| import gradio as gr | |
| from transformers import AutoImageProcessor, SiglipForImageClassification | |
| from PIL import Image | |
| import torch | |
| model_name = "prithivMLmods/Dog-Breed-120" | |
| processor = AutoImageProcessor.from_pretrained(model_name) | |
| model = SiglipForImageClassification.from_pretrained(model_name) | |
| model.eval() | |
| def predict(image): | |
| inputs = processor(images=image, return_tensors="pt") | |
| with torch.no_grad(): | |
| outputs = model(**inputs) | |
| logits = outputs.logits | |
| probs = torch.nn.functional.softmax(logits, dim=1).squeeze() | |
| top_idx = probs.argmax().item() | |
| label = model.config.id2label[top_idx] | |
| confidence = probs[top_idx].item() | |
| return f"{label} ({round(confidence*100, 1)}%)" | |
| demo = gr.Interface( | |
| fn=predict, | |
| inputs=gr.Image(type="pil"), | |
| outputs="text", | |
| title="Dog Breed Classifier 🐶", | |
| description="Upload a dog photo to identify its breed using the prithivMLmods/Dog-Breed-120 model." | |
| ) | |
| demo.launch() | |