Upload 6 files

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+example_images/000002.png filter=lfs diff=lfs merge=lfs -text
+example_images/000011.jpg filter=lfs diff=lfs merge=lfs -text
+example_images/000048.png filter=lfs diff=lfs merge=lfs -text
+example_images/Chihiro_13.PNG filter=lfs diff=lfs merge=lfs -text
+example_images/Kiki_01.PNG filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+from transformers import pipeline
+
+# Load your fine-tuned ViT model
+vit_classifier = pipeline("image-classification", model="itsJasminZWIN/chihiro-classifier")
+
+# Load CLIP zero-shot model
+clip_detector = pipeline(model="openai/clip-vit-base-patch32", task="zero-shot-image-classification")
+
+# Labels for both models
+label_names = ['chihiro', 'not chihiro']
+
+# Define classification function
+def classify_character(image):
+    # Run ViT classifier
+    vit_results = vit_classifier(image)
+    vit_output = {res['label']: round(res['score'], 4) for res in vit_results}
+
+    # Run CLIP zero-shot classifier
+    clip_results = clip_detector(image, candidate_labels=label_names)
+    clip_output = {res['label']: round(res['score'], 4) for res in clip_results}
+
+    return {
+        "ViT Classification": vit_output,
+        "CLIP Zero-Shot Classification": clip_output
+    }
+
+# Optional: Example images from your project (use actual image paths or URLs)
+example_images = [
+    ["example_images/Chihiro_13.PNG"],
+    ["example_images/000048.png"],
+    ["example_images/000011.jpg"],
+    ["example_images/000002.png."],
+    ["example_images/Kiki_01.PNG"]
+    
+]
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=classify_character,
+    inputs=gr.Image(type="filepath"),
+    outputs=gr.JSON(),
+    title="Chihiro Classifier: ViT vs. CLIP",
+    description="Upload an image to see if it's Chihiro or not, using both a fine-tuned ViT model and CLIP zero-shot comparison.",
+    examples=example_images
+)
+
+iface.launch()