File size: 5,627 Bytes
e5d6a79
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34ce06f
 
083695f
34ce06f
acde58c
083695f
34ce06f
083695f
34ce06f
 
083695f
 
 
 
 
 
 
 
 
 
34ce06f
 
 
 
8225e0d
 
 
 
 
 
 
3e6e3cf
8225e0d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
  - text-to-image
  - stable-diffusion
  - safetensors
  - stable-diffusion-xl
---
<style>
  .title{
    font-size: 2.5em;
    letter-spacing: 0.01em;
    padding: 0.5em 0;
  }
	.thumbwidth{
	max-width: 180px;
	}
	.font_red{
		color:red
	}
</style>

<a id="test03"></a> 
<h1 class="title">
<span>lora_zipang_flux_test</span>
</h1>
-Training was based on a merged model of dev1 and lora test**.<br/>
<br/>

### -Trigger
```bash
japanese, european
```
<br/>

* [test04](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/lora_zipang_flux_test04.safetensors?download=true) +350 images
```bash
myjc, 13yo
```

* [test03](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/lora_zipang_flux_test03.safetensors?download=true) +920 images
```bash
myjsh, 12yo
```

 <br/>
<a id="test02"></a> 
<h1 class="title">
<span>myjsm_flux_test02</span>
</h1>
  -It is a test lora of poor quality with only a few images learned.<br/>
 -trained 273 images.<br/>
  <br/>
Found a slightly better training setting.
But still hard to find things that don't show up in flux.
 <br/>
 <br/>
 
[Download:test02](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjsm_flux_test02.safetensors?download=true) <br/>
<br/>
The model used for the test is [Flux Fusion DS v0 GGUF Q4_0 (UNET)](https://civitai.com/models/630820?modelVersionId=736086) and [v0 GGUF Q5_K_M](https://civitai.com/models/630820?modelVersionId=765575).
<table>
  <tr>
    <td colspan="3">
      <div>
GGUF Q4_0 + t5xxl_fp8_e4m3fn : 4step
      </div>
	  </td>
	</tr>
<tr>
<td>
<a href="https://img99.pixhost.to/images/126/507626249_20240827094724_fusionds_v0_q4_456078958.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626249_20240827094724_fusionds_v0_q4_456078958.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626251_20240827103511_fusionds_v0_q4_482040669.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626251_20240827103511_fusionds_v0_q4_482040669.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626253_20240827112528_fusionds_v0_q4_1816421730.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626253_20240827112528_fusionds_v0_q4_1816421730.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>
  <tr>
    <td colspan="3">
      <div>
GGUF Q5_K_M. + t5-v1_1-xxl-encoder-Q5_K_M : 12step
      </div>
	  </td>
	</tr>
<tr>
<td>
<a href="https://img99.pixhost.to/images/126/507626250_20240827102458_fluxfusionds_v0_q5_k_m_2418428235.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626250_20240827102458_fluxfusionds_v0_q5_k_m_2418428235.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626252_20240827110802_fluxfusionds_v0_q5_k_m_3216545735.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626252_20240827110802_fluxfusionds_v0_q5_k_m_3216545735.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626256_20240827121409_fluxfusionds_v0_q5_k_m_2982180625.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626256_20240827121409_fluxfusionds_v0_q5_k_m_2982180625.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>
</table>
-refer to png info
<br />

## - sample prompt
[<img src=https://t99.pixhost.to/thumbs/126/507626257_20240827124249_fusionds_v0_q4_642879771.jpg />](https://img99.pixhost.to/images/126/507626257_20240827124249_fusionds_v0_q4_642879771.jpg)
```bash
9yo, myjsm, japanese,
photograph of a girl sitting on a brick pavement with a pink umbrella in front of her. She is wearing a white camisole and a blue skirt with a anime print. She has shoulder-length dark hair and is smiling at the camera.
bangs, black eyes, skirt, rain
<lora:myjsm_flux_test02:1>
Steps: 4, Sampler: Euler, Schedule type: Simple, CFG scale: 1, Seed: 642879771, Size: 792x1056, Model hash: 5e21feb505, Model: FusionDS_v0_Q4, Lora hashes: "myjsm_flux_test02: 3fdff20b7d65", Version: f2.0.1v1.10.1-previous-419-gf82029c5c, Module 1: ae, Module 2: clip_l, Module 3: t5xxl_fp8_e4m3fn
```
<br />

## - trigger
```bash
myjsm, japanese, 9yo, 
and native english
```
<br />

## -Train Settings
```bash
base model: flux1-dev.safetensors
vae/text encoder: clip_l.safetensors, t5xxl_fp8_e4m3fn.safetensors, ae.safetensors
tag: caption (native eng) + tags (danbooru)

--network_module "networks.lora_flux"
--gradient_checkpointing
--cache_latents
--cache_latents_to_disk
--cache_text_encoder_outputs
--cache_text_encoder_outputs_to_disk
--enable_bucket
--bucket_no_upscale
--optimizer_type "AdamW8bit"
--optimizer_args "weight_decay=0.01" "betas=0.9,0.999"
--learning_rate=0.0002
--network_dim=32
--network_alpha=4
--network_train_unet_only
--mixed_precision "bf16"
--save_precision "bf16"
--full_bf16
--loss_type "l2"
--huber_schedule "snr"
--model_prediction_type "raw"
--discrete_flow_shift 3
--timestep_sampling "sigma"
--max_grad_norm=1
--max_timestep=1000
--min_snr_gamma=5
--min_timestep=100
--noise_offset=0.0375
--adaptive_noise_scale=0.00375
--apply_t5_attn_mask
--split_mode
--network_args "loraplus_unet_lr_ratio=16" "train_blocks=single"
```
<br />