Create app.py

app.py

@@ -0,0 +1,49 @@
+# impoprt packages
+import torch
+import requests
+from PIL import Image
+from transformers import BlipProcessor, BlipForConditionalGeneration, AutoTokenizer, pipeline
+import sentencepiece
+import gradio as gr
+
+# Image captioning model
+processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
+model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
+
+# Translate en to ar
+model_translater = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-big-en-ar")
+
+# conditional image captioning (with prefix-)
+def image_captioning(image_url, prefix="a "):
+  """ Return text (As str) to describe an image """
+  # Get the image by image_url and convert it to RGB
+  raw_image = Image.open(requests.get(image_url, stream=True).raw).convert('RGB')
+
+  # Process the image
+  inputs = processor(raw_image, prefix, return_tensors="pt")
+
+  # Generate text to describe the image
+  output = model.generate(**inputs)
+
+  # Decode the output
+  output = processor.decode(output[0], skip_special_tokens=True, max_length=80)
+  return output
+
+def translate_text(text, to="ar"):
+  """ Return translated text """
+  translated_text = model_translater(str(text))
+  return translated_text[0]['translation_text']
+
+def image_captioning_ar(image_url, prefix = "a "):
+  if image_url:
+    text = image_captioning(image_url, prefix=prefix)
+    return translate_text(text)
+  return null
+    
+imageCaptioning_interface = gr.Interface(
+    fn = image_captioning_ar,
+    inputs=gr.inputs.Textbox(lines = 7, label = 'Image url'),
+    outputs=gr.outputs.Textbox(label="Caption"),
+    title = 'Image captioning',
+)
+imageCaptioning_interface.launch()