Add image classifier CPU Space

README.md

@@ -1,12 +1,13 @@
 ---
-title: Image Classifier Cpu
-emoji: 🦀
-colorFrom: gray
-colorTo: pink
+title: Image Classifier CPU
+colorFrom: indigo
+colorTo: blue
 sdk: gradio
-sdk_version: 6.10.0
 app_file: app.py
 pinned: false
+license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Image Classifier CPU
+
+Free CPU image classifier using `google/vit-base-patch16-224`.

app.py

@@ -0,0 +1,39 @@
+import gradio as gr
+
+from image_classifier.service import ImageClassifierService
+
+
+service = ImageClassifierService()
+
+
+def classify_image(image):
+    return service.classify(image)
+
+
+with gr.Blocks(
+    title="Image Classifier CPU",
+    theme=gr.themes.Soft(primary_hue="indigo", secondary_hue="blue"),
+) as demo:
+    gr.Markdown(
+        """
+        # Image Classifier CPU
+        Upload an image and get top predicted labels on free CPU.
+        """
+    )
+
+    image_input = gr.Image(type="pil", label="Input Image")
+    run_button = gr.Button("Classify", variant="primary")
+
+    top_label_output = gr.Textbox(label="Top Label", lines=1)
+    top_results_output = gr.Textbox(label="Top Results", lines=6)
+    status_output = gr.Textbox(label="Status", lines=2)
+
+    run_button.click(
+        fn=classify_image,
+        inputs=[image_input],
+        outputs=[top_label_output, top_results_output, status_output],
+    )
+
+
+if __name__ == "__main__":
+    demo.launch()

image_classifier/__init__.py

@@ -0,0 +1,3 @@
+from .service import ImageClassifierService
+
+__all__ = ["ImageClassifierService"]

image_classifier/service.py

@@ -0,0 +1,48 @@
+import os
+
+import torch
+
+
+MODEL_ID = "google/vit-base-patch16-224"
+
+
+class ImageClassifierService:
+    def __init__(self):
+        self.pipe = None
+        cpu_count = os.cpu_count() or 1
+        torch.set_num_threads(max(1, min(4, cpu_count)))
+
+    def classify(self, image):
+        if image is None:
+            return "", "", "Upload an image first."
+
+        try:
+            results = self._run_model(image)
+            top = results[0]
+            top_label = top["label"]
+            formatted = self._format_results(results)
+            return top_label, formatted, f"Classified image with {MODEL_ID}."
+        except Exception as exc:
+            return "", "", f"Image classification failed: {type(exc).__name__}: {exc}"
+
+    def _load_pipeline(self):
+        if self.pipe is not None:
+            return
+
+        from transformers import pipeline
+
+        self.pipe = pipeline(
+            "image-classification",
+            model=MODEL_ID,
+            device=-1,
+        )
+
+    def _run_model(self, image):
+        self._load_pipeline()
+        return self.pipe(image, top_k=5)
+
+    def _format_results(self, results):
+        lines = []
+        for item in results:
+            lines.append(f"{item['label']}: {item['score'] * 100:.1f}%")
+        return "\n".join(lines)

requirements.txt

@@ -0,0 +1,6 @@
+gradio>=5.23.0
+huggingface_hub>=0.34.0,<1.0
+Pillow>=10.4.0
+safetensors>=0.5.3
+torch>=2.3.0
+transformers>=4.49.0