Add extended description, tags, and improved README formatting

This project demonstrates how to fine-tune and deploy a lightweight NLP model for news classification. Although trained on a small sample dataset, the model generalizes basic categories reasonably well. You can further fine-tune it using larger and richer datasets such as ag_news.

Perfect for:

Students learning NLP

Fine-tuning transformers on custom tasks

Deploying small models with Hugging Face and 🤗 Transformers

README.md

@@ -1,30 +1,34 @@
 ---
 license: apache-2.0
 language:
-- en
+  - en
 metrics:
-- accuracy
+  - accuracy
 base_model:
-- distilbert/distilbert-base-uncased-finetuned-sst-2-english
+  - distilbert-base-uncased
 pipeline_tag: text-classification
 library_name: transformers
+tags:
+  - text-classification
+  - news
+  - distilbert
+  - huggingface
+  - ag_news
 ---
+
 # 📚 DistilBERT News Classifier by Dheeraj
 
 A text classification model fine-tuned on news headlines to classify them into one of four categories:
 
-- **0 → Sports**  
-- **1 → Business**  
-- **2 → Tech**  
-- **3 → Science**
-
----
+- `0 → Sports`  
+- `1 → Business`  
+- `2 → Tech`  
+- `3 → Science`  
 
 ## 🧠 Model Info
 
-This model is based on [`distilbert-base-uncased`](https://huggingface.co/distilbert-base-uncased) and trained using Hugging Face's `Trainer` API.
-
----
+This model is built using [DistilBERT](https://huggingface.co/distilbert-base-uncased) and fine-tuned using Hugging Face's `Trainer` API.  
+It is trained on a small dataset of sample headlines, inspired by AG News, for demonstration and educational purposes.
 
 ## 🚀 Example Usage
 
@@ -33,4 +37,4 @@ from transformers import pipeline
 
 classifier = pipeline("text-classification", model="Dheeraj3103/distilbert-news-classifier-dheeraj")
 result = classifier("NASA finds evidence of water on Mars.")
-print(result)
+print(result)