Spaces:
Runtime error
Runtime error
Delete image_to_text.py
Browse files- image_to_text.py +0 -31
image_to_text.py
DELETED
@@ -1,31 +0,0 @@
|
|
1 |
-
from transformers import BlipProcessor, BlipForConditionalGeneration
|
2 |
-
from PIL import Image
|
3 |
-
import spaces
|
4 |
-
|
5 |
-
@spaces.GPU
|
6 |
-
class ImageToText:
|
7 |
-
"""
|
8 |
-
Class to handle Image-to-Text captioning using BLIP.
|
9 |
-
"""
|
10 |
-
def __init__(self):
|
11 |
-
# Initialize the processor and model
|
12 |
-
print("Loading Image-to-Text model...")
|
13 |
-
self.processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
|
14 |
-
self.model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
|
15 |
-
print("Image-to-Text model loaded successfully.")
|
16 |
-
|
17 |
-
|
18 |
-
async def generate_caption(self, image):
|
19 |
-
"""
|
20 |
-
Generate a descriptive caption for an uploaded image.
|
21 |
-
|
22 |
-
Args:
|
23 |
-
image (PIL.Image): The image to caption.
|
24 |
-
|
25 |
-
Returns:
|
26 |
-
str: The generated caption.
|
27 |
-
"""
|
28 |
-
inputs = self.processor(image, return_tensors="pt")
|
29 |
-
out = self.model.generate(**inputs)
|
30 |
-
caption = self.processor.decode(out[0], skip_special_tokens=True)
|
31 |
-
return caption
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|