Update README.md

README.md

@@ -1,98 +1,160 @@
 ---
 license: apache-2.0
 language:
-  - en
+- en
 tags:
-  - bio
-  - personality
-  - tags
-  - extraction
-  - spiceechat
-  - tiny-model
-  - work-in-progress
+- bio-to-tags
+- tag-generation
+- smollm2
+- text-generation
+- personality
+- interests
+- spiceechat
 pipeline_tag: text-generation
+library_name: transformers
+---
+
+<p align="center">
+  <img src="https://huggingface.co/SpiceeChat/Bio2Tags-Qwen3.5-4B-SFT/resolve/main/Spiceechat.png" 
+       alt="SpiceeChat" 
+       width="1100" 
+       height="1000" 
+       style="border-radius: 50%; object-fit: cover;">
+</p>
+
+<p align="center">
+  <a href="https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct"><img src="https://img.shields.io/badge/SmolLM2-360M-blue?logo=huggingface" alt="SmolLM2"></a>
+  <a href="https://github.com/unslothai/unsloth"><img src="https://img.shields.io/badge/Fine‑Tuned-QLoRA-green" alt="QLoRA"></a>
+  <a href="https://huggingface.co/SpiceeChat"><img src="https://img.shields.io/badge/SpiceeChat-🔥-orange" alt="SpiceeChat"></a>
+  <a href="https://www.apache.org/licenses/LICENSE-2.0"><img src="https://img.shields.io/badge/License-Apache%202.0-yellow" alt="License"></a>
+</p>
+
 ---
 
 # 🏷️ Bio2Tags-Lite
 
-> *The full real README is stuck in traffic. Will arrive tomorrow. Please download on half faith. 🤞*
+**Because reading between the lines shouldn't require a psychology degree.**
 
----
+Bio2Tags-Lite is a fine-tuned SmolLM2-360M model that reads personal biographies and returns clean, structured personality tags. Feed it a dating bio, a LinkedIn summary, or whatever someone wrote about themselves at 2am — it'll tell you what kind of person they actually are.
+
+No rambling. No fluff. Just tags.
 
-## 🚧 What's Going On?
+---
 
-You're looking at a placeholder because:
+## ✨ Features
 
-1. **The real README is in traffic.** LA rush hour. It's bad out there.
-2. **This model was born 30 minutes ago.** It still doesn't know what a 401(k) is.
-3. **I am training like 4 models right now.** 
+- **Lightweight**: 360M parameters — runs on hardware that would make a gamer cry
+- **Fast**: Inference in milliseconds, because nobody has time to wait
+- **Structured Output**: Clean comma-separated tags, every time
+- **Plug & Play**: Works with Transformers out of the box, no PhD required
+- **SpiceeChat Pipeline**: Pairs with Cinder-1.5B like peanut butter and heartbreak
 
 ---
 
-## 🤨 So What Does This Model Actually Do?
+## 🧪 Example
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained(
+    "SpiceeChat/Bio2Tags-Lite",
+    torch_dtype="auto",
+    device_map="auto",
+)
+tokenizer = AutoTokenizer.from_pretrained("SpiceeChat/Bio2Tags-Lite")
+
+def get_tags(bio):
+    prompt = f"Extract personality tags from the bio below. Output ONLY comma-separated tags, nothing else.\n\nBio: {bio}\n\nTags:"
+    messages = [{"role": "user", "content": prompt}]
+    formatted = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    inputs = tokenizer(formatted, return_tensors="pt").to(model.device)
+    outputs = model.generate(**inputs, max_new_tokens=50, temperature=0.7, do_sample=True)
+    return tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True).strip()
+
+# Try it
+print(get_tags("I love hiking at dawn, painting watercolors, and deep conversations about philosophy."))
+# Output: nature-lover, artist, intellectual, deep-thinker
+```
 
-It reads a dating bio and outputs personality tags.
+---
 
-**Input:**
-> *"I'm a retired teacher who gardens, reads history books, and bakes sourdough."*
+## 📊 Sample Outputs
 
-**Output:**
-> *intellectual, family-oriented, gardener, history-buff, old-soul*
+| Bio | Tags |
+|-----|------|
+| "I'm a software engineer who loves late-night coding and playing jazz piano." | tech-savvy, creative, night-owl, music-enthusiast, artistic |
+| "I spend my weekends trail running and evenings reading classic literature." | adventurous, nature-lover, bookworm, intellectual, quiet |
+| "I'm a retired teacher who gardens, reads history books, and bakes sourdough." | intellectual, family-oriented, gardener, history-buff, old-soul |
+| "As a digital nomad, my office changes weekly — from Bali cafes to Alpine cabins." | adventurous, creative, digital-nomad, spontaneous, tech-savvy |
 
-That's it. No small talk. No life advice. Just tags.
+*(Yes, the sourdough one is a stereotype. Yes, it's also always accurate.)*
 
 ---
 
-## 🙏 Why "Download on Half Faith"?
+## 📦 Installation
 
-Because it works about 50% of the time right now. The other 50%? Let's call it "creative interpretation." We're working on it. The 360M version is much better than the original 135M prototype that once tagged everyone as "adventurous, creative, empathetic" regardless of input.
+```bash
+pip install transformers torch accelerate
+```
 
-**We'll get there. Just not today. Today is chaos.**
+That's it. No ritual sacrifices, no config files, no Stack Overflow rabbit holes.
 
