Spaces:
Running
on
A100
Running
on
A100
Avijit Ghosh
commited on
Commit
·
ab041ea
1
Parent(s):
8a06c57
add gpu wrapper
Browse files
app.py
CHANGED
@@ -10,6 +10,7 @@ import matplotlib.pyplot as plt
|
|
10 |
from matplotlib.colors import hex2color
|
11 |
import stone
|
12 |
import os
|
|
|
13 |
|
14 |
# Define model initialization functions
|
15 |
def load_model(model_name):
|
@@ -47,27 +48,29 @@ def load_model(model_name):
|
|
47 |
raise ValueError("Unknown model name")
|
48 |
return pipeline
|
49 |
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
|
57 |
-
|
58 |
-
|
59 |
|
60 |
# Initialize the default model
|
61 |
default_model = "stabilityai/sdxl-turbo"
|
62 |
|
63 |
pipeline_text2image = load_model(default_model)
|
64 |
|
|
|
65 |
def getimgen(prompt):
|
66 |
return pipeline_text2image(prompt=prompt, guidance_scale=0.0, num_inference_steps=2).images[0]
|
67 |
|
68 |
blip_processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
|
69 |
blip_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large", torch_dtype=torch.float16).to("cuda")
|
70 |
|
|
|
71 |
def blip_caption_image(image, prefix):
|
72 |
inputs = blip_processor(image, prefix, return_tensors="pt").to("cuda", torch.float16)
|
73 |
out = blip_model.generate(**inputs)
|
@@ -106,6 +109,7 @@ def skintoneplot(hex_codes):
|
|
106 |
ax.add_patch(plt.Rectangle((0, 0), 1, 1, color=sorted_hex_codes[i]))
|
107 |
return fig
|
108 |
|
|
|
109 |
def generate_images_plots(prompt, model_name):
|
110 |
global pipeline_text2image
|
111 |
pipeline_text2image = load_model(model_name)
|
|
|
10 |
from matplotlib.colors import hex2color
|
11 |
import stone
|
12 |
import os
|
13 |
+
import spaces
|
14 |
|
15 |
# Define model initialization functions
|
16 |
def load_model(model_name):
|
|
|
48 |
raise ValueError("Unknown model name")
|
49 |
return pipeline
|
50 |
|
51 |
+
choices=[
|
52 |
+
"stabilityai/sdxl-turbo",
|
53 |
+
"runwayml/stable-diffusion-v1-5",
|
54 |
+
"ByteDance/SDXL-Lightning",
|
55 |
+
"segmind/SSD-1B"
|
56 |
+
]
|
57 |
|
58 |
+
for model_name in choices:
|
59 |
+
load_model(model_name)
|
60 |
|
61 |
# Initialize the default model
|
62 |
default_model = "stabilityai/sdxl-turbo"
|
63 |
|
64 |
pipeline_text2image = load_model(default_model)
|
65 |
|
66 |
+
@spaces.GPU
|
67 |
def getimgen(prompt):
|
68 |
return pipeline_text2image(prompt=prompt, guidance_scale=0.0, num_inference_steps=2).images[0]
|
69 |
|
70 |
blip_processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
|
71 |
blip_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large", torch_dtype=torch.float16).to("cuda")
|
72 |
|
73 |
+
@spaces.GPU
|
74 |
def blip_caption_image(image, prefix):
|
75 |
inputs = blip_processor(image, prefix, return_tensors="pt").to("cuda", torch.float16)
|
76 |
out = blip_model.generate(**inputs)
|
|
|
109 |
ax.add_patch(plt.Rectangle((0, 0), 1, 1, color=sorted_hex_codes[i]))
|
110 |
return fig
|
111 |
|
112 |
+
@spaces.GPU
|
113 |
def generate_images_plots(prompt, model_name):
|
114 |
global pipeline_text2image
|
115 |
pipeline_text2image = load_model(model_name)
|