{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6ebab6ca",
   "metadata": {
    "papermill": {
     "duration": 0.004567,
     "end_time": "2026-06-13T02:38:35.471970+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:35.467403+00:00",
     "status": "completed"
    }
   },
   "source": [
    "# Diffusion-TS: Interpretable Diffusion with Conditional Generation\n",
    "\n",
    "**Chapter 5: Synthetic Data Generation**\n",
    "**Section Reference**: Section 5.5 (Diffusion Models for Financial Time Series)\n",
    "\n",
    "**Docker image**: `ml4t-gpu`\n",
    "\n",
    "> **GPU recommended**: This notebook trains models with PyTorch/CUDA. It will run on CPU\n",
    "> but training may be very slow. For GPU acceleration:\n",
    "> ```bash\n",
    "> docker compose run --rm ml4t-gpu python 05_synthetic_data/05_diffusion_ts.py\n",
    "> ```\n",
    "\n",
    "\n",
    "## Purpose\n",
    "\n",
    "This notebook implements **Diffusion-TS** (Yuan & Qiao, ICLR 2024), a diffusion\n",
    "model that decomposes the denoising prediction into **trend** (polynomial regression)\n",
    "and **seasonal** (Fourier basis) components. This interpretable structure encourages\n",
    "the model to separate slow drift from periodic patterns, analogous to classical STL\n",
    "decomposition but learned end-to-end within the diffusion framework.\n",
    "\n",
    "Unlike vanilla diffusion models that predict noise $\\varepsilon$, Diffusion-TS\n",
    "predicts $x_0$ directly. The Fourier-domain loss further regularizes spectral\n",
    "fidelity -- critical for preserving autocorrelation structure in financial returns.\n",
    "\n",
    "## Learning Objectives\n",
    "\n",
    "By completing this notebook, you will:\n",
    "- Implement a diffusion model with **interpretable trend+seasonal decomposition**\n",
    "- Train with a combined **time-domain + Fourier-domain** loss\n",
    "- Use **DDIM** fast sampling to reduce generation from 500 to 50 reverse steps\n",
    "- Build a **regime classifier** on noised data and apply **classifier guidance**\n",
    "  to generate regime-conditional synthetic returns\n",
    "- Evaluate unconditional and conditional generation quality\n",
    "\n",
    "## Cross-References\n",
    "\n",
    "- **Upstream**: ETF Universe loader (`data`)\n",
    "- **Downstream**: Regime-conditioned synthetic data for stress testing (Ch 20)\n",
    "- **Book**: Section 5.5 discusses diffusion + conditional generation\n",
    "- **Related**: [`02_tailgan_tail_risk`](02_tailgan_tail_risk.ipynb) (GAN), [`03_sigcwgan_signatures`](03_sigcwgan_signatures.ipynb) (GAN)\n",
    "\n",
    "---\n",
    "\n",
    "## From GANs to Interpretable Diffusion\n",
    "\n",
    "GANs face training instability, mode collapse, and limited interpretability.\n",
    "Diffusion models solve the first two by replacing adversarial training with a\n",
    "simple regression objective. Diffusion-TS goes further by decomposing the\n",
    "denoising network's output into components with known semantics:\n",
    "\n",
    "$$\\hat{x}_0 = \\text{Trend}(z) + \\text{Season}(z) + \\text{Residual}(z)$$\n",
    "\n",
    "where $z$ is the latent representation at diffusion step $t$. The trend block\n",
    "uses polynomial regression, the seasonal block selects top-$k$ Fourier modes,\n",
    "and the residual captures everything else. This makes the model's behavior\n",
    "inspectable -- you can visualize what the model attributes to drift versus\n",
    "cyclical patterns.\n",
    "\n",
    "## References\n",
    "\n",
    "- **Paper**: Yuan, X. & Qiao, Y. (2024). \"Diffusion-TS: Interpretable Diffusion\n",
    "  for General Time Series Generation.\" ICLR 2024.\n",
    "- **Code**: https://github.com/Y-debug-sys/Diffusion-TS\n",
    "\n",
    "## Key Adaptation Decisions\n",
    "\n",
    "1. **No clamp(-1, 1)**: The original code clamps predicted $x_0$ to [-1, 1] for\n",
    "   image-like data. Financial returns are StandardScaler-normalized (unbounded),\n",
    "   so we remove the clamp entirely.\n",
    "2. **Predict $x_0$** (not noise): The model outputs $\\hat{x}_0 = \\text{trend} +\n",
    "   \\text{season}$, then derives noise analytically. This pairs naturally with the\n",
    "   decomposition.\n",
    "3. **Fourier loss preserved**: The frequency-domain regularizer matches spectral\n",
    "   properties -- essential for autocorrelation fidelity in returns.\n",
    "4. **Classifier guidance**: A separate Transformer classifier trained on noised\n",
    "   sequences enables regime-conditional generation at sampling time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "570812f8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:35.479751Z",
     "iopub.status.busy": "2026-06-13T02:38:35.479643Z",
     "iopub.status.idle": "2026-06-13T02:38:38.332327Z",
     "shell.execute_reply": "2026-06-13T02:38:38.331639Z"
    },
    "papermill": {
     "duration": 2.857374,
     "end_time": "2026-06-13T02:38:38.332818+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:35.475444+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "\"\"\"Diffusion-TS: Interpretable Diffusion with Conditional Generation.\"\"\"\n",
    "\n",
    "import json\n",
    "import math\n",
    "import warnings\n",
    "from copy import deepcopy\n",
    "from datetime import UTC, datetime\n",
    "from pathlib import Path\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import polars as pl\n",
    "import seaborn as sns\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from einops import rearrange, reduce, repeat\n",
    "from hmmlearn.hmm import GaussianHMM\n",
    "from IPython.display import Image, display\n",
    "from scipy import stats\n",
    "from scipy.stats import kurtosis as calc_kurtosis\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "\n",
    "from data import load_etfs\n",
    "from utils.paths import get_chapter_dir, get_output_dir\n",
    "from utils.reproducibility import set_global_seeds\n",
    "from utils.style import COLORS, plot_fidelity_comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a1b47af",
   "metadata": {
    "papermill": {
     "duration": 0.003313,
     "end_time": "2026-06-13T02:38:38.339836+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.336523+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## Diffusion Process Overview\n",
    "\n",
    "Diffusion models work by gradually adding noise (forward process) and then\n",
    "learning to reverse it (denoising). Diffusion-TS adds interpretable\n",
    "trend+seasonal decomposition to the denoiser."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "32457965",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.347247Z",
     "iopub.status.busy": "2026-06-13T02:38:38.347155Z",
     "iopub.status.idle": "2026-06-13T02:38:38.349129Z",
     "shell.execute_reply": "2026-06-13T02:38:38.348681Z"
    },
    "papermill": {
     "duration": 0.006609,
     "end_time": "2026-06-13T02:38:38.349700+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.343091+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "ASSETS_DIR = get_chapter_dir(5) / \"assets\"\n",
    "if (ASSETS_DIR / \"diffusion_forward_reverse.jpeg\").exists():\n",
    "    display(Image(ASSETS_DIR / \"diffusion_forward_reverse.jpeg\", width=800))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03b1294a",
   "metadata": {
    "papermill": {
     "duration": 0.003485,
     "end_time": "2026-06-13T02:38:38.358858+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.355373+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## Configuration\n",
    "\n",
    "The original Diffusion-TS paper uses seq_length=24, 6 stocks, and 10,000 epochs.\n",
    "We adapt for financial applications with the following considerations:\n",
    "\n",
    "| Parameter | Paper | Ours | Rationale |\n",
    "|-----------|-------|------|-----------|\n",
    "| seq_length | 24 | 60 | ~3 months of trading context |\n",
    "| feature_size | 6 | 20 | Balance diversity vs complexity |\n",
    "| epochs | 10,000 | 10,000 | Match paper for convergence |\n",
    "| lr | 1e-5→8e-4 | 1e-5→8e-4 | Warmup schedule from paper |\n",
    "\n",
    "With 20 assets × 60 timesteps = 1,200 values per sample (vs paper's 144),\n",
