mjbuehler commited on
Commit
d34cc1b
·
verified ·
1 Parent(s): fcb21e8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +25 -2
README.md CHANGED
@@ -33,11 +33,11 @@ pip install -U diffusers
33
 
34
  These are LoRA adaption weights for the FLUX.1 [dev] model (```black-forest-labs/FLUX.1-dev```). The base model is, and you must first get access to it before loading this LoRA adapter.
35
 
36
- This LoRA adapter has rank=64 and alpha=64, trained for 4,000 steps. Earlier checkpoints are available in this repository as well (you can load these via the ```adapter``` parameter, see example below).
37
 
38
  ## Trigger keywords
39
 
40
- The model was fine-tuned with a set of ~1,600 images of biological materials, structures, shapes and other images of nature, using the keyword \bioinspired\>.
41
 
42
  You should use \<bioinspired\> to trigger these features during image generation.
43
 
@@ -139,6 +139,7 @@ grid
139
 
140
 
141
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/qdnFLSWbzXOKeZdAUzNMH.png)
 
142
  Image generation - Example #2:
143
 
144
  ```python
@@ -160,6 +161,28 @@ grid
160
 
161
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/kMTHiPszlXnZoBT-4klLY.png)
162
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
163
 
164
  ```bibtext
165
  @article{BioinspiredFluxBuehler2024,
 
33
 
34
  These are LoRA adaption weights for the FLUX.1 [dev] model (```black-forest-labs/FLUX.1-dev```). The base model is, and you must first get access to it before loading this LoRA adapter.
35
 
36
+ This LoRA adapter has rank=64 and alpha=64, trained for 16,000 steps. Earlier checkpoints are available in this repository as well (you can load these via the ```adapter``` parameter, see example below).
37
 
38
  ## Trigger keywords
39
 
40
+ The model was fine-tuned with a set of ~1,600 images of biological materials, structures, shapes and other images of nature, using the keyword \<bioinspired\>.
41
 
42
  You should use \<bioinspired\> to trigger these features during image generation.
43
 
 
139
 
140
 
141
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/qdnFLSWbzXOKeZdAUzNMH.png)
142
+
143
  Image generation - Example #2:
144
 
145
  ```python
 
161
 
162
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/kMTHiPszlXnZoBT-4klLY.png)
163
 
164
+ Image generation - Example #3:
165
+
166
+ ```python
167
+ prompt = "An architectural design in the style of <bioinspired>. The structure itself features key design elements as in <bioinspired>."
168
+ num_samples =1
169
+ num_rows =1
170
+ n_steps=50
171
+ guidance_scale=5.
172
+ all_images = []
173
+ for _ in range(num_rows):
174
+ image = pipeline(prompt,num_inference_steps=n_steps,num_images_per_prompt=num_samples,
175
+ guidance_scale=guidance_scale,).images
176
+
177
+ all_images.extend(image)
178
+
179
+ grid = image_grid(all_images, num_rows, num_samples, save_individual_files=True, )
180
+ grid
181
+ ```
182
+
183
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/5t4cG5s7-Yf6bzBmNFqbQ.png)
184
+
185
+
186
 
187
  ```bibtext
188
  @article{BioinspiredFluxBuehler2024,