Upload folder using huggingface_hub

README.md

@@ -25,5 +25,5 @@ narrates it, and — within strict evidence gates — recalibrates it.
   re-checked by a code validator (full walk-forward acceptance criteria, 90-day rate limit, hard rails).
 - **MarketGuruLLM** (fine-tuned Qwen2.5-14B) narrates every rebalance and decision in plain English.
 
-The page is **read-only**: live positions, equity vs S&P 500 since inception, agent health, the latest decision
+The page is **read-only**: live positions, the fund vs the S&P 500 since inception, agent health, the latest decision
 with its reasoning, and a dry-run rebalance preview. Backtested results, paper trading — not investment advice.

app.py

@@ -154,7 +154,7 @@ def load_health():
         r = json.load(open(hf_hub_download("s-ttp/marketfm-panel", "deepquant_health.json", repo_type="dataset",
                                            token=tok, force_download=True)))
         icon = {"OK": "🟢", "WATCH": "🟡", "ALERT": "🔴"}
-        lines = [f"### 🩺 Agent health — {icon.get(r.get('overall'), '⚪')} {r.get('overall')} <sub>(checked {r.get('date')})</sub>"]
+        lines = [f"### 🩺 System health — {icon.get(r.get('overall'), '⚪')} {r.get('overall')} <sub>(checked {r.get('date')})</sub>"]
         for c in r.get("checks", []):
             lines.append(f"- {icon.get(c['status'], '⚪')} **{c['name']}** — {c['detail']}")
         try:
@@ -163,13 +163,30 @@ def load_health():
             tr = rr.get("tracking", {})
             inc = rr.get("incumbent", {})
             flag = "🔁" if rr.get("changed") else ("🟡" if rr.get("triggered") else "🟢")
-            lines.append(f"- {flag} **Autonomous recalibration agent** (weekly) — 8-wk gap vs SPY "
-                         f"{tr.get('gap_8w_pp', '—')}pp ({tr.get('pctile_8w', '—')}th pctile of backtest) · "
-                         f"implementation drift {tr.get('impl_drift_pp', '—')}pp · {rr.get('decision', '')[:160]}")
-            lines.append(f"<sub>Live strategy config v{inc.get('version', '?')}: signal {inc.get('signal')} · regime "
-                         f"{inc.get('regime')} · defense {inc.get('defense')} · cadence {inc.get('cadence_weeks')}w. "
-                         f"Recalibration only adopts a variant that beats the incumbent on a full 20-year walk-forward, "
-                         f"net of cost, across folds — never a re-fit to recent weeks; max one change per 90 days.</sub>")
+            SIGMAP = {"n_mom_12_1": "momentum (12-1)", "n_roe": "ROE", "n_fcf_yield": "FCF yield",
+                      "n_roa": "ROA", "n_vol_13w": "low volatility"}
+            DEFMAP = {"mom60_agg20_gld20": "60% stocks + 20% bonds + 20% gold",
+                      "mom40_agg30_gld30": "40% stocks + 30% bonds + 30% gold",
+                      "agg50_gld50": "50% bonds + 50% gold", "lowvol": "low-volatility stocks"}
+            dec = rr.get("decision", "")
+            if dec.startswith("[agentic] "):
+                dec = dec[10:]
+            act, _, why = dec.partition(":")
+            why = (why.strip()
+                   .replace("adopt_rate_limit_ok=false", "a strategy change is currently rate-limited")
+                   .replace("ADOPT is infeasible", "no change is possible this cycle")
+                   .replace("validator-approved candidates", "validated better variants"))
+            cut = why.find(". ")
+            why = (why[:cut + 1] if 0 < cut <= 230 else (why[:230].rsplit(" ", 1)[0] + "…" if len(why) > 230 else why))
+            lines.append(f"- {flag} **Autonomous recalibration agent** (weekly, Analyst → Critic → PM) — latest decision "
+                         f"**{act.strip()}**: {why}")
+            sig = " + ".join(SIGMAP.get(x, x) for x in (inc.get("signal") or []))
+            lines.append(f"<sub>Tracking: 8-week gap vs S&P {tr.get('gap_8w_pp', '—')}pp — the {tr.get('pctile_8w', '—')}th "
+                         f"percentile of the strategy's 20-year history · implementation drift {tr.get('impl_drift_pp', '—')}pp. "
+                         f"Live strategy v{inc.get('version', '?')}: {sig} selection · {inc.get('regime', '?')} trend rule · "
+                         f"defense = {DEFMAP.get(inc.get('defense'), inc.get('defense'))} · rebalanced every "
+                         f"{inc.get('cadence_weeks', 1)} week(s). A change is adopted only if it beats the incumbent on the "
+                         f"full 20-year walk-forward, net of costs — never a re-fit to recent weeks; at most one change per 90 days.</sub>")
         except Exception:
             pass
         lines.append("<sub>Deterministic daily checks: live tracking gap located in the 20-year backtest distribution, "
@@ -268,41 +285,17 @@ with gr.Blocks(title="DeepQuant — live paper monitor", theme=gr.themes.Soft())
         "S&P 500's ~11%**, higher Sharpe, −35% max drawdown vs the index's −49%.\n"
         "<sub>Fully autonomous, paper trading, this page is read-only. Backtested results — past performance ≠ future. "
         "Not investment advice.</sub>")
-    with gr.Accordion("🧬 Why hybrid? Who decides what — and why", open=False):
-        gr.Markdown(
-            "The lesson from building this system: **neither pure determinism nor pure agency is enough.** Rules can't "
-            "weigh ambiguous evidence; models can't be allowed to improvise with money. So each job goes to the framework "
-            "that has *earned* it:\n\n"
-            "| Responsibility | Framework | Why |\n|---|---|---|\n"
-            "| Stock ranking & target weights | **Deterministic** (MarketFM factors + fixed rules) | Validated on a 20-year "
-            "walk-forward, survivorship-free, net of costs — and reproducible to the penny |\n"
-            "| Order execution & risk rails | **Deterministic** | Execution is arithmetic — the last place you want a model "
-            "hallucinating. The rails (long-only, caps, no leverage) are code no one can override |\n"
-            "| Daily monitoring | **Deterministic** | Statistical checks against the strategy's own history — e.g. \"is this "
-            "drawdown inside the envelope of 280 historical 8-week windows?\" |\n"
-            "| Weekly assessment & judgment | **Agentic** — a role-separated crew (Analyst → adversarial Critic → PM) on "
-            "Claude Opus 5 | Weighing noisy, conflicting evidence is judgment; separating the roles makes each argument "
-            "survive an adversarial challenge before any decision |\n"
-            "| Strategy changes | **Hybrid** | The crew may adopt *only* variants that beat the incumbent across a bounded "
-            "menu (selection signal × regime rule × defense sleeve × cadence) on the full 20-year walk-forward, net of cost — "
-            "re-checked by a code validator; max one change per 90 days. Trigger on recent data, validate on all data |\n"
-            "| Emergency brake | **Hybrid** | The crew can pause execution up to 2 weeks on operational anomalies; holding "
-            "the validated book is always the safe default — and any AI failure automatically resolves to HOLD |\n"
-            "| Explanations | **Agentic** (MarketGuruLLM) | Language is what language models are for |\n\n"
-            "Every weekly decision is logged in full — the Analyst's assessment, the Critic's objections, the PM's ruling "
-            "and the evidence they cite — and the current strategy configuration is versioned, so the fund's entire "
-            "decision history is auditable. **AI proposes and judges; validated evidence admits; code enforces.**")
     with gr.Tabs():
         with gr.Tab("📊 Live account"):
             stats = gr.Markdown()
             health = gr.Markdown()
             rebal = gr.Markdown()
-            gr.Markdown("### Holdings")
-            tbl = gr.Dataframe(wrap=True, interactive=False)
-            gr.Markdown("### Equity vs S&P 500")
+            gr.Markdown("### 📈 DeepQuant fund vs S&P 500")
             plot = gr.LinePlot(x="date", y="value", color="series", height=320,
                                y_title="Account value ($)", x_title="")
             note = gr.Markdown(visible=False)
+            gr.Markdown("### 📋 Current holdings")
+            tbl = gr.Dataframe(wrap=True, interactive=False)
             with gr.Row():
                 updated = gr.Markdown("_loading…_")
                 btn = gr.Button("↻ Refresh now", scale=0)
@@ -319,6 +312,31 @@ with gr.Blocks(title="DeepQuant — live paper monitor", theme=gr.themes.Soft())
                 with gr.Column():
                     gr.Markdown("**Proposed orders** — sells first, then buys")
                     ord_tbl = gr.Dataframe(interactive=False, wrap=True)
+    with gr.Accordion("🧬 Why hybrid? Who decides what — and why", open=False):
+        gr.Markdown(
+            "The lesson from building this system: **neither pure determinism nor pure agency is enough.** Rules can't "
+            "weigh ambiguous evidence; models can't be allowed to improvise with money. So each job goes to the framework "
+            "that has *earned* it:\n\n"
+            "| Responsibility | Framework | Why |\n|---|---|---|\n"
+            "| Stock ranking & target weights | **Deterministic** (MarketFM factors + fixed rules) | Validated on a 20-year "
+            "walk-forward, survivorship-free, net of costs — and reproducible to the penny |\n"
+            "| Order execution & risk rails | **Deterministic** | Execution is arithmetic — the last place you want a model "
+            "hallucinating. The rails (long-only, caps, no leverage) are code no one can override |\n"
+            "| Daily monitoring | **Deterministic** | Statistical checks against the strategy's own history — e.g. \"is this "
+            "drawdown inside the envelope of 280 historical 8-week windows?\" |\n"
+            "| Weekly assessment & judgment | **Agentic** — a role-separated crew (Analyst → adversarial Critic → PM) on "
+            "Claude Opus 5 | Weighing noisy, conflicting evidence is judgment; separating the roles makes each argument "
+            "survive an adversarial challenge before any decision |\n"
+            "| Strategy changes | **Hybrid** | The crew may adopt *only* variants that beat the incumbent across a bounded "
+            "menu (selection signal × regime rule × defense sleeve × cadence) on the full 20-year walk-forward, net of cost — "
+            "re-checked by a code validator; max one change per 90 days. Trigger on recent data, validate on all data |\n"
+            "| Emergency brake | **Hybrid** | The crew can pause execution up to 2 weeks on operational anomalies; holding "
+            "the validated book is always the safe default — and any AI failure automatically resolves to HOLD |\n"
+            "| Explanations | **Agentic** (MarketGuruLLM) | Language is what language models are for |\n\n"
+            "Every weekly decision is logged in full — the Analyst's assessment, the Critic's objections, the PM's ruling "
+            "and the evidence they cite — and the current strategy configuration is versioned, so the fund's entire "
+            "decision history is auditable. **AI proposes and judges; validated evidence admits; code enforces.**")
+
     timer = gr.Timer(1800)
 
     acct_outs = [stats, tbl, plot, note, updated]