demo fix commit

app.py

@@ -0,0 +1,51 @@
+# import gradio as gr
+
+# def greet(image):
+#     return "Shape  " + image.shape + "!!"
+
+# iface = gr.Interface(fn=greet, inputs="image", outputs="text")
+# iface.launch()
+
+
+import sys
+from IPython.display import display, HTML
+from BLIP.models.blip import blip_decoder
+from google_drive_downloader import GoogleDriveDownloader as gdd
+from PIL import Image
+import requests
+import torch
+from torchvision import transforms
+from torchvision.transforms.functional import InterpolationMode
+from urllib.parse import urlparse
+from google_drive_downloader import GoogleDriveDownloader as gdd
+
+
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+
+transform = transforms.Compose([
+        transforms.Resize((image_size,image_size),interpolation=InterpolationMode.BICUBIC),
+        transforms.ToTensor(),
+        transforms.Normalize((0.48145466, 0.4578275, 0.40821073), (0.26862954, 0.26130258, 0.27577711))
+        ]) 
+
+model_url = "https://technionmail-my.sharepoint.com/personal/snoamr_campus_technion_ac_il/_layouts/15/download.aspx?share=EZxgXQaBXGREgDsQiaTcwAAB0z8jQA_hgAnwwPQDt8Dgew"
+model = blip_decoder(pretrained=model_url, image_size=384, vit='base')
+model.eval()
+model = model.to(device)
+
+def inference(raw_image):
+    image = transform(raw_image).unsqueeze(0).to(device)
+    with torch.no_grad():
+          caption = model.generate(image, sample=False, num_beams=1, max_length=60, min_length=5) 
+      return caption[0]
+
+
+inputs = [gr.Image(type='pil', interactive=False),]
+outputs = gr.outputs.Textbox(label="Caption")
+
+title = "FuseCap"
+description = "Gradio demo for FuseCap: Leveraging Large Language Models to Fuse Visual Data into Enriched Image Captions. This demo features a BLIP-based model, trained using FuseCap."
+
+article = "place holder"
+gr.Interface(inference, inputs, outputs, title=title, description=description, article=article, examples=[['birthday_dog.jpeg']]).launch(enable_queue=True)