Update README.md

README.md

@@ -1,12 +1,29 @@
----
-title: Hate Speech Classifier
-emoji: 🐢
-colorFrom: purple
-colorTo: blue
-sdk: gradio
-sdk_version: 5.23.1
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# ToxiFilter: AI-Based Hate Speech Classifier 🚫
+
+A fine-tuned BERT model that detects hate speech and offensive language in real-time messages. Developed as part of a capstone project, it powers a Gradio-based chat simulation where offensive content is automatically censored.
+
+## 🔍 Model Info
+
+- **Base**: `bert-base-uncased`
+- **Task**: Binary Classification
+- **Labels**:
+  - `1`: Hate/Offensive
+  - `0`: Not Offensive
+- **Accuracy**: ~92%
+- **Dataset**: [tdavidson/hate_speech_offensive](https://huggingface.co/datasets/tdavidson/hate_speech_offensive)  
+  (Labels 0 and 1 combined as "offensive")
+
+## 🛠 Usage
+
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+model = AutoModelForSequenceClassification.from_pretrained("chaitravi/hate-speech-classifier")
+tokenizer = AutoTokenizer.from_pretrained("chaitravi/hate-speech-classifier")
+
+def classify(text):
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
+    outputs = model(**inputs)
+    pred = torch.argmax(outputs.logits, dim=1).item()
+    return "Hate/Offensive" if pred == 1 else "Not Offensive"