Update README.md

README.md

@@ -8,62 +8,45 @@ tags:
 - gguf
 - llama.cpp
 - phi-3
+pipeline_tag: text-classification
 ---
-# 📧 Email Classifier (Phi-3 Mini GGUF)
+# 📧 Email Classifier (Phi-3 Mini Fine-Tuned GGUF)
 
 ## 🧠 Model Overview
 
-This repository contains a **GGUF quantized version** of a Phi-3 Mini model intended for email classification tasks.
+This model is a **fine-tuned version of Phi-3 Mini (4K Instruct)** optimized for **email classification tasks** such as spam detection and categorization.
 
 * **Base Model:** Phi-3 Mini 4K Instruct
-* **Format:** GGUF
+* **Fine-tuning:** Supervised fine-tuning on email dataset
+* **Format:** GGUF (for efficient local inference)
 * **Quantization:** Q5_K_M
-* **Framework:** llama.cpp / compatible runtimes
 
 ---
 
-## 🚀 Intended Use
+## 📚 Dataset
 
-This model is designed for:
+The model was trained on:
 
-* Email classification (spam / important / categories)
-* Lightweight local inference
-* Edge or CPU-based deployments
+* **Dataset:** `json22322/email-classifier`
+* Contains labeled email samples for classification tasks (e.g., spam vs non-spam)
 
 ---
 
-## ⚙️ Usage
-
-### Using llama.cpp
+## 🎯 Task
 
-```bash
-./main -m Phi-3-mini-4k-instruct-Q5_K_M.gguf -p "Classify this email: ..."
-```
+* Email classification
+* Spam detection
+* Priority categorization
 
 ---
 
-## 📦 Files
-
-| File Name                          | Description          |
-| ---------------------------------- | -------------------- |
-| Phi-3-mini-4k-instruct-Q5_K_M.gguf | Quantized model file |
-
----
-
-## 📊 Model Details
-
-* Context Length: 4K tokens
-* Architecture: Transformer
-* Quantization Type: Q5_K_M
-* Precision: 5-bit
-
----
+## ⚙️ Usage
 
-## ⚠️ Limitations
+### Using llama.cpp
 
-* Not fine-tuned specifically for email classification
-* May produce inconsistent results depending on prompt
-* Requires prompt engineering for best performance
+```bash
+./main -m Phi-3-mini-4k-instruct-Q5_K_M.gguf -p "Classify this email: Subject: ... Body: ..."
+```
 
 ---
 
@@ -84,22 +67,39 @@ Spam
 
 ---
 
+## ⚠️ Limitations
+
+* Performance depends on dataset quality
+* May not generalize to unseen domains
+* Sensitive to prompt phrasing
+
+---
+
+## 📦 Files
+
+| File   | Description                |
+| ------ | -------------------------- |
+| *.gguf | Quantized fine-tuned model |
+
+---
+
 ## 📌 Notes
 
-* This is a **base model**, not a fine-tuned classifier
-* For production use, fine-tuning is recommended
+* Optimized for **local inference (CPU-friendly)**
+* Compatible with **llama.cpp and GGUF runtimes**
 
 ---
 
 ## 📜 License
 
-Specify the original Phi-3 license here.
+Same as base model (Phi-3).
 
 ---
 
 ## 🙌 Acknowledgements
 
 * Microsoft for Phi-3
-* llama.cpp for GGUF support
+* Hugging Face
+* llama.cpp
 
----
+---