add code for inference with BlipForConditionalGeneration

app.py

@@ -1,7 +1,21 @@
 import gradio as gr
+from PIL import Image
+import requests
 
-def greet(name):
-    return "Hello " + name + "!!"
+from transformers import BlipProcessor, BlipForConditionalGeneration
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+
+model_id = "Salesforce/blip-image-captioning-base"
+
+model = BlipForConditionalGeneration.from_pretrained(model_id)
+processor = BlipProcessor.from_pretrained(model_id)
+
+def launch(input):
+    image = Image.open(requests.get(input, stream=True).raw).convert('RGB')
+    inputs = processor(image, return_tensors="pt")
+    out = model.generate(**inputs)
+    return processor.decode(out[0], skip_special_tokens=True)
+
+
+iface = gr.Interface(launch, inputs="text", outputs="text")
 iface.launch()

requirements.txt

@@ -0,0 +1,2 @@
+git+https://github.com/huggingface/transformers.git@main
+torch