    "we train longer to capture the richer cross-asset structure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ba6f64f1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.366610Z",
     "iopub.status.busy": "2026-06-13T02:38:38.366531Z",
     "iopub.status.idle": "2026-06-13T02:38:38.368550Z",
     "shell.execute_reply": "2026-06-13T02:38:38.368179Z"
    },
    "papermill": {
     "duration": 0.0066,
     "end_time": "2026-06-13T02:38:38.368804+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.362204+00:00",
     "status": "completed"
    },
    "tags": [
     "parameters"
    ]
   },
   "outputs": [],
   "source": [
    "# Production defaults (Yuan & Qiao, ICLR 2024)\n",
    "SEQ_LENGTH = 60  # ~3 months of trading context (paper uses 24)\n",
    "FEATURE_SIZE = 20  # Number of ETFs (paper uses 6)\n",
    "N_LAYER_DEC = 4  # Decoder layers (paper uses 2; we use 4 for more capacity)\n",
    "TIMESTEPS = 500  # Forward diffusion steps\n",
    "SAMPLING_TIMESTEPS = 50  # DDIM fast sampling steps\n",
    "EPOCHS = 10000  # Training epochs (paper uses 10000)\n",
    "BATCH_SIZE = 64  # Training batch size\n",
    "WARMUP_STEPS = 500  # LR warmup steps\n",
    "GRADIENT_ACCUMULATE_EVERY = 2  # Gradient accumulation steps\n",
    "CLASSIFIER_EPOCHS = 1500  # Regime classifier training epochs\n",
    "SCHEDULER_PATIENCE = 500  # ReduceLROnPlateau patience\n",
    "N_SYNTHETIC = 500  # Number of unconditional synthetic sequences\n",
    "N_COND = 100  # Number of conditional samples per regime\n",
    "SEED = 42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "81b95f7c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.376225Z",
     "iopub.status.busy": "2026-06-13T02:38:38.376155Z",
     "iopub.status.idle": "2026-06-13T02:38:38.415694Z",
     "shell.execute_reply": "2026-06-13T02:38:38.415298Z"
    },
    "papermill": {
     "duration": 0.044098,
     "end_time": "2026-06-13T02:38:38.416275+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.372177+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "set_global_seeds(SEED)\n",
    "\n",
    "# Configuration\n",
    "RETRAIN = False  # Set True to retrain even if checkpoint exists\n",
    "\n",
    "CONFIG = {\n",
    "    # Sequence/architecture - scaled from paper (24×6) to financial use case\n",
    "    \"seq_length\": SEQ_LENGTH,\n",
    "    \"feature_size\": FEATURE_SIZE,\n",
    "    \"d_model\": 64,\n",
    "    \"n_heads\": 4,\n",
    "    \"n_layer_enc\": 2,\n",
    "    \"n_layer_dec\": N_LAYER_DEC,\n",
    "    # Diffusion process\n",
    "    \"timesteps\": TIMESTEPS,\n",
    "    \"sampling_timesteps\": SAMPLING_TIMESTEPS,\n",
    "    \"eta\": 0.0,  # Deterministic DDIM (eta=1 made variance worse, not better)\n",
    "    \"beta_schedule\": \"cosine\",\n",
    "    \"loss_type\": \"l1\",\n",
    "    # Training - match paper's schedule\n",
    "    \"epochs\": EPOCHS,\n",
    "    \"batch_size\": BATCH_SIZE,\n",
    "    \"lr\": 1e-5,  # Paper's base_lr (not warmup target)\n",
    "    \"warmup_lr\": 8e-4,  # Paper's warmup target\n",
    "    \"warmup_steps\": WARMUP_STEPS,\n",
    "    \"ema_decay\": 0.995,\n",
    "    \"gradient_accumulate_every\": GRADIENT_ACCUMULATE_EVERY,\n",
    "    # Data split\n",
    "    \"start_date\": \"2005-01-01\",\n",
    "    \"holdout_start\": \"2024-01-01\",\n",
    "    # Classifier guidance for regime-conditional generation\n",
    "    \"classifier_epochs\": CLASSIFIER_EPOCHS,\n",
    "    \"classifier_lr\": 5e-4,\n",
    "    # Regime-specific guidance: minority class (high-vol) needs gentler steering\n",
    "    # to prevent mode collapse (all samples pushed to extreme volatility).\n",
    "    # Low-Vol (majority): standard guidance, temperature=1, eta=0.5\n",
    "    # High-Vol (minority): lower scale, higher temperature for diversity, eta=0.7\n",
    "    \"guidance_settings\": {\n",
    "        0: {\"scale\": 0.75, \"temperature\": 1.0, \"eta\": 0.5},  # Low-Vol\n",
    "        1: {\"scale\": 0.3, \"temperature\": 2.0, \"eta\": 0.7},  # High-Vol: gentler\n",
    "    },\n",
    "    \"n_regimes\": 2,  # Low-Vol vs High-Vol (imbalanced data precludes 3-way)\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a7ed483a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.424309Z",
     "iopub.status.busy": "2026-06-13T02:38:38.424231Z",
     "iopub.status.idle": "2026-06-13T02:38:38.426433Z",
     "shell.execute_reply": "2026-06-13T02:38:38.426154Z"
    },
    "papermill": {
     "duration": 0.007099,
     "end_time": "2026-06-13T02:38:38.427002+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.419903+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n"
     ]
    }
   ],
   "source": [
    "# Output paths and reproducibility\n",
    "OUTPUT_DIR = get_output_dir(5, \"diffusion_ts\")\n",
    "CHECKPOINT_DIR = OUTPUT_DIR / \"checkpoints\"\n",
    "CHECKPOINT_DIR.mkdir(parents=True, exist_ok=True)\n",
    "CHECKPOINT_PATH = CHECKPOINT_DIR / \"diffusion_ts_model.pt\"\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f\"Using device: {device}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "946816df",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003361,
     "end_time": "2026-06-13T02:38:38.433815+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.430454+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 1. Data Loading and Preparation\n",
    "\n",
    "We load daily ETF returns and create overlapping sequences. The temporal\n",
    "train/holdout split at 2024-01-01 ensures unbiased TSTR evaluation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "23f8f466",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.441248Z",
     "iopub.status.busy": "2026-06-13T02:38:38.441177Z",
     "iopub.status.idle": "2026-06-13T02:38:38.443692Z",
     "shell.execute_reply": "2026-06-13T02:38:38.443377Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.006842,
     "end_time": "2026-06-13T02:38:38.443995+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.437153+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "def load_returns_data(start_date: str, n_assets: int) -> tuple[np.ndarray, np.ndarray, list[str]]:\n",
    "    \"\"\"Load daily returns for ETF assets with temporal split.\"\"\"\n",
    "    df = load_etfs()\n",
    "    start_dt = pl.lit(start_date).str.to_date()\n",
    "\n",
    "    returns_df = (\n",
    "        df.filter(pl.col(\"timestamp\") >= start_dt)\n",
    "        .sort([\"symbol\", \"timestamp\"])\n",
    "        .with_columns(pl.col(\"close\").pct_change().over(\"symbol\").alias(\"return\"))\n",
    "        .pivot(on=\"symbol\", index=\"timestamp\", values=\"return\")\n",
    "        .sort(\"timestamp\")\n",
    "        .drop_nulls()\n",
    "    )\n",
    "\n",
    "    data_cols = [c for c in returns_df.columns if c != \"timestamp\"][:n_assets]\n",
    "    timestamps = returns_df.select(\"timestamp\").to_numpy().flatten()\n",
    "    returns = returns_df.select(data_cols).to_numpy().astype(np.float32)\n",
    "\n",
    "    print(f\"Loaded {len(returns)} days of returns for {len(data_cols)} assets\")\n",
    "    print(f\"Date range: {timestamps[0]} to {timestamps[-1]}\")\n",
    "    return returns, timestamps, data_cols"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03295520",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003393,
     "end_time": "2026-06-13T02:38:38.450885+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.447492+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Create Overlapping Sequences\n",
    "\n",
    "Sliding windows maximize sample count from limited financial data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2cb88685",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.458331Z",
     "iopub.status.busy": "2026-06-13T02:38:38.458261Z",
     "iopub.status.idle": "2026-06-13T02:38:38.460099Z",
     "shell.execute_reply": "2026-06-13T02:38:38.459785Z"
    },
    "papermill": {
     "duration": 0.006223,
     "end_time": "2026-06-13T02:38:38.460455+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.454232+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "def create_sequences(data: np.ndarray, seq_length: int) -> np.ndarray:\n",
    "    \"\"\"Create overlapping sequences from time series data.\"\"\"\n",
    "    n_seq = len(data) - seq_length + 1\n",
    "    seqs = np.zeros((n_seq, seq_length, data.shape[1]), dtype=np.float32)\n",
