Update README.md

README.md

@@ -3,6 +3,9 @@ license: apache-2.0
 datasets:
 - yunusserhat/TurkishFoods-25
 ---
+# TurkishFoods-25
+
+> **TurkishFoods-25** is a computer vision model fine-tuned from `google/siglip2-base-patch16-224` for multi-class food image classification. It is trained to identify 25 traditional Turkish dishes using the `SiglipForImageClassification` architecture.
 
 ```py
 Classification Report:
@@ -37,4 +40,99 @@ hunkar_begendi     0.9583    0.9274    0.9426       248
       accuracy                         0.9186      9168
      macro avg     0.9234    0.9216    0.9220      9168
   weighted avg     0.9194    0.9186    0.9186      9168
-```
+```
+
+---
+
+## Label Space: 25 Classes
+
+The model classifies food images into the following Turkish dishes:
+
+```json
+"id2label": {
+  "0": "asure",
+  "1": "baklava",
+  "2": "biber_dolmasi",
+  "3": "borek",
+  "4": "cig_kofte",
+  "5": "enginar",
+  "6": "et_sote",
+  "7": "gozleme",
+  "8": "hamsi",
+  "9": "hunkar_begendi",
+  "10": "icli_kofte",
+  "11": "ispanak",
+  "12": "izmir_kofte",
+  "13": "karniyarik",
+  "14": "kebap",
+  "15": "kisir",
+  "16": "kuru_fasulye",
+  "17": "lahmacun",
+  "18": "lokum",
+  "19": "manti",
+  "20": "mucver",
+  "21": "pirinc_pilavi",
+  "22": "simit",
+  "23": "taze_fasulye",
+  "24": "yaprak_sarma"
+}
+```
+
+---
+
+## Install Requirements
+
+```bash
+pip install -q transformers torch pillow gradio
+```
+
+---
+
+## Inference Script
+
+```python
+import gradio as gr
+from transformers import AutoImageProcessor, SiglipForImageClassification
+from PIL import Image
+import torch
+
+model_name = "prithivMLmods/TurkishFoods-25"  # Replace with your Hugging Face repo
+model = SiglipForImageClassification.from_pretrained(model_name)
+processor = AutoImageProcessor.from_pretrained(model_name)
+
+id2label = {
+    "0": "asure", "1": "baklava", "2": "biber_dolmasi", "3": "borek", "4": "cig_kofte",
+    "5": "enginar", "6": "et_sote", "7": "gozleme", "8": "hamsi", "9": "hunkar_begendi",
+    "10": "icli_kofte", "11": "ispanak", "12": "izmir_kofte", "13": "karniyarik", "14": "kebap",
+    "15": "kisir", "16": "kuru_fasulye", "17": "lahmacun", "18": "lokum", "19": "manti",
+    "20": "mucver", "21": "pirinc_pilavi", "22": "simit", "23": "taze_fasulye", "24": "yaprak_sarma"
+}
+
+def predict_food(image):
+    image = Image.fromarray(image).convert("RGB")
+    inputs = processor(images=image, return_tensors="pt")
+    with torch.no_grad():
+        logits = model(**inputs).logits
+        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
+    return {id2label[str(i)]: round(probs[i], 3) for i in range(len(probs))}
+
+iface = gr.Interface(
+    fn=predict_food,
+    inputs=gr.Image(type="numpy"),
+    outputs=gr.Label(num_top_classes=5, label="Top Turkish Foods"),
+    title="TurkishFoods-25 Classifier",
+    description="Upload a food image to identify one of 25 Turkish dishes."
+)
+
+if __name__ == "__main__":
+    iface.launch()
+```
+
+---
+
+## Applications
+
+* Turkish cuisine image datasets
+* Food delivery or smart restaurant apps
+* Culinary learning platforms
+* Nutrition tracking via image-based recognition