Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
tags:
|
| 6 |
+
- bio
|
| 7 |
+
- personality
|
| 8 |
+
- tags
|
| 9 |
+
- extraction
|
| 10 |
+
- spiceechat
|
| 11 |
+
- tiny-model
|
| 12 |
+
- work-in-progress
|
| 13 |
+
pipeline_tag: text-generation
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# 🏷️ Bio2Tags-Lite
|
| 17 |
+
|
| 18 |
+
> *The full real README is stuck in traffic. Will arrive tomorrow. Please download on half faith. 🤞*
|
| 19 |
+
|
| 20 |
+
---
|
| 21 |
+
|
| 22 |
+
## 🚧 What's Going On?
|
| 23 |
+
|
| 24 |
+
You're looking at a placeholder because:
|
| 25 |
+
|
| 26 |
+
1. **The real README is in traffic.** LA rush hour. It's bad out there.
|
| 27 |
+
2. **This model was born 30 minutes ago.** It still doesn't know what a 401(k) is.
|
| 28 |
+
3. **I am training like 4 models right now.**
|
| 29 |
+
|
| 30 |
+
---
|
| 31 |
+
|
| 32 |
+
## 🤨 So What Does This Model Actually Do?
|
| 33 |
+
|
| 34 |
+
It reads a dating bio and outputs personality tags.
|
| 35 |
+
|
| 36 |
+
**Input:**
|
| 37 |
+
> *"I'm a retired teacher who gardens, reads history books, and bakes sourdough."*
|
| 38 |
+
|
| 39 |
+
**Output:**
|
| 40 |
+
> *intellectual, family-oriented, gardener, history-buff, old-soul*
|
| 41 |
+
|
| 42 |
+
That's it. No small talk. No life advice. Just tags.
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
+
## 🙏 Why "Download on Half Faith"?
|
| 47 |
+
|
| 48 |
+
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.
|
| 49 |
+
|
| 50 |
+
**We'll get there. Just not today. Today is chaos.**
|
| 51 |
+
|
| 52 |
+
---
|
| 53 |
+
|
| 54 |
+
## ⚡ Quick Test (If You're Brave)
|
| 55 |
+
|
| 56 |
+
```python
|
| 57 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 58 |
+
|
| 59 |
+
model = AutoModelForCausalLM.from_pretrained("SpiceeChat/Bio2Tags-Lite", dtype="auto", device_map="auto")
|
| 60 |
+
tokenizer = AutoTokenizer.from_pretrained("SpiceeChat/Bio2Tags-Lite")
|
| 61 |
+
|
| 62 |
+
bio = "I love hiking at sunrise and brewing craft beer on weekends."
|
| 63 |
+
prompt = f"Extract personality tags from the bio below. Output ONLY comma-separated tags, nothing else.\n\nBio: {bio}\n\nTags:"
|
| 64 |
+
messages = [{"role": "user", "content": prompt}]
|
| 65 |
+
formatted = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
| 66 |
+
inputs = tokenizer(formatted, return_tensors="pt").to(model.device)
|
| 67 |
+
outputs = model.generate(**inputs, max_new_tokens=50, temperature=0.7)
|
| 68 |
+
tags = tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
|
| 69 |
+
print(tags)
|
| 70 |
+
```
|
| 71 |
+
|
| 72 |
+
---
|
| 73 |
+
|
| 74 |
+
## 📍 Status
|
| 75 |
+
|
| 76 |
+
| Thing | Status |
|
| 77 |
+
|-------|--------|
|
| 78 |
+
| Model | 🟡 Works 50% of the time |
|
| 79 |
+
| Real README | 🔴 Stuck in traffic, ETA tomorrow |
|
| 80 |
+
| Developer | 🟡 Running on caffeine and blind faith |
|
| 81 |
+
| Cinder-1.5B | 🟡 Still training on Kaggle |
|
| 82 |
+
| Atom-350M | ⚪ Coming soon |
|
| 83 |
+
| Sleep | 🔴 Not happening |
|
| 84 |
+
|
| 85 |
+
---
|
| 86 |
+
|
| 87 |
+
## 🧠 Part of SpiceeChat
|
| 88 |
+
|
| 89 |
+
Built by **QuantaSparkLabs** for **SpiceeChat** — tools and AI to help people navigate the messy world of dating.
|
| 90 |
+
|
| 91 |
+
- 🌐 [dating-fatigue.com](https://dating-fatigue.com)
|
| 92 |
+
- 🔥 [Cinder-1.5B](https://huggingface.co/SpiceeChat/Cinder-1.5B)
|
| 93 |
+
- 🏷️ [Bio2Tags-Lite](https://huggingface.co/SpiceeChat/Bio2Tags-Lite) ← You are here
|
| 94 |
+
|
| 95 |
+
---
|
| 96 |
+
|
| 97 |
+
<div align="center">
|
| 98 |
+
<sub>🚗 The real README is on the 405. It'll be here tomorrow. Probably.</sub>
|
| 99 |
+
</div>
|