    "    for i in range(n_seq):\n",
    "        seqs[i] = data[i : i + seq_length]\n",
    "    return seqs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0af12a0c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.469019Z",
     "iopub.status.busy": "2026-06-13T02:38:38.468943Z",
     "iopub.status.idle": "2026-06-13T02:38:38.535521Z",
     "shell.execute_reply": "2026-06-13T02:38:38.535094Z"
    },
    "papermill": {
     "duration": 0.072037,
     "end_time": "2026-06-13T02:38:38.535933+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.463896+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 1894 days of returns for 20 assets\n",
      "Date range: 2018-06-20 to 2025-12-31\n",
      "\n",
      "Train: 1,392 days | Holdout: 502 days\n",
      "Train sequences: (1333, 60, 20) | Holdout: (443, 60, 20)\n"
     ]
    }
   ],
   "source": [
    "all_returns, all_timestamps, asset_names = load_returns_data(\n",
    "    CONFIG[\"start_date\"], CONFIG[\"feature_size\"]\n",
    ")\n",
    "n_assets = all_returns.shape[1]\n",
    "\n",
    "# Temporal split\n",
    "holdout_dt = np.datetime64(CONFIG[\"holdout_start\"])\n",
    "train_mask = all_timestamps < holdout_dt\n",
    "\n",
    "returns = all_returns[train_mask]\n",
    "holdout_returns = all_returns[~train_mask]\n",
    "\n",
    "print(f\"\\nTrain: {len(returns):,} days | Holdout: {len(holdout_returns):,} days\")\n",
    "\n",
    "sequences = create_sequences(returns, CONFIG[\"seq_length\"])\n",
    "holdout_sequences = create_sequences(holdout_returns, CONFIG[\"seq_length\"])\n",
    "print(f\"Train sequences: {sequences.shape} | Holdout: {holdout_sequences.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48bb6e37",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.009807,
     "end_time": "2026-06-13T02:38:38.552770+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.542963+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 2. Utility Modules\n",
    "\n",
    "These building blocks support the Transformer architecture: sinusoidal\n",
    "timestep embeddings, learnable positional encoding, and adaptive layer\n",
    "normalization that conditions on the diffusion timestep."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ac17bf73",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.569292Z",
     "iopub.status.busy": "2026-06-13T02:38:38.569089Z",
     "iopub.status.idle": "2026-06-13T02:38:38.572296Z",
     "shell.execute_reply": "2026-06-13T02:38:38.571600Z"
    },
    "papermill": {
     "duration": 0.011849,
     "end_time": "2026-06-13T02:38:38.572734+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.560885+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Inline helpers (from Diffusion-TS model_utils)\n",
    "def extract(a, t, x_shape):\n",
    "    \"\"\"Gather values from `a` at indices `t`, reshape for broadcasting.\"\"\"\n",
    "    b, *_ = t.shape\n",
    "    out = a.gather(-1, t)\n",
    "    return out.reshape(b, *((1,) * (len(x_shape) - 1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0d950fa1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.587723Z",
     "iopub.status.busy": "2026-06-13T02:38:38.587532Z",
     "iopub.status.idle": "2026-06-13T02:38:38.591326Z",
     "shell.execute_reply": "2026-06-13T02:38:38.590776Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.012525,
     "end_time": "2026-06-13T02:38:38.592107+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.579582+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class SinusoidalPosEmb(nn.Module):\n",
    "    \"\"\"Sinusoidal positional embedding for diffusion timestep.\"\"\"\n",
    "\n",
    "    def __init__(self, dim):\n",
    "        super().__init__()\n",
    "        self.dim = dim\n",
    "\n",
    "    def forward(self, x):\n",
    "        half_dim = self.dim // 2\n",
    "        emb = math.log(10000) / (half_dim - 1)\n",
    "        emb = torch.exp(torch.arange(half_dim, device=x.device) * -emb)\n",
    "        emb = x[:, None] * emb[None, :]\n",
    "        return torch.cat((emb.sin(), emb.cos()), dim=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "542c006d",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.005477,
     "end_time": "2026-06-13T02:38:38.604714+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.599237+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Adaptive Layer Normalization\n",
    "\n",
    "AdaLayerNorm modulates the normalized activations by a scale and shift\n",
    "derived from the diffusion timestep embedding. This gives each Transformer\n",
    "layer information about the current noise level."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ca8646f9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.632651Z",
     "iopub.status.busy": "2026-06-13T02:38:38.631972Z",
     "iopub.status.idle": "2026-06-13T02:38:38.642532Z",
     "shell.execute_reply": "2026-06-13T02:38:38.641054Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.027748,
     "end_time": "2026-06-13T02:38:38.643583+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.615835+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class AdaLayerNorm(nn.Module):\n",
    "    \"\"\"Layer norm conditioned on diffusion timestep via scale+shift.\"\"\"\n",
    "\n",
    "    def __init__(self, n_embd):\n",
    "        super().__init__()\n",
    "        self.emb = SinusoidalPosEmb(n_embd)\n",
    "        self.silu = nn.SiLU()\n",
    "        self.linear = nn.Linear(n_embd, n_embd * 2)\n",
    "        self.layernorm = nn.LayerNorm(n_embd, elementwise_affine=False)\n",
    "\n",
    "    def forward(self, x, timestep, label_emb=None):\n",
    "        emb = self.emb(timestep)\n",
    "        if label_emb is not None:\n",
    "            emb = emb + label_emb\n",
    "        emb = self.linear(self.silu(emb)).unsqueeze(1)\n",
    "        scale, shift = torch.chunk(emb, 2, dim=2)\n",
    "        return self.layernorm(x) * (1 + scale) + shift"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fac88f8d",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.01404,
     "end_time": "2026-06-13T02:38:38.668667+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.654627+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Learnable Positional Encoding and Conv Embedding\n",
    "\n",
    "LearnablePositionalEncoding adds a learned position vector to each timestep.\n",
    "Conv_MLP projects the input feature dimension to the model dimension using\n",
    "a 1D convolution, which captures local patterns in the feature axis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b99e1dcc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.688526Z",
     "iopub.status.busy": "2026-06-13T02:38:38.688414Z",
     "iopub.status.idle": "2026-06-13T02:38:38.690749Z",
     "shell.execute_reply": "2026-06-13T02:38:38.690419Z"
    },
    "papermill": {
     "duration": 0.008537,
     "end_time": "2026-06-13T02:38:38.691088+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.682551+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class LearnablePositionalEncoding(nn.Module):\n",
    "    \"\"\"Learned positional encoding added to sequence embeddings.\"\"\"\n",
    "\n",
    "    def __init__(self, d_model, dropout=0.1, max_len=1024):\n",
    "        super().__init__()\n",
    "        self.dropout = nn.Dropout(p=dropout)\n",
    "        self.pe = nn.Parameter(torch.empty(1, max_len, d_model))\n",
    "        nn.init.uniform_(self.pe, -0.02, 0.02)\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.dropout(x + self.pe[:, : x.size(1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "face08a6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.698568Z",
     "iopub.status.busy": "2026-06-13T02:38:38.698483Z",
     "iopub.status.idle": "2026-06-13T02:38:38.700430Z",
     "shell.execute_reply": "2026-06-13T02:38:38.700129Z"
    },
    "papermill": {
     "duration": 0.006156,
     "end_time": "2026-06-13T02:38:38.700758+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.694602+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class Transpose(nn.Module):\n",
    "    \"\"\"Transpose wrapper for use in nn.Sequential.\"\"\"\n",
    "\n",
    "    def __init__(self, shape: tuple):\n",
    "        super().__init__()\n",
    "        self.shape = shape\n",
    "\n",
    "    def forward(self, x):\n",
    "        return x.transpose(*self.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1be19421",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.708241Z",
     "iopub.status.busy": "2026-06-13T02:38:38.708168Z",
     "iopub.status.idle": "2026-06-13T02:38:38.710164Z",
     "shell.execute_reply": "2026-06-13T02:38:38.709930Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.006186,
     "end_time": "2026-06-13T02:38:38.710472+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.704286+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class Conv_MLP(nn.Module):\n",
    "    \"\"\"1D convolution embedding: features → model dimension.\"\"\"\n",
    "\n",
    "    def __init__(self, in_dim, out_dim, resid_pdrop=0.0):\n",
    "        super().__init__()\n",
    "        self.sequential = nn.Sequential(\n",
    "            Transpose(shape=(1, 2)),\n",
    "            nn.Conv1d(in_dim, out_dim, 3, stride=1, padding=1),\n",
    "            nn.Dropout(p=resid_pdrop),\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.sequential(x).transpose(1, 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14b4b3b1",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003383,
     "end_time": "2026-06-13T02:38:38.718463+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.715080+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 3. Interpretable Decomposition\n",
    "\n",
    "The key innovation of Diffusion-TS: each decoder layer extracts **trend**\n",
    "and **seasonal** components from its intermediate representation.\n",
    "\n",
    "- **TrendBlock**: Learns a polynomial basis (degree 3) via 1D convolutions,\n",
    "  then multiplies by a polynomial space $[t, t^2, t^3]$ to produce a smooth trend.\n",
    "- **FourierLayer**: Computes the DFT of the latent, selects top-$k$ frequencies\n",
    "  by magnitude, then reconstructs via inverse DFT. This extracts dominant\n",
    "  periodic patterns without assuming a fixed period.\n",
    "\n",
    "Across decoder layers, trend and seasonal residuals accumulate, building up\n",
    "the full decomposition progressively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "89bbf7f5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.725717Z",
     "iopub.status.busy": "2026-06-13T02:38:38.725649Z",
     "iopub.status.idle": "2026-06-13T02:38:38.728259Z",
     "shell.execute_reply": "2026-06-13T02:38:38.727745Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.006692,
     "end_time": "2026-06-13T02:38:38.728525+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.721833+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class TrendBlock(nn.Module):\n",
    "    \"\"\"Polynomial regression on latent representation → smooth trend.\"\"\"\n",
    "\n",
    "    def __init__(self, in_dim, out_dim, in_feat, out_feat, act):\n",
    "        super().__init__()\n",
    "        trend_poly = 3\n",
    "        self.trend = nn.Sequential(\n",
    "            nn.Conv1d(in_channels=in_dim, out_channels=trend_poly, kernel_size=3, padding=1),\n",
    "            act,\n",
    "            Transpose(shape=(1, 2)),\n",
    "            nn.Conv1d(in_feat, out_feat, 3, stride=1, padding=1),\n",
    "        )\n",
    "        lin_space = torch.arange(1, out_dim + 1, 1) / (out_dim + 1)\n",
    "        self.poly_space = torch.stack([lin_space ** float(p + 1) for p in range(trend_poly)], dim=0)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.trend(x).transpose(1, 2)\n",
    "        trend_vals = torch.matmul(x.transpose(1, 2), self.poly_space.to(x.device))\n",
    "        return trend_vals.transpose(1, 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eca6bdcc",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003365,
     "end_time": "2026-06-13T02:38:38.735323+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.731958+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### FourierLayer: Top-k Frequency Selection\n",
    "\n",
    "Rather than using all Fourier coefficients, the layer selects the top-$k$\n",
    "frequencies (by magnitude) and reconstructs only those. This acts as a\n",
    "learned bandpass filter that adapts to the data's spectral content."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7563927d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.743106Z",
     "iopub.status.busy": "2026-06-13T02:38:38.743001Z",
     "iopub.status.idle": "2026-06-13T02:38:38.746769Z",
     "shell.execute_reply": "2026-06-13T02:38:38.746414Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.008224,
     "end_time": "2026-06-13T02:38:38.747058+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.738834+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class FourierLayer(nn.Module):\n",
    "    \"\"\"Extract seasonal component via top-k inverse DFT.\"\"\"\n",
    "\n",
    "    def __init__(self, d_model, low_freq=1, factor=1):\n",
    "        super().__init__()\n",
    "        self.d_model = d_model\n",
    "        self.factor = factor\n",
    "        self.low_freq = low_freq\n",
    "\n",
    "    def forward(self, x):\n",
    "        \"\"\"x: (b, t, d)\"\"\"\n",
    "        b, t, d = x.shape\n",
    "        x_freq = torch.fft.rfft(x, dim=1)\n",
    "\n",
    "        if t % 2 == 0:\n",
    "            x_freq = x_freq[:, self.low_freq : -1]\n",
    "            f = torch.fft.rfftfreq(t)[self.low_freq : -1]\n",
    "        else:\n",
    "            x_freq = x_freq[:, self.low_freq :]\n",
    "            f = torch.fft.rfftfreq(t)[self.low_freq :]\n",
    "\n",
    "        x_freq, index_tuple = self.topk_freq(x_freq)\n",
    "        f = repeat(f, \"f -> b f d\", b=x_freq.size(0), d=x_freq.size(2)).to(x_freq.device)\n",
    "        f = rearrange(f[index_tuple], \"b f d -> b f () d\").to(x_freq.device)\n",
    "        return self.extrapolate(x_freq, f, t)\n",
    "\n",
    "    def extrapolate(self, x_freq, f, t):\n",
    "        x_freq = torch.cat([x_freq, x_freq.conj()], dim=1)\n",
    "        f = torch.cat([f, -f], dim=1)\n",
    "        t_range = rearrange(torch.arange(t, dtype=torch.float), \"t -> () () t ()\").to(x_freq.device)\n",
    "        amp = rearrange(x_freq.abs(), \"b f d -> b f () d\")\n",
    "        phase = rearrange(x_freq.angle(), \"b f d -> b f () d\")\n",
    "        x_time = amp * torch.cos(2 * math.pi * f * t_range + phase)\n",
    "        return reduce(x_time, \"b f t d -> b t d\", \"sum\")\n",
    "\n",
    "    def topk_freq(self, x_freq):\n",
    "        length = x_freq.shape[1]\n",
    "        top_k = int(self.factor * math.log(max(length, 2)))\n",
    "        top_k = max(top_k, 1)\n",
    "        values, indices = torch.topk(x_freq.abs(), top_k, dim=1, largest=True, sorted=True)\n",
    "        mesh_a, mesh_b = torch.meshgrid(\n",
    "            torch.arange(x_freq.size(0)), torch.arange(x_freq.size(2)), indexing=\"ij\"\n",
    "        )\n",
    "        index_tuple = (mesh_a.unsqueeze(1), indices, mesh_b.unsqueeze(1))\n",
    "        x_freq = x_freq[index_tuple]\n",
    "        return x_freq, index_tuple"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5a6d35f",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003594,
     "end_time": "2026-06-13T02:38:38.754228+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.750634+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 4. Transformer Architecture\n",
    "\n",
    "The encoder-decoder Transformer processes noised sequences. The **encoder**\n",
    "creates a contextual representation conditioned on the diffusion timestep.\n",
    "The **decoder** cross-attends to the encoder output and, at each layer,\n",
    "extracts trend and seasonal residuals that accumulate across layers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d328d5d6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.761832Z",
     "iopub.status.busy": "2026-06-13T02:38:38.761727Z",
     "iopub.status.idle": "2026-06-13T02:38:38.764991Z",
     "shell.execute_reply": "2026-06-13T02:38:38.764371Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.007672,
     "end_time": "2026-06-13T02:38:38.765361+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.757689+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class FullAttention(nn.Module):\n",
    "    \"\"\"Multi-head self-attention.\"\"\"\n",
    "\n",
    "    def __init__(self, n_embd, n_head, attn_pdrop=0.0, resid_pdrop=0.0):\n",
    "        super().__init__()\n",
    "        assert n_embd % n_head == 0\n",
    "        self.key = nn.Linear(n_embd, n_embd)\n",
    "        self.query = nn.Linear(n_embd, n_embd)\n",
    "        self.value = nn.Linear(n_embd, n_embd)\n",
    "        self.attn_drop = nn.Dropout(attn_pdrop)\n",
    "        self.resid_drop = nn.Dropout(resid_pdrop)\n",
    "        self.proj = nn.Linear(n_embd, n_embd)\n",
    "        self.n_head = n_head\n",
    "\n",
    "    def forward(self, x, mask=None):\n",
    "        B, T, C = x.size()\n",
    "        k = self.key(x).view(B, T, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        q = self.query(x).view(B, T, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        v = self.value(x).view(B, T, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))\n",
    "        att = F.softmax(att, dim=-1)\n",
    "        att = self.attn_drop(att)\n",
    "        y = att @ v\n",
    "        y = y.transpose(1, 2).contiguous().view(B, T, C)\n",
    "        return self.resid_drop(self.proj(y))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc29138e",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003546,
     "end_time": "2026-06-13T02:38:38.772720+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.769174+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Cross-Attention\n",
    "\n",
    "The decoder cross-attends to the encoder's output. Queries come from the\n",
    "decoder, while keys and values come from the encoder -- this lets each\n",
    "decoder position gather relevant context from the full encoded sequence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8e32e57b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.780500Z",
     "iopub.status.busy": "2026-06-13T02:38:38.780385Z",
     "iopub.status.idle": "2026-06-13T02:38:38.783519Z",
     "shell.execute_reply": "2026-06-13T02:38:38.783169Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.007572,
     "end_time": "2026-06-13T02:38:38.783800+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.776228+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class CrossAttention(nn.Module):\n",
    "    \"\"\"Multi-head cross-attention (decoder queries, encoder keys/values).\"\"\"\n",
    "\n",
    "    def __init__(self, n_embd, condition_embd, n_head, attn_pdrop=0.0, resid_pdrop=0.0):\n",
    "        super().__init__()\n",
    "        assert n_embd % n_head == 0\n",
    "        self.key = nn.Linear(condition_embd, n_embd)\n",
    "        self.query = nn.Linear(n_embd, n_embd)\n",
    "        self.value = nn.Linear(condition_embd, n_embd)\n",
    "        self.attn_drop = nn.Dropout(attn_pdrop)\n",
    "        self.resid_drop = nn.Dropout(resid_pdrop)\n",
    "        self.proj = nn.Linear(n_embd, n_embd)\n",
    "        self.n_head = n_head\n",
    "\n",
    "    def forward(self, x, encoder_output, mask=None):\n",
    "        B, T, C = x.size()\n",
    "        B, T_E, _ = encoder_output.size()\n",
    "        k = self.key(encoder_output).view(B, T_E, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        q = self.query(x).view(B, T, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        v = self.value(encoder_output).view(B, T_E, self.n_head, C // self.n_head).transpose(1, 2)\n",
    "        att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))\n",
    "        att = F.softmax(att, dim=-1)\n",
    "        att = self.attn_drop(att)\n",
    "        y = att @ v\n",
    "        y = y.transpose(1, 2).contiguous().view(B, T, C)\n",
    "        return self.resid_drop(self.proj(y))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c2a44cd",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.00349,
     "end_time": "2026-06-13T02:38:38.790973+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.787483+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Encoder and Decoder Blocks\n",
    "\n",
    "Each encoder block applies adaptive layer norm → self-attention → FFN.\n",
    "Each decoder block adds cross-attention to the encoder output, then\n",
    "splits the representation into two branches: one for trend extraction\n",
    "(polynomial regression) and one for seasonal extraction (Fourier layer)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ec57d54f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.798802Z",
     "iopub.status.busy": "2026-06-13T02:38:38.798707Z",
     "iopub.status.idle": "2026-06-13T02:38:38.801305Z",
     "shell.execute_reply": "2026-06-13T02:38:38.800897Z"
    },
    "papermill": {
     "duration": 0.007088,
     "end_time": "2026-06-13T02:38:38.801702+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.794614+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class EncoderBlock(nn.Module):\n",
    "    \"\"\"Transformer encoder block with timestep-conditioned AdaLayerNorm.\"\"\"\n",
    "\n",
    "    def __init__(self, n_embd=64, n_head=4, attn_pdrop=0.0, resid_pdrop=0.0, mlp_hidden_times=4):\n",
    "        super().__init__()\n",
    "        self.ln1 = AdaLayerNorm(n_embd)\n",
    "        self.ln2 = nn.LayerNorm(n_embd)\n",
    "        self.attn = FullAttention(n_embd, n_head, attn_pdrop, resid_pdrop)\n",
    "        self.mlp = nn.Sequential(\n",
    "            nn.Linear(n_embd, mlp_hidden_times * n_embd),\n",
    "            nn.GELU(),\n",
    "            nn.Linear(mlp_hidden_times * n_embd, n_embd),\n",
    "            nn.Dropout(resid_pdrop),\n",
    "        )\n",
    "\n",
    "    def forward(self, x, timestep, mask=None):\n",
    "        x = x + self.attn(self.ln1(x, timestep), mask=mask)\n",
    "        x = x + self.mlp(self.ln2(x))\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "702f63ae",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.809677Z",
     "iopub.status.busy": "2026-06-13T02:38:38.809518Z",
     "iopub.status.idle": "2026-06-13T02:38:38.812035Z",
     "shell.execute_reply": "2026-06-13T02:38:38.811671Z"
    },
    "papermill": {
     "duration": 0.007008,
     "end_time": "2026-06-13T02:38:38.812335+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.805327+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class Encoder(nn.Module):\n",
    "    \"\"\"Stack of encoder blocks.\"\"\"\n",
    "\n",
    "    def __init__(self, n_layer=2, n_embd=64, n_head=4, attn_pdrop=0.0, resid_pdrop=0.0):\n",
    "        super().__init__()\n",
    "        self.blocks = nn.ModuleList(\n",
    "            [EncoderBlock(n_embd, n_head, attn_pdrop, resid_pdrop) for _ in range(n_layer)]\n",
    "        )\n",
    "\n",
    "    def forward(self, x, t):\n",
    "        for block in self.blocks:\n",
    "            x = block(x, t)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ad25d9fe",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.820264Z",
     "iopub.status.busy": "2026-06-13T02:38:38.820179Z",
     "iopub.status.idle": "2026-06-13T02:38:38.823391Z",
     "shell.execute_reply": "2026-06-13T02:38:38.823048Z"
    },
    "papermill": {
     "duration": 0.007878,
     "end_time": "2026-06-13T02:38:38.823783+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.815905+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class DecoderBlock(nn.Module):\n",
    "    \"\"\"Decoder block: self-attn + cross-attn + trend/season extraction.\"\"\"\n",
    "\n",
    "    def __init__(\n",
    "        self,\n",
    "        n_channel,\n",
    "        n_feat,\n",
    "        n_embd=64,\n",
    "        n_head=4,\n",
    "        attn_pdrop=0.0,\n",
    "        resid_pdrop=0.0,\n",
    "        mlp_hidden_times=4,\n",
    "        condition_dim=64,\n",
    "    ):\n",
    "        super().__init__()\n",
    "        self.ln1 = AdaLayerNorm(n_embd)\n",
    "        self.ln2 = nn.LayerNorm(n_embd)\n",
    "        self.ln1_1 = AdaLayerNorm(n_embd)\n",
    "\n",
    "        self.attn1 = FullAttention(n_embd, n_head, attn_pdrop, resid_pdrop)\n",
    "        self.attn2 = CrossAttention(n_embd, condition_dim, n_head, attn_pdrop, resid_pdrop)\n",
    "\n",
    "        act = nn.GELU()\n",
    "        self.trend = TrendBlock(n_channel, n_channel, n_embd, n_feat, act=act)\n",
    "        self.seasonal = FourierLayer(d_model=n_embd)\n",
    "\n",
    "        self.mlp = nn.Sequential(\n",
    "            nn.Linear(n_embd, mlp_hidden_times * n_embd),\n",
    "            nn.GELU(),\n",
    "            nn.Linear(mlp_hidden_times * n_embd, n_embd),\n",
    "            nn.Dropout(resid_pdrop),\n",
    "        )\n",
    "        self.proj = nn.Conv1d(n_channel, n_channel * 2, 1)\n",
    "        self.linear = nn.Linear(n_embd, n_feat)\n",
    "\n",
    "    def forward(self, x, encoder_output, timestep, mask=None):\n",
    "        x = x + self.attn1(self.ln1(x, timestep), mask=mask)\n",
    "        x = x + self.attn2(self.ln1_1(x, timestep), encoder_output, mask=mask)\n",
    "        x1, x2 = self.proj(x).chunk(2, dim=1)\n",
    "        trend, season = self.trend(x1), self.seasonal(x2)\n",
    "        x = x + self.mlp(self.ln2(x))\n",
    "        m = torch.mean(x, dim=1, keepdim=True)\n",
    "        return x - m, self.linear(m), trend, season"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3fce7e10",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.831544Z",
     "iopub.status.busy": "2026-06-13T02:38:38.831475Z",
     "iopub.status.idle": "2026-06-13T02:38:38.834181Z",
     "shell.execute_reply": "2026-06-13T02:38:38.833895Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.007268,
     "end_time": "2026-06-13T02:38:38.834708+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.827440+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class Decoder(nn.Module):\n",
    "    \"\"\"Stack of decoder blocks, accumulating trend and seasonal components.\"\"\"\n",
    "\n",
    "    def __init__(\n",
    "        self,\n",
    "        n_channel,\n",
    "        n_feat,\n",
    "        n_embd=64,\n",
    "        n_head=4,\n",
    "        n_layer=4,\n",
    "        attn_pdrop=0.0,\n",
    "        resid_pdrop=0.0,\n",
    "        condition_dim=64,\n",
    "    ):\n",
    "        super().__init__()\n",
    "        self.d_model = n_embd\n",
    "        self.n_feat = n_feat\n",
    "        self.blocks = nn.ModuleList(\n",
    "            [\n",
    "                DecoderBlock(\n",
    "                    n_feat=n_feat,\n",
    "                    n_channel=n_channel,\n",
    "                    n_embd=n_embd,\n",
    "                    n_head=n_head,\n",
    "                    attn_pdrop=attn_pdrop,\n",
    "                    resid_pdrop=resid_pdrop,\n",
    "                    condition_dim=condition_dim,\n",
    "                )\n",
    "                for _ in range(n_layer)\n",
    "            ]\n",
    "        )\n",
    "\n",
    "    def forward(self, x, t, enc):\n",
    "        b, c, _ = x.shape\n",
    "        mean = []\n",
    "        season = torch.zeros((b, c, self.d_model), device=x.device)\n",
    "        trend = torch.zeros((b, c, self.n_feat), device=x.device)\n",
    "        for block in self.blocks:\n",
    "            x, residual_mean, residual_trend, residual_season = block(x, enc, t)\n",
    "            season += residual_season\n",
    "            trend += residual_trend\n",
    "            mean.append(residual_mean)\n",
    "        mean = torch.cat(mean, dim=1)\n",
    "        return x, mean, trend, season"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6bcf27b",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003347,
     "end_time": "2026-06-13T02:38:38.841465+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.838118+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Full Transformer\n",
    "\n",
    "The Transformer wraps encoder and decoder with input/output projections.\n",
    "The final output combines: (1) accumulated trend across decoder layers,\n",
    "(2) seasonal component projected back to feature space, and (3) a residual\n",
    "from the decoder output. This produces $\\hat{x}_0 = \\text{trend} + \\text{season\\_error}$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "25eb4749",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.848776Z",
     "iopub.status.busy": "2026-06-13T02:38:38.848705Z",
     "iopub.status.idle": "2026-06-13T02:38:38.852042Z",
     "shell.execute_reply": "2026-06-13T02:38:38.851742Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.00756,
     "end_time": "2026-06-13T02:38:38.852387+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.844827+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class DiffusionTransformer(nn.Module):\n",
    "    \"\"\"Encoder-decoder Transformer with interpretable trend+seasonal decomposition.\"\"\"\n",
    "\n",
    "    def __init__(\n",
    "        self,\n",
    "        n_feat,\n",
    "        n_channel,\n",
    "        n_layer_enc=2,\n",
    "        n_layer_dec=4,\n",
    "        n_embd=64,\n",
    "        n_heads=4,\n",
    "        attn_pdrop=0.0,\n",
    "        resid_pdrop=0.0,\n",
    "        mlp_hidden_times=4,\n",
    "        max_len=2048,\n",
    "    ):\n",
    "        super().__init__()\n",
    "        self.emb = Conv_MLP(n_feat, n_embd, resid_pdrop=resid_pdrop)\n",
    "        self.inverse = Conv_MLP(n_embd, n_feat, resid_pdrop=resid_pdrop)\n",
    "\n",
    "        kernel_size, padding = (1, 0) if n_feat < 32 and n_channel < 64 else (5, 2)\n",
    "\n",
    "        self.combine_s = nn.Conv1d(\n",
    "            n_embd,\n",
    "            n_feat,\n",
    "            kernel_size=kernel_size,\n",
    "            stride=1,\n",
    "            padding=padding,\n",
    "            padding_mode=\"circular\",\n",
    "            bias=False,\n",
    "        )\n",
    "        self.combine_m = nn.Conv1d(\n",
    "            n_layer_dec,\n",
    "            1,\n",
    "            kernel_size=1,\n",
    "            stride=1,\n",
    "            padding=0,\n",
    "            padding_mode=\"circular\",\n",
    "            bias=False,\n",
    "        )\n",
    "\n",
    "        self.encoder = Encoder(n_layer_enc, n_embd, n_heads, attn_pdrop, resid_pdrop)\n",
    "        self.pos_enc = LearnablePositionalEncoding(n_embd, dropout=resid_pdrop, max_len=max_len)\n",
    "\n",
    "        self.decoder = Decoder(\n",
    "            n_channel,\n",
    "            n_feat,\n",
    "            n_embd,\n",
    "            n_heads,\n",
    "            n_layer_dec,\n",
    "            attn_pdrop,\n",
    "            resid_pdrop,\n",
    "            condition_dim=n_embd,\n",
    "        )\n",
    "        self.pos_dec = LearnablePositionalEncoding(n_embd, dropout=resid_pdrop, max_len=max_len)\n",
    "\n",
    "    def forward(self, x, t, return_res=False):\n",
    "        emb = self.emb(x)\n",
    "        inp_enc = self.pos_enc(emb)\n",
    "        enc_cond = self.encoder(inp_enc, t)\n",
    "\n",
    "        inp_dec = self.pos_dec(emb)\n",
    "        output, mean, trend, season = self.decoder(inp_dec, t, enc_cond)\n",
    "\n",
    "        res = self.inverse(output)\n",
    "        res_m = torch.mean(res, dim=1, keepdim=True)\n",
    "        season_error = self.combine_s(season.transpose(1, 2)).transpose(1, 2) + res - res_m\n",
    "        trend = self.combine_m(mean) + res_m + trend\n",
    "\n",
    "        if return_res:\n",
    "            return trend, self.combine_s(season.transpose(1, 2)).transpose(1, 2), res - res_m\n",
    "\n",
    "        return trend, season_error"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6218f1b2",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003374,
     "end_time": "2026-06-13T02:38:38.859321+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.855947+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 5. Diffusion Process\n",
    "\n",
    "The `DiffusionTS` class implements the full DDPM framework with $x_0$\n",
    "prediction. Key methods:\n",
    "- `q_sample`: Forward process -- add noise to clean data\n",
    "- `model_predictions`: Get model's $\\hat{x}_0$ and derived noise\n",
    "- `p_mean_variance`: Compute posterior $p(x_{t-1}|x_t)$ -- **no clamp**\n",
    "- `_train_loss`: L1 + Fourier loss (time + frequency domain)\n",
    "- `fast_sample`: DDIM-style accelerated sampling\n",
    "\n",
    "**Critical adaptation**: We remove `clamp(-1, 1)` from `p_mean_variance`\n",
    "and `model_predictions`. Financial returns are StandardScaler-normalized\n",
    "but unbounded -- clamping would truncate the distribution tails."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3f691d22",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.866808Z",
     "iopub.status.busy": "2026-06-13T02:38:38.866737Z",
     "iopub.status.idle": "2026-06-13T02:38:38.868697Z",
     "shell.execute_reply": "2026-06-13T02:38:38.868337Z"
    },
    "papermill": {
     "duration": 0.006203,
     "end_time": "2026-06-13T02:38:38.868914+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.862711+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "def cosine_beta_schedule(timesteps, s=0.008):\n",
    "    \"\"\"Cosine schedule (Nichol & Dhariwal 2021).\"\"\"\n",
    "    steps = timesteps + 1\n",
    "    x = torch.linspace(0, timesteps, steps, dtype=torch.float64)\n",
    "    alphas_cumprod = torch.cos(((x / timesteps) + s) / (1 + s) * math.pi * 0.5) ** 2\n",
    "    alphas_cumprod = alphas_cumprod / alphas_cumprod[0]\n",
    "    betas = 1 - (alphas_cumprod[1:] / alphas_cumprod[:-1])\n",
    "    return torch.clip(betas, 0, 0.999)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0ee58004",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.876619Z",
     "iopub.status.busy": "2026-06-13T02:38:38.876546Z",
     "iopub.status.idle": "2026-06-13T02:38:38.888294Z",
     "shell.execute_reply": "2026-06-13T02:38:38.887949Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.016296,
     "end_time": "2026-06-13T02:38:38.888678+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.872382+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class DiffusionTS(nn.Module):\n",
    "    \"\"\"Diffusion-TS: x_0-prediction diffusion with Fourier loss.\"\"\"\n",
    "\n",
    "    def __init__(\n",
    "        self,\n",
    "        seq_length,\n",
    "        feature_size,\n",
    "        n_layer_enc=2,\n",
    "        n_layer_dec=4,\n",
    "        d_model=64,\n",
    "        timesteps=500,\n",
    "        sampling_timesteps=None,\n",
    "        loss_type=\"l1\",\n",
    "        n_heads=4,\n",
    "        mlp_hidden_times=4,\n",
    "        eta=0.0,\n",
    "        reg_weight=None,\n",
    "    ):\n",
    "        super().__init__()\n",
    "        self.eta = eta\n",
    "        self.seq_length = seq_length\n",
    "        self.feature_size = feature_size\n",
    "        self.ff_weight = reg_weight if reg_weight is not None else math.sqrt(seq_length) / 5\n",
    "\n",
    "        self.model = DiffusionTransformer(\n",
    "            n_feat=feature_size,\n",
    "            n_channel=seq_length,\n",
    "            n_layer_enc=n_layer_enc,\n",
    "            n_layer_dec=n_layer_dec,\n",
    "            n_heads=n_heads,\n",
    "            mlp_hidden_times=mlp_hidden_times,\n",
    "            max_len=seq_length,\n",
    "            n_embd=d_model,\n",
    "        )\n",
    "\n",
    "        betas = cosine_beta_schedule(timesteps)\n",
    "        alphas = 1.0 - betas\n",
    "        alphas_cumprod = torch.cumprod(alphas, dim=0)\n",
    "        alphas_cumprod_prev = F.pad(alphas_cumprod[:-1], (1, 0), value=1.0)\n",
    "\n",
    "        self.num_timesteps = int(timesteps)\n",
    "        self.loss_type = loss_type\n",
    "        self.sampling_timesteps = (\n",
    "            sampling_timesteps if sampling_timesteps is not None else timesteps\n",
    "        )\n",
    "        self.fast_sampling = self.sampling_timesteps < timesteps\n",
    "\n",
    "        def register(name, val):\n",
    "            self.register_buffer(name, val.to(torch.float32))\n",
    "\n",
    "        register(\"betas\", betas)\n",
    "        register(\"alphas_cumprod\", alphas_cumprod)\n",
    "        register(\"alphas_cumprod_prev\", alphas_cumprod_prev)\n",
    "        register(\"sqrt_alphas_cumprod\", torch.sqrt(alphas_cumprod))\n",
    "        register(\"sqrt_one_minus_alphas_cumprod\", torch.sqrt(1.0 - alphas_cumprod))\n",
    "        register(\"log_one_minus_alphas_cumprod\", torch.log(1.0 - alphas_cumprod))\n",
    "        register(\"sqrt_recip_alphas_cumprod\", torch.sqrt(1.0 / alphas_cumprod))\n",
    "        register(\"sqrt_recipm1_alphas_cumprod\", torch.sqrt(1.0 / alphas_cumprod - 1))\n",
    "\n",
    "        posterior_variance = betas * (1.0 - alphas_cumprod_prev) / (1.0 - alphas_cumprod)\n",
    "        register(\"posterior_variance\", posterior_variance)\n",
    "        register(\"posterior_log_variance_clipped\", torch.log(posterior_variance.clamp(min=1e-20)))\n",
    "        register(\n",
    "            \"posterior_mean_coef1\",\n",
    "            betas * torch.sqrt(alphas_cumprod_prev) / (1.0 - alphas_cumprod),\n",
    "        )\n",
    "        register(\n",
    "            \"posterior_mean_coef2\",\n",
    "            (1.0 - alphas_cumprod_prev) * torch.sqrt(alphas) / (1.0 - alphas_cumprod),\n",
    "        )\n",
    "        register(\n",
    "            \"loss_weight\",\n",
    "            torch.sqrt(alphas) * torch.sqrt(1.0 - alphas_cumprod) / betas / 100,\n",
    "        )\n",
    "\n",
    "    def predict_noise_from_start(self, x_t, t, x0):\n",
    "        return (extract(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t - x0) / extract(\n",
    "            self.sqrt_recipm1_alphas_cumprod, t, x_t.shape\n",
    "        )\n",
    "\n",
    "    def q_posterior(self, x_start, x_t, t):\n",
    "        posterior_mean = (\n",
    "            extract(self.posterior_mean_coef1, t, x_t.shape) * x_start\n",
    "            + extract(self.posterior_mean_coef2, t, x_t.shape) * x_t\n",
    "        )\n",
    "        posterior_variance = extract(self.posterior_variance, t, x_t.shape)\n",
    "        posterior_log_variance = extract(self.posterior_log_variance_clipped, t, x_t.shape)\n",
    "        return posterior_mean, posterior_variance, posterior_log_variance\n",
    "\n",
    "    def output(self, x, t):\n",
    "        \"\"\"Model forward: predict x_0 as trend + season.\"\"\"\n",
    "        trend, season = self.model(x, t)\n",
    "        return trend + season\n",
    "\n",
    "    def model_predictions(self, x, t):\n",
    "        \"\"\"Predict x_0 (NO clamp -- returns are unbounded) and derive noise.\"\"\"\n",
    "        x_start = self.output(x, t)\n",
    "        pred_noise = self.predict_noise_from_start(x, t, x_start)\n",
    "        return pred_noise, x_start\n",
    "\n",
    "    def p_mean_variance(self, x, t):\n",
    "        \"\"\"Posterior mean and variance -- NO clamp on x_start.\"\"\"\n",
    "        _, x_start = self.model_predictions(x, t)\n",
    "        model_mean, posterior_variance, posterior_log_variance = self.q_posterior(\n",
    "            x_start=x_start, x_t=x, t=t\n",
    "        )\n",
    "        return model_mean, posterior_variance, posterior_log_variance, x_start\n",
    "\n",
    "    def p_sample(self, x, t: int, cond_fn=None, model_kwargs=None):\n",
    "        \"\"\"Single DDPM reverse step with optional classifier guidance.\"\"\"\n",
    "        batched_times = torch.full((x.shape[0],), t, device=x.device, dtype=torch.long)\n",
    "        model_mean, _, model_log_variance, x_start = self.p_mean_variance(x=x, t=batched_times)\n",
    "        noise = torch.randn_like(x) if t > 0 else 0.0\n",
    "        if cond_fn is not None:\n",
    "            model_mean = self.condition_mean(\n",
    "                cond_fn,\n",
    "                model_mean,\n",
    "                model_log_variance,\n",
    "                x,\n",
    "                t=batched_times,\n",
    "                model_kwargs=model_kwargs,\n",
    "            )\n",
    "        return model_mean + (0.5 * model_log_variance).exp() * noise, x_start\n",
    "\n",
    "    @torch.no_grad()\n",
    "    def sample(self, shape):\n",
    "        \"\"\"Full DDPM reverse sampling (all timesteps).\"\"\"\n",
    "        img = torch.randn(shape, device=self.betas.device)\n",
    "        for t in reversed(range(self.num_timesteps)):\n",
    "            img, _ = self.p_sample(img, t)\n",
    "        return img\n",
    "\n",
    "    @torch.no_grad()\n",
    "    def fast_sample(self, shape):\n",
    "        \"\"\"DDIM-style accelerated sampling.\"\"\"\n",
    "        batch, total_timesteps = shape[0], self.num_timesteps\n",
    "        sampling_timesteps, eta = self.sampling_timesteps, self.eta\n",
    "\n",
    "        times = torch.linspace(-1, total_timesteps - 1, steps=sampling_timesteps + 1)\n",
    "        times = list(reversed(times.int().tolist()))\n",
    "        time_pairs = list(zip(times[:-1], times[1:], strict=False))\n",
    "\n",
    "        img = torch.randn(shape, device=self.betas.device)\n",
    "\n",
    "        for time, time_next in time_pairs:\n",
    "            time_cond = torch.full((batch,), time, device=self.betas.device, dtype=torch.long)\n",
    "            pred_noise, x_start = self.model_predictions(img, time_cond)\n",
    "\n",
    "            if time_next < 0:\n",
    "                img = x_start\n",
    "                continue\n",
    "\n",
    "            alpha = self.alphas_cumprod[time]\n",
    "            alpha_next = self.alphas_cumprod[time_next]\n",
    "            sigma = eta * ((1 - alpha / alpha_next) * (1 - alpha_next) / (1 - alpha)).sqrt()\n",
    "            c = (1 - alpha_next - sigma**2).sqrt()\n",
    "            noise = torch.randn_like(img)\n",
    "            img = x_start * alpha_next.sqrt() + c * pred_noise + sigma * noise\n",
    "\n",
    "        return img\n",
    "\n",
    "    @torch.no_grad()\n",
    "    def fast_sample_cond(self, shape, cond_fn=None, model_kwargs=None, eta=None):\n",
    "        \"\"\"DDIM-style sampling with classifier guidance.\"\"\"\n",
    "        batch, total_timesteps = shape[0], self.num_timesteps\n",
    "        sampling_timesteps = self.sampling_timesteps\n",
    "        eta = eta if eta is not None else self.eta\n",
    "\n",
    "        times = torch.linspace(-1, total_timesteps - 1, steps=sampling_timesteps + 1)\n",
    "        times = list(reversed(times.int().tolist()))\n",
    "        time_pairs = list(zip(times[:-1], times[1:], strict=False))\n",
    "\n",
    "        img = torch.randn(shape, device=self.betas.device)\n",
    "\n",
    "        for time, time_next in time_pairs:\n",
    "            time_cond = torch.full((batch,), time, device=self.betas.device, dtype=torch.long)\n",
    "            pred_noise, x_start = self.model_predictions(img, time_cond)\n",
    "\n",
    "            if cond_fn is not None:\n",
    "                _, x_start = self.condition_score(\n",
    "                    cond_fn, x_start, img, time_cond, model_kwargs=model_kwargs\n",
    "                )\n",
    "                pred_noise = self.predict_noise_from_start(img, time_cond, x_start)\n",
    "\n",
    "            if time_next < 0:\n",
    "                img = x_start\n",
    "                continue\n",
    "\n",
    "            alpha = self.alphas_cumprod[time]\n",
    "            alpha_next = self.alphas_cumprod[time_next]\n",
    "            sigma = eta * ((1 - alpha / alpha_next) * (1 - alpha_next) / (1 - alpha)).sqrt()\n",
    "            c = (1 - alpha_next - sigma**2).sqrt()\n",
    "            noise = torch.randn_like(img)\n",
    "            img = x_start * alpha_next.sqrt() + c * pred_noise + sigma * noise\n",
    "\n",
    "        return img\n",
    "\n",
    "    @torch.no_grad()\n",
    "    def sample_cond(self, shape, cond_fn=None, model_kwargs=None, eta=None):\n",
    "        \"\"\"Full DDPM reverse sampling with classifier guidance.\n",
    "\n",
    "        Note: eta parameter is ignored for full DDPM (inherently stochastic).\n",
    "        \"\"\"\n",
    "        img = torch.randn(shape, device=self.betas.device)\n",
    "        for t in reversed(range(self.num_timesteps)):\n",
    "            img, _ = self.p_sample(img, t, cond_fn=cond_fn, model_kwargs=model_kwargs)\n",
    "        return img\n",
    "\n",
    "    def generate_mts(self, batch_size=16, model_kwargs=None, cond_fn=None, eta=None):\n",
    "        \"\"\"Entry point: generate multivariate time series.\n",
    "\n",
    "        Args:\n",
    "            eta: Stochasticity for DDIM sampling. eta=0 is deterministic, eta=1 is full DDPM.\n",
    "                 For conditional sampling, eta>0 adds diversity and prevents mode collapse.\n",
    "        \"\"\"\n",
    "        shape = (batch_size, self.seq_length, self.feature_size)\n",
    "        if cond_fn is not None:\n",
    "            sample_fn = self.fast_sample_cond if self.fast_sampling else self.sample_cond\n",
    "            return sample_fn(shape, cond_fn=cond_fn, model_kwargs=model_kwargs, eta=eta)\n",
    "        sample_fn = self.fast_sample if self.fast_sampling else self.sample\n",
    "        return sample_fn(shape)\n",
    "\n",
    "    @property\n",
    "    def loss_fn(self):\n",
    "        if self.loss_type == \"l1\":\n",
    "            return F.l1_loss\n",
    "        elif self.loss_type == \"l2\":\n",
    "            return F.mse_loss\n",
    "        raise ValueError(f\"invalid loss type {self.loss_type}\")\n",
    "\n",
    "    def q_sample(self, x_start, t, noise=None):\n",
    "        \"\"\"Forward process: add noise to x_0.\"\"\"\n",
    "        if noise is None:\n",
    "            noise = torch.randn_like(x_start)\n",
    "        return (\n",
    "            extract(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start\n",
    "            + extract(self.sqrt_one_minus_alphas_cumprod, t, x_start.shape) * noise\n",
    "        )\n",
    "\n",
    "    def _train_loss(self, x_start, t, target=None, noise=None):\n",
    "        \"\"\"Combined time-domain (L1) + frequency-domain (Fourier) loss.\"\"\"\n",
    "        if noise is None:\n",
    "            noise = torch.randn_like(x_start)\n",
    "        if target is None:\n",
    "            target = x_start\n",
    "\n",
    "        x = self.q_sample(x_start=x_start, t=t, noise=noise)\n",
    "        model_out = self.output(x, t)\n",
    "\n",
    "        train_loss = self.loss_fn(model_out, target, reduction=\"none\")\n",
    "\n",
    "        # Fourier loss: match spectral content\n",
    "        fft1 = torch.fft.fft(model_out.transpose(1, 2), norm=\"forward\")\n",
    "        fft2 = torch.fft.fft(target.transpose(1, 2), norm=\"forward\")\n",
    "        fft1, fft2 = fft1.transpose(1, 2), fft2.transpose(1, 2)\n",
    "        fourier_loss = self.loss_fn(\n",
    "            torch.real(fft1), torch.real(fft2), reduction=\"none\"\n",
    "        ) + self.loss_fn(torch.imag(fft1), torch.imag(fft2), reduction=\"none\")\n",
    "        train_loss = train_loss + self.ff_weight * fourier_loss\n",
    "\n",
    "        train_loss = reduce(train_loss, \"b ... -> b (...)\", \"mean\")\n",
    "        train_loss = train_loss * extract(self.loss_weight, t, train_loss.shape)\n",
    "        return train_loss.mean()\n",
    "\n",
    "    def forward(self, x, **kwargs):\n",
    "        b, c, n, device = *x.shape, x.device\n",
    "        assert n == self.feature_size, f\"expected {self.feature_size} features, got {n}\"\n",
    "        t = torch.randint(0, self.num_timesteps, (b,), device=device).long()\n",
    "        return self._train_loss(x_start=x, t=t, **kwargs)\n",
    "\n",
    "    def return_components(self, x, t: int):\n",
    "        \"\"\"Return trend, seasonal, residual decomposition for visualization.\"\"\"\n",
    "        b, c, n, device = *x.shape, x.device\n",
    "        t_tensor = torch.tensor([t]).repeat(b).to(device)\n",
    "        x_noised = self.q_sample(x, t_tensor)\n",
    "        trend, season, residual = self.model(x_noised, t_tensor, return_res=True)\n",
    "        return trend, season, residual, x_noised\n",
    "\n",
    "    def condition_mean(self, cond_fn, mean, log_variance, x, t, model_kwargs=None):\n",
    "        \"\"\"Shift mean by σ² · ∇_x log p(y|x) for classifier guidance.\"\"\"\n",
    "        gradient = cond_fn(x=x, t=t, **(model_kwargs or {}))\n",
    "        return mean.float() + torch.exp(log_variance) * gradient.float()\n",
    "\n",
    "    def condition_score(self, cond_fn, x_start, x, t, model_kwargs=None):\n",
    "        \"\"\"Score-based conditioning (Song et al. 2020).\"\"\"\n",
    "        alpha_bar = extract(self.alphas_cumprod, t, x.shape)\n",
    "        eps = self.predict_noise_from_start(x, t, x_start)\n",
    "        eps = eps - (1 - alpha_bar).sqrt() * cond_fn(x=x, t=t, **(model_kwargs or {}))\n",
    "        pred_xstart = (\n",
    "            extract(self.sqrt_recip_alphas_cumprod, t, x.shape) * x\n",
    "            - extract(self.sqrt_recipm1_alphas_cumprod, t, x.shape) * eps\n",
    "        )\n",
    "        model_mean, _, _ = self.q_posterior(x_start=pred_xstart, x_t