 ---
 
-## ⚡ Quick Test (If You're Brave)
+## 🎯 Use Cases
 
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+- **Dating Apps**: Tag user bios automatically for smarter matching — because "I like long walks on the beach" means something very different than "I like long walks on the beach at 3am alone"
+- **Social Media**: Generate relevant hashtags from profile descriptions
+- **Recommender Systems**: Build personality-based recommendation engines
+- **Content Analysis**: Extract structured metadata from unstructured text
+- **SpiceeChat Pipeline**: Feed extracted tags into Cinder-1.5B for personalized compatibility advice
 
-model = AutoModelForCausalLM.from_pretrained("SpiceeChat/Bio2Tags-Lite", dtype="auto", device_map="auto")
-tokenizer = AutoTokenizer.from_pretrained("SpiceeChat/Bio2Tags-Lite")
+---
 
-bio = "I love hiking at sunrise and brewing craft beer on weekends."
-prompt = f"Extract personality tags from the bio below. Output ONLY comma-separated tags, nothing else.\n\nBio: {bio}\n\nTags:"
-messages = [{"role": "user", "content": prompt}]
-formatted = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-inputs = tokenizer(formatted, return_tensors="pt").to(model.device)
-outputs = model.generate(**inputs, max_new_tokens=50, temperature=0.7)
-tags = tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
-print(tags)
-```
+## 🛠️ Technical Details
+
+| Detail | Value |
+|--------|-------|
+| **Base Model** | [SmolLM2-360M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct) |
+| **Fine-tuning Method** | QLoRA (4-bit quantization, rank-16 adapters) |
+| **Training Framework** | Unsloth |
+| **Training Data** | 1,387 hand-crafted (bio, tags) pairs |
+| **Epochs** | 3 |
+| **Learning Rate** | 1e-4 |
+| **Sequence Length** | 512 tokens |
+| **Hardware Used** | Google Colab T4 (free tier — yes, really) |
+| **Final Size** | 724 MB (FP16) |
+| **Min VRAM Required** | ~1.5 GB |
 
 ---
 
-## 📍 Status
+## ⚠️ Limitations
 
-| Thing | Status |
-|-------|--------|
-| Model | 🟡 Works 50% of the time |
-| Real README | 🔴 Stuck in traffic, ETA tomorrow |
-| Developer | 🟡 Running on caffeine and blind faith |
-| Cinder-1.5B | 🟡 Still training on Kaggle |
-| Sleep | 🔴 Not happening |
+- **English only**: Other languages may produce results ranging from "creative" to "confidently wrong"
+- **Training data size**: 1,387 examples is a solid start — more data is always on the roadmap
+- **Tag granularity**: Captures the salient stuff, not every quirk (the model can't detect if someone is secretly obsessed with true crime podcasts)
+- **Edge cases**: Very short bios, emoji-heavy text, or deeply abstract descriptions may surprise you
 
 ---
 
-## 🧠 Part of SpiceeChat
+## 🧠 Part of the SpiceeChat Ecosystem
+
+Bio2Tags-Lite is a core component of the SpiceeChat AI pipeline:
+
+- 🏷️ **Bio2Tags-Lite** → Extracts personality tags from bios
+- 🔥 **[Cinder-1.5B](https://huggingface.co/SpiceeChat/Cinder-1.5B)** → Personalized dating advice powered by those tags
+- 🌐 **[dating-fatigue.com](https://dating-fatigue.com)** → Live tools for real humans trying to find real love
+
+---
 
-Built for **SpiceeChat** — tools and AI to help people navigate the messy world of dating.
+## 📜 License
 
-- 🌐 [dating-fatigue.com](https://dating-fatigue.com)
-- 🔥 [Cinder-1.5B](https://huggingface.co/SpiceeChat/Cinder-1.5B)
-- 🏷️ [Bio2Tags-Lite](https://huggingface.co/SpiceeChat/Bio2Tags-Lite) ← You are here
+Apache 2.0 — use it, modify it, ship it. Just give SpiceeChat a nod.
 
 ---
 
 <div align="center">
-  <sub>🚗 The real README is on the 405. It'll be here tomorrow. Probably.</sub>
-</div>
+  <sub>Built with ❤️ by <b>SpiceeChat</b></sub>
+  <br>
+  <sub>🔗 <a href="https://huggingface.co/SpiceeChat">huggingface.co/SpiceeChat</a></sub>
+</div>