Update app.py

app.py

@@ -10,10 +10,8 @@ import os
 os.environ['CUDA_VISIBLE_DEVICES'] = ''
 
 device = torch.device("cpu")
-model = SomeModel()
-model = model.to(device)
 processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
-model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large").to("cuda")
+model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large").to("cpu")
 pipe = StableAudioPipeline.from_pretrained("stabilityai/stable-audio-open-1.0", torch_dtype=torch.float16)
 pipe = pipe.to("cpu")
 
@@ -36,7 +34,7 @@ with gr.Blocks(theme=gr.themes.Ocean(primary_hue="pink", neutral_hue="indigo", f
 
     def describir(url):
       raw_image = Image.open(requests.get(url, stream=True).raw).convert('RGB')
-      inputs = processor(raw_image, return_tensors="pt").to("cuda")
+      inputs = processor(raw_image, return_tensors="pt").to("cpu")
       out = model.generate(**inputs)
       return processor.decode(out[0], skip_special_tokens=True)
 
@@ -45,7 +43,7 @@ with gr.Blocks(theme=gr.themes.Ocean(primary_hue="pink", neutral_hue="indigo", f
         negative_prompt = "Low quality."
 
         # set the seed for generator
-        generator = torch.Generator("cuda").manual_seed(0)
+        generator = torch.Generator("cpu").manual_seed(0)
 
         # run the generation
         audio = pipe(