SG161222 commited on
Commit
5d6d970
Β·
verified Β·
1 Parent(s): f98c8c0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +31 -7
README.md CHANGED
@@ -3,10 +3,34 @@ license: other
3
  license_name: flux-1-dev-non-commercial-license
4
  license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
5
  ---
6
- <b>RealFlux 1.0b Training Status:</b><br><br>
7
- [ β˜‘ ] Collecting images into a new dataset<br>
8
- [ β˜‘ ] Image captioning<br>
9
- [ β˜‘ ] Preparing the dataset for training<br>
10
- [ β˜‘ ] Training<br>
11
- [ β–Ά ] Testing the second model (the first model is unsuccessful)<br>
12
- [ ☐ ] Release
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  license_name: flux-1-dev-non-commercial-license
4
  license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
5
  ---
6
+ <h1>RealFlux 1.0b Model Training (Third Attempt)</h1>
7
+
8
+ <h2>Stage 1: Data Preparation</h2>
9
+ <ul>
10
+ <li>[ β˜‘ ] Collect images for the dataset</li>
11
+ <li>[ β˜‘ ] Image captioning</li>
12
+ </ul>
13
+
14
+ <h2>Stage 2: Model Training</h2>
15
+ <ul>
16
+ <li>[ β˜‘ ] Prepare dataset for model input</li>
17
+ <li>[ β–Ά ] Model training</li>
18
+ </ul>
19
+
20
+ <h2>Stage 3: Testing</h2>
21
+ <ul>
22
+ <li>[ ☐ ] Model testing</li>
23
+ </ul>
24
+
25
+ <h2>Stage 4: Release</h2>
26
+ <ul>
27
+ <li>[ ☐ ] Final validation</li>
28
+ <li>[ ☐ ] Release Compact version and Transformer</li>
29
+ <li>[ ☐ ] Release GGUF Q8 version</li>
30
+ </ul>
31
+
32
+ <h2>Comments</h2>
33
+
34
+ <p>The first version of the model turned out to be unsuccessful due to significant anatomical errors. The second version of the model is being trained on a reassembled dataset with better-chosen training parameters.</p>
35
+
36
+ <p>The second version of the model has even fewer anatomical errors and text errors. The model has been sent for refinement with a slightly modified dataset.</p>