Update README.md

README.md

@@ -1,6 +1,21 @@
 ---
 license: apache-2.0
+datasets:
+- Vedant3907/Hindi-Sign-Language-Dataset
+language:
+- en
+base_model:
+- google/siglip2-base-patch16-224
+pipeline_tag: image-classification
+library_name: transformers
+tags:
+- Hindi-Sign-Language-Detection
+- SigLIP2
+- 93M
 ---
+# Hindi-Sign-Language-Detection
+
+> Hindi-Sign-Language-Detection is a vision-language model fine-tuned from google/siglip2-base-patch16-224 for multi-class image classification. It is trained to detect and classify Hindi sign language hand gestures into corresponding Devanagari characters using the SiglipForImageClassification architecture.
 
 ```py
 Classification Report:
@@ -39,4 +54,116 @@ Classification Report:
     accuracy                         0.9425     16177
    macro avg     0.9433    0.9426    0.9413     16177
 weighted avg     0.9457    0.9425    0.9425     16177
-```
+```
+
+---
+
+## Label Space: 29 Classes
+
+The model classifies a hand sign into one of the following 29 Hindi characters:
+
+```json
+"id2label": {
+  "0": "ऋ",
+  "1": "क",
+  "2": "ख",
+  "3": "ग",
+  "4": "घ",
+  "5": "ङ",
+  "6": "च",
+  "7": "छ",
+  "8": "ज",
+  "9": "झ",
+  "10": "ट",
+  "11": "ठ",
+  "12": "ड",
+  "13": "ण",
+  "14": "त",
+  "15": "थ",
+  "16": "द",
+  "17": "न",
+  "18": "प",
+  "19": "फ",
+  "20": "ब",
+  "21": "भ",
+  "22": "म",
+  "23": "य",
+  "24": "र",
+  "25": "ल",
+  "26": "व",
+  "27": "स",
+  "28": "ह"
+}
+```
+
+---
+
+## Install Dependencies
+
+```bash
+pip install -q transformers torch pillow gradio
+```
+
+---
+
+## Inference Code
+
+```python
+import gradio as gr
+from transformers import AutoImageProcessor, SiglipForImageClassification
+from PIL import Image
+import torch
+
+# Load model and processor
+model_name = "prithivMLmods/Hindi-Sign-Language-Detection"  # Replace with actual path
+model = SiglipForImageClassification.from_pretrained(model_name)
+processor = AutoImageProcessor.from_pretrained(model_name)
+
+# Hindi label mapping
+id2label = {
+    "0": "ऋ", "1": "क", "2": "ख", "3": "ग", "4": "घ",
+    "5": "ङ", "6": "च", "7": "छ", "8": "ज", "9": "झ",
+    "10": "ट", "11": "ठ", "12": "ड", "13": "ण", "14": "त",
+    "15": "थ", "16": "द", "17": "न", "18": "प", "19": "फ",
+    "20": "ब", "21": "भ", "22": "म", "23": "य", "24": "र",
+    "25": "ल", "26": "व", "27": "स", "28": "ह"
+}
+
+def classify_hindi_sign(image):
+    image = Image.fromarray(image).convert("RGB")
+    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().tolist()
+
+    prediction = {
+        id2label[str(i)]: round(probs[i], 3) for i in range(len(probs))
+    }
+
+    return prediction
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=classify_hindi_sign,
+    inputs=gr.Image(type="numpy"),
+    outputs=gr.Label(num_top_classes=3, label="Hindi Sign Classification"),
+    title="Hindi-Sign-Language-Detection",
+    description="Upload an image of a Hindi sign language hand gesture to identify the corresponding character."
+)
+
+if __name__ == "__main__":
+    iface.launch()
+```
+
+---
+
+## Intended Use
+
+Hindi-Sign-Language-Detection can be used in:
+
+* Educational tools for learning Indian sign language.
+* Assistive technology for hearing and speech-impaired individuals.
+* Real-time sign-to-text translation applications.
+* Human-computer interaction for Hindi users.