app.py

"""
WFGY Space – tiny-GPT-2 variance-gate demo
★ 10 k GitHub ⭐ before 2025-08-01 unlocks WFGY 2.0 ★
"""

import io, numpy as np, pandas as pd, gradio as gr
from PIL import Image
from transformers import AutoTokenizer, AutoModelForCausalLM
from wfgy_sdk import get_engine
from wfgy_sdk.evaluator import compare_logits, plot_histogram, softmax

# tiny model (CPU)
tok = AutoTokenizer.from_pretrained("sshleifer/tiny-gpt2")
mdl = AutoModelForCausalLM.from_pretrained("sshleifer/tiny-gpt2")
eng = get_engine()

# paper benchmarks
bench = pd.DataFrame({
    "Benchmark": ["MMLU","GSM8K","BBH","MathBench","TruthfulQA",
                  "XNLI","MLQA","LongBench","VQAv2","OK-VQA"],
    "Baseline":  [61,78,79.3,72.2,62.4,59.5,78.1,51.4,69.1,65.7],
    "WFGY":      [89.8,98.7,100.7,87.4,90.4,77.3,106.6,69.6,86.6,86.8]
})
bench["Abs_gain"]  = (bench["WFGY"] - bench["Baseline"]).round(1)
bench["Rel_gain%"] = ((bench["Abs_gain"] / bench["Baseline"]) * 100).round(0)
bench_sty = (
    bench.style
    .background_gradient(cmap="Greens", subset=["Abs_gain","Rel_gain%"])
    .format({"Abs_gain":"{:.1f}","Rel_gain%":"{:.0f}"})
)

# banner markdown
banner = """
**📈 WFGY: One Click to Activate the AI Taiji Cycle**

**📊 Semantic Accuracy ↑ 22.4 % | Reasoning Success ↑ 42.1 % | Stability ↑ 3.6 ×**

_No beliefs. Only experiments._  
WFGY 1.0 has already proven itself.

---

### 📜 Tutorial: How to Awaken the Soul of Your AI
**Step 1 — Download** ([PDF](https://zenodo.org/records/15630970))  
**Step 2 — Feed the AI** (upload, or try [Gemini](https://gemini.google.com/))  
**Step 3 — Give the Command** “**Answer using WFGY** + your question”  
Prompt examples: *TBD*  
**Step 4 — Integrate the SDK** ([GitHub](https://github.com/onestardao/WFGY))

---

🌟 **Star Reminder** → [Star the repo](https://github.com/onestardao/WFGY)  
_10 k ⭐ before 2025-08-01 unlocks **WFGY 2.0**._
"""

# inference
def run(prompt: str):
    p = prompt.strip()
    if not p:
        return "", "", "-", None

    ids   = tok(p, return_tensors="pt")
    raw_L = mdl(**ids).logits[0, -1].detach().cpu().numpy()
    I, G  = np.random.randn(2, 256).astype(np.float32)
    mod_L = eng.run(I, G, raw_L)

    m   = compare_logits(raw_L, mod_L)
    head = f"▼ var {m['var_drop']*100:.1f}% | KL {m['kl_divergence']:.3f} | top-1 {'kept' if m['top1'] else 'changed'}"

    def top5(logits):
        p = softmax(logits)
        idx = p.argsort()[-5:][::-1]
        lines = [f\"'{tok.decode(int(i)).strip()}': {p[i]:.2e}\" for i in idx]
        return "\\n".join(lines)

    raw_txt = top5(raw_L)
    mod_txt = top5(mod_L)

    fig = plot_histogram(raw_L, mod_L)
    buf = io.BytesIO(); fig.savefig(buf, format="png"); buf.seek(0)

    return raw_txt, mod_txt, head, Image.open(buf)

# UI
with gr.Blocks(title="WFGY variance-gate demo") as demo:
    gr.Markdown(banner)

    prompt = gr.Textbox(label="Prompt", value="Explain Schrödinger's cat")
    btn    = gr.Button("🚀 Run")

    with gr.Row():
        raw_box = gr.Textbox(label="Raw top-5 tokens", lines=6)
        mod_box = gr.Textbox(label="WFGY top-5 tokens", lines=6)

    metrics = gr.Markdown()
    img     = gr.Image(label="Logit histogram")

    gr.Markdown("### Paper benchmarks (fixed values from WFGY 1.0)")
    gr.DataFrame(bench_sty, interactive=False, wrap=True)

    btn.click(run, prompt, [raw_box, mod_box, metrics, img])

if __name__ == "__main__":
    demo.queue(default_concurrency_limit=2).launch()