Update app.py

app.py

@@ -71,11 +71,13 @@ def shot(image, labels_text, model_name, hypothesis_template_prefix, hypothesis_
            hypothesis_template=hypothesis_template)
     return {dic["label"]: dic["score"] for dic in res}
 
+#clear, winter, day, front, moderate occlusion
+
 iface = gr.Interface(shot, 
             inputs, 
             "label", 
             examples=[["festival.jpg", "lantern, firecracker, couplet", "ViT/B-16", "a photo of a {}", "in {} {} {} from {} with {}.", "clear, autumn, day, side, light occlusion"],
-                     ["car.png", "car, bike, truck", "ViT/B-16", "a photo of a {}", "in {} {} {} from {} with {}.", "clear, winter, day, front, moderate occlusion"]], 
+                     ["car.png", "car, bike, truck", "ViT/B-16", "a photo of a {}", "in {} {} {} from {} with {}.", ""]], 
             description="""<p>Chinese CLIP is a contrastive-learning-based vision-language foundation model pretrained on large-scale Chinese data. For more information, please refer to the paper and official github. Also, Chinese CLIP has already been merged into Huggingface Transformers! <br><br>
             Paper: <a href='https://arxiv.org/pdf/2403.02714'>https://arxiv.org/pdf/2403.02714</a> <br>
             To begin with the demo, provide a picture (either upload manually, or select from the given examples) and add class labels one by one. Optionally, you can also add template as a prefix to the class labels. <br>""",