Create app.py

app.py

@@ -0,0 +1,212 @@
+import gradio as gr
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import seaborn as sns
+import logging
+import requests
+from typing import List, Tuple, Dict, Optional
+from datasets import load_dataset
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import LabelEncoder
+
+# --- CONFIG & LOGGING ---
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+REPO_CONFIG = {
+    "Core (Clean)": {
+        "repo": "QSBench/QSBench-Core-v1.0.0-demo",
+        "meta_url": "https://huggingface.co/datasets/QSBench/QSBench-Core-v1.0.0-demo/raw/metadata/meta/meta.json",
+        "report_url": "https://huggingface.co/datasets/QSBench/QSBench-Core-v1.0.0-demo/raw/metadata/meta/report.json"
+    },
+    "Depolarizing Noise": {
+        "repo": "QSBench/QSBench-Depolarizing-Demo-v1.0.0",
+        "meta_url": "https://huggingface.co/datasets/QSBench/QSBench-Depolarizing-Demo-v1.0.0/raw/meta/meta/meta.json",
+        "report_url": "https://huggingface.co/datasets/QSBench/QSBench-Depolarizing-Demo-v1.0.0/raw/meta/meta/report.json"
+    },
+    "Amplitude Damping": {
+        "repo": "QSBench/QSBench-Amplitude-v1.0.0-demo",
+        "meta_url": "https://huggingface.co/datasets/QSBench/QSBench-Amplitude-v1.0.0-demo/raw/meta/meta/meta.json",
+        "report_url": "https://huggingface.co/datasets/QSBench/QSBench-Amplitude-v1.0.0-demo/raw/meta/meta/report.json"
+    },
+    "Transpilation (10q)": {
+        "repo": "QSBench/QSBench-Transpilation-v1.0.0-demo",
+        "meta_url": "https://huggingface.co/datasets/QSBench/QSBench-Transpilation-v1.0.0-demo/raw/meta/meta/meta.json",
+        "report_url": "https://huggingface.co/datasets/QSBench/QSBench-Transpilation-v1.0.0-demo/raw/meta/meta/report.json"
+    }
+}
+
+# Columns that are NOT features
+NON_FEATURE_COLS = {
+    "sample_id", "sample_seed", "circuit_hash", "split", "circuit_qasm", 
+    "qasm_raw", "qasm_transpiled", "circuit_type_resolved", "circuit_type_requested",
+    "noise_type", "noise_prob", "observable_bases", "observable_mode", "backend_device", 
+    "precision_mode", "circuit_signature", "entanglement", "shots", "gpu_requested", "gpu_available"
+}
+
+_ASSET_CACHE = {}
+
+def load_all_assets(key: str) -> Dict:
+    if key not in _ASSET_CACHE:
+        logger.info(f"Fetching {key}...")
+        ds = load_dataset(REPO_CONFIG[key]["repo"])
+        meta = requests.get(REPO_CONFIG[key]["meta_url"]).json()
+        report = requests.get(REPO_CONFIG[key]["report_url"]).json()
+        _ASSET_CACHE[key] = {"df": pd.DataFrame(ds["train"]), "meta": meta, "report": report}
+    return _ASSET_CACHE[key]
+
+# --- UI LOGIC ---
+
+def load_guide_content():
+    """Reads the content of GUIDE.md from the local directory."""
+    try:
+        with open("GUIDE.md", "r", encoding="utf-8") as f:
+            return f.read()
+    except FileNotFoundError:
+        return "### ⚠️ Error: GUIDE.md not found. Please ensure it is in the root directory."
+
+def sync_ml_metrics(ds_name: str):
+    """Extracts numerical features available for classification."""
+    assets = load_all_assets(ds_name)
+    df = assets["df"]
+    numeric_cols = df.select_dtypes(include=[np.number]).columns.tolist()
+    
+    valid_features = [
+        c for c in numeric_cols 
+        if c not in NON_FEATURE_COLS 
+        and not any(prefix in c for prefix in ["ideal_", "noisy_", "error_", "sign_"])
+    ]
+    
+    # Pre-select logical structural indicators
+    defaults = [f for f in ["gate_entropy", "meyer_wallach", "adjacency", "depth", "cx_count"] if f in valid_features]
+    return gr.update(choices=valid_features, value=defaults or valid_features[:5])
+
+def train_classifier(ds_name: str, features: List[str]):
+    """Trains a Classifier to identify the Circuit Family based on topology."""
+    if not features: return None, "### ❌ Error: No features selected."
+    assets = load_all_assets(ds_name)
+    df = assets["df"]
+    
+    target_col = "circuit_type_requested"
+    if target_col not in df.columns:
+        return None, f"### ❌ Error: Target column '{target_col}' not found."
+
+    # Data Cleaning
+    train_df = df.dropna(subset=features + [target_col])
+    X = train_df[features]
+    y = train_df[target_col]
+
+    # Encoding targets
+    le = LabelEncoder()
+    y_encoded = le.fit_transform(y)
+    class_names = le.classes_
+
+    X_train, X_test, y_train, y_test = train_test_split(X, y_encoded, test_size=0.2, random_state=42)
+
+    # Classification Model
+    clf = RandomForestClassifier(n_estimators=100, max_depth=12, n_jobs=-1, random_state=42)
+    clf.fit(X_train, y_train)
+    preds = clf.predict(X_test)
+
+    # Metrics
+    acc = accuracy_score(y_test, preds)
+    
+    # Visualization
+    sns.set_theme(style="whitegrid", context="talk")
+    fig, axes = plt.subplots(1, 2, figsize=(20, 8))
+
+    # 1. Confusion Matrix
+    cm = confusion_matrix(y_test, preds)
+    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', 
+                xticklabels=class_names, yticklabels=class_names, ax=axes[0], cbar=False)
+    axes[0].set_title(f"Confusion Matrix (Accuracy: {acc:.2%})")
+    axes[0].set_xlabel("Predicted Family")
+    axes[0].set_ylabel("Actual Family")
+
+    # 2. Feature Importance
+    importances = clf.feature_importances_
+    indices = np.argsort(importances)[-10:] # Top 10
+    axes[1].barh([features[i] for i in indices], importances[indices], color='#16a085')
+    axes[1].set_title("Top Structural Discriminators")
+
+    plt.tight_layout()
+    
+    report_dict = classification_report(y_test, preds, target_names=class_names)
+    summary = f"### 🏆 Classification Results\n**Overall Accuracy:** {acc:.2%}\n\n**Detailed Report:**\n```\n{report_dict}\n```"
+    
+    return fig, summary
+
+def update_explorer(ds_name: str, split_name: str):
+    """Updates the data view for the Explorer tab."""
+    assets = load_all_assets(ds_name)
+    df = assets["df"]
+    unique_splits = df["split"].unique().tolist() if "split" in df.columns else ["train"]
+    
+    if "split" in df.columns:
+        filtered_df = df[df["split"] == split_name]
+        if filtered_df.empty:
+            split_name = unique_splits[0]
+            filtered_df = df[df["split"] == split_name]
+    else:
+        filtered_df = df
+
+    display_df = filtered_df.head(10)
+    raw = display_df["qasm_raw"].iloc[0] if "qasm_raw" in display_df.columns and not display_df.empty else "// N/A"
+    tr = display_df["qasm_transpiled"].iloc[0] if "qasm_transpiled" in display_df.columns and not display_df.empty else "// N/A"
+    
+    return gr.update(choices=unique_splits, value=split_name), display_df, raw, tr, f"### 📋 {ds_name} Explorer"
+
+# --- INTERFACE ---
+with gr.Blocks(theme=gr.themes.Soft(), title="QSBench Classifier") as demo:
+    gr.Markdown("# 🌌 QSBench: Circuit Family Classifier")
+    gr.Markdown("Identify circuit types (QFT, HEA, RANDOM, etc.) using high-level structural complexity metrics.")
+    
+    with gr.Tabs():
+        with gr.TabItem("🔎 Dataset Explorer"):
+            meta_txt = gr.Markdown("### Loading...")
+            with gr.Row():
+                ds_sel = gr.Dropdown(list(REPO_CONFIG.keys()), value="Core (Clean)", label="Dataset Type")
+                sp_sel = gr.Dropdown(["train"], value="train", label="Subset (Split)")
+            data_view = gr.Dataframe(interactive=False)
+            with gr.Row():
+                c_raw = gr.Code(label="Original QASM (Logic)", language="python")
+                c_tr = gr.Code(label="Transpiled QASM (Hardware-ready)", language="python")
+
+        with gr.TabItem("🧠 Classification Model"):
+            gr.Markdown("Predict the **Circuit Family** by analyzing topology signatures.")
+            with gr.Row():
+                with gr.Column(scale=1):
+                    ml_ds_sel = gr.Dropdown(list(REPO_CONFIG.keys()), value="Core (Clean)", label="Environment")
+                    ml_feat_sel = gr.CheckboxGroup(label="Structural Features", choices=[])
+                    train_btn = gr.Button("Run Classification", variant="primary")
+                with gr.Column(scale=2):
+                    p_out = gr.Plot()
+                    t_out = gr.Markdown()
+
+        with gr.TabItem("📖 User Guide"):
+            meth_md = gr.Markdown(value=load_guide_content())
+
+    gr.Markdown(f"""
+---
+### 🔗 Project Resources
+[**🌐 Website**](https://qsbench.github.io) | [**🤗 Hugging Face**](https://huggingface.co/QSBench) | [**💻 GitHub**](https://github.com/QSBench)
+    """)
+
+    # --- EVENTS ---
+    # Explorer events
+    ds_sel.change(update_explorer, [ds_sel, sp_sel], [sp_sel, data_view, c_raw, c_tr, meta_txt])
+    sp_sel.change(update_explorer, [ds_sel, sp_sel], [sp_sel, data_view, c_raw, c_tr, meta_txt])
+    
+    # ML events
+    ml_ds_sel.change(sync_ml_metrics, [ml_ds_sel], [ml_feat_sel])
+    train_btn.click(train_classifier, [ml_ds_sel, ml_feat_sel], [p_out, t_out])
+    
+    # Initial Load
+    demo.load(update_explorer, [ds_sel, sp_sel], [sp_sel, data_view, c_raw, c_tr, meta_txt])
+    demo.load(sync_ml_metrics, [ml_ds_sel], [ml_feat_sel])
+
+if __name__ == "__main__":
+    demo.launch()