## Model Architecture and Objective

Enhance your model&#39;s quality and sharpness using your own pre-trained Unet.


## Compute Infrastructure

More information needed

### Hardware

8xA100

### Software

Fully trained only with Kohya S &amp; Shih-Ying Yeh (Kohaku-BlueLeaf)
https://arxiv.org/abs/2108.06098

# Citation

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

**BibTeX:**

More information needed

**APA:**

@misc{LyCORIS,
  author       = &#34;Shih-Ying Yeh (Kohaku-BlueLeaf), Yu-Guan Hsieh, Zhidong Gao&#34;,
  title        = &#34;LyCORIS - Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion&#34;,
  howpublished = &#34;\url{https://github.com/KohakuBlueleaf/LyCORIS}&#34;,
  month        = &#34;March&#34;,
  year         = &#34;2023&#34;
}

# Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->

More information needed

# More Information [optional]

More information needed

# Model Card Authors [optional]

<!-- This section provides another layer of transparency and accountability. Whose views is this model card representing? How many voices were included in its construction? Etc. -->

  idle stoev

# Model Card Contact

[email protected]

# How to Get Started with the Model

Use the code below to get started with the model.

<details>
<summary> Click to expand </summary>

For a1111
Install https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris

Download di.FFUSION.ai-tXe-FXAA to /models/Lycoris

Option1:

Insert <lyco:di.FFUSION.ai-tXe-FXAA:1.0> to prompt
No need to split Unet and Text Enc as its only TX encoder there.

You can go up to 2x weights

Option2: If you need it always ON (ex run a batch from txt file) then you can go to settings / Quicksettings list



add sd_lyco


restart and you should have a drop-down now 🤟 🥃

</details>