Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import CLIPFeatureExtractor
|
2 |
+
from safety_checker import StableDiffusionSafetyChecker
|
3 |
+
import torch
|
4 |
+
from PIL import Image
|
5 |
+
import gradio as gr
|
6 |
+
from pathlib import Path
|
7 |
+
|
8 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
9 |
+
safety_checker = StableDiffusionSafetyChecker.from_pretrained(
|
10 |
+
"CompVis/stable-diffusion-safety-checker"
|
11 |
+
).to(device)
|
12 |
+
feature_extractor = CLIPFeatureExtractor.from_pretrained("openai/clip-vit-base-patch32")
|
13 |
+
|
14 |
+
|
15 |
+
import gradio as gr
|
16 |
+
|
17 |
+
|
18 |
+
def image_classifier(files):
|
19 |
+
images = [Image.open(file).convert("RGB").resize((512, 512)) for file in files]
|
20 |
+
|
21 |
+
safety_checker_input = feature_extractor(images, return_tensors="pt").to(device)
|
22 |
+
has_nsfw_concepts = safety_checker(
|
23 |
+
images=[images], clip_input=safety_checker_input.pixel_values.to(torch.float16)
|
24 |
+
)
|
25 |
+
results = [
|
26 |
+
{"has_nsfw": nsfw, "file": Path(file).name}
|
27 |
+
for (nsfw, file) in zip(has_nsfw_concepts, files)
|
28 |
+
]
|
29 |
+
return {"results": results}
|
30 |
+
|
31 |
+
|
32 |
+
demo = gr.Interface(
|
33 |
+
fn=image_classifier,
|
34 |
+
inputs=gr.File(file_count="multiple", file_types=["image"]),
|
35 |
+
outputs="json",
|
36 |
+
api_name="classify",
|
37 |
+
)
|
38 |
+
demo.launch()
|