onkarsus13 commited on
Commit
8ad8e8f
1 Parent(s): b90b282

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -9
README.md CHANGED
@@ -8,12 +8,10 @@ you will find the code to run the model [here](https://github.com/Onkarsus13/Dif
8
 
9
  For direct inference
10
 
11
- step 1: Clone the github repo to get the customized ControlNet-StableDiffusion-inpaint Pipeline implimentation
12
-
13
  `git clone https://github.com/Onkarsus13/Diff_SceneTextEraser`
14
 
15
  Step2: Go into the repository and install repository, dependency
16
-
17
  ```
18
  cd Diff_SceneTextEraser
19
  pip install -e ".[torch]"
@@ -37,18 +35,14 @@ import math
37
  import os
38
 
39
  device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
40
-
41
  model_path = "onkarsus13/controlnet_stablediffusion_scenetextEraser"
42
 
43
  pipe = StableDiffusionControlNetSceneTextErasingPipeline.from_pretrained(model_path)
44
-
45
  pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
46
-
47
  pipe.to(device)
48
 
49
  # pipe.enable_xformers_memory_efficient_attention()
50
  pipe.enable_model_cpu_offload()
51
-
52
  generator = torch.Generator(device).manual_seed(1)
53
 
54
  image = Image.open("<path to scene text image>").resize((512, 512))
@@ -63,7 +57,5 @@ image = pipe(
63
  controlnet_conditioning_scale=1.0,
64
  guidance_scale=1.0
65
  ).images[0]
66
-
67
  image.save('test1.png')
68
-
69
  ```
 
8
 
9
  For direct inference
10
 
11
+ step 1: Clone the GitHub repo to get the customized ControlNet-StableDiffusion-inpaint Pipeline Implementation
 
12
  `git clone https://github.com/Onkarsus13/Diff_SceneTextEraser`
13
 
14
  Step2: Go into the repository and install repository, dependency
 
15
  ```
16
  cd Diff_SceneTextEraser
17
  pip install -e ".[torch]"
 
35
  import os
36
 
37
  device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
38
  model_path = "onkarsus13/controlnet_stablediffusion_scenetextEraser"
39
 
40
  pipe = StableDiffusionControlNetSceneTextErasingPipeline.from_pretrained(model_path)
 
41
  pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
 
42
  pipe.to(device)
43
 
44
  # pipe.enable_xformers_memory_efficient_attention()
45
  pipe.enable_model_cpu_offload()
 
46
  generator = torch.Generator(device).manual_seed(1)
47
 
48
  image = Image.open("<path to scene text image>").resize((512, 512))
 
57
  controlnet_conditioning_scale=1.0,
58
  guidance_scale=1.0
59
  ).images[0]
 
60
  image.save('test1.png')
 
61
  ```