Bootloader commited on
Commit
e6589c0
β€’
1 Parent(s): cfa7334

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -19
README.md CHANGED
@@ -2,44 +2,48 @@
2
  license: unknown
3
  ---
4
 
5
- # Ai to Ai LoRA
 
 
6
 
7
- ## Model Details
8
 
9
  - **Model Name:** Ai2Ai_v1
10
- - **Description:** a model trained on some images collected from pinterest and those images were ai generated as well
11
  - **Version:** v1
12
- - **Date of Release:** I forgot!
13
- - **Framework Used:** Kohya trainer colab
14
 
15
- ## Intended Use
16
 
17
- This model is designed as a LoRA. It is recommended for generating anime girls. It is not suitable for full body and clothes.
18
 
19
- ## Performance
20
 
21
- - **Accuracy:** high on face and hat, low on body
22
- - **Precision:** high on face and hat, low on body
23
 
24
- ## Limitations and Bias
25
 
26
- - **Limitations:** a small LoRA, generates identical faces
27
- - **Bias Considerations:** jpt artifacts, cheek bulge, too powerfull
28
- ## Training Data
29
 
30
- - **Data Sources:** Pinterest
31
- - **Data Size:** 82 images, around 12 Mb
32
 
 
 
 
33
 
34
- ## Examples
35
 
36
  Example 1:
37
  ![Example 1](image_url_for_example_1)
38
 
39
- Caption: [Description of the example]
40
 
41
  Example 2:
42
  ![Example 2](image_url_for_example_2)
43
 
44
- Caption: [Description of the example]
45
 
 
 
2
  license: unknown
3
  ---
4
 
5
+ <div align="center">
6
+ <h1>πŸš€ Model Card: Ai2Ai_v1 πŸš€</h1>
7
+ </div>
8
 
9
+ ## 🌟 Model Details
10
 
11
  - **Model Name:** Ai2Ai_v1
12
+ - **Description:** A cutting-edge LoRA trained on AI-generated images sourced from Pinterest.
13
  - **Version:** v1
14
+ - **Release Date:** [Release date - Insert Date]
15
+ - **Framework:** Kohya Trainer Colab
16
 
17
+ ## 🌍 Intended Use
18
 
19
+ This revolutionary LoRA (Limitless Realm AI) is designed for generating stunning anime girls with incredible creativity. It excels in generating detailed facial features and captivating hats. However, it's not optimized for rendering full body and clothing designs.
20
 
21
+ ## πŸ“Š Performance
22
 
23
+ - **Accuracy:** High on face and hat generation, lower on body details
24
+ - **Precision:** High on face and hat generation, lower on body details
25
 
26
+ ## ⚑ Limitations and Bias
27
 
28
+ - **Limitations:** This LoRA model is relatively small, resulting in the generation of somewhat identical faces.
29
+ - **Bias Considerations:** Beware of potential artifacts such as "jpt artifacts," "cheek bulge," and the model's potential to be overly powerful in certain aspects.
 
30
 
31
+ ## 🎯 Training Data
 
32
 
33
+ - **Data Sources:** Images from Pinterest
34
+ - **Data Size:** 82 images, approximately 12 MB
35
+ - **Data Preprocessing:** AI-generated images carefully curated for training quality
36
 
37
+ ## 🌟 Examples
38
 
39
  Example 1:
40
  ![Example 1](image_url_for_example_1)
41
 
42
+ *Caption: [Description of the example]*
43
 
44
  Example 2:
45
  ![Example 2](image_url_for_example_2)
46
 
47
+ *Caption: [Description of the example]*
48
 
49
+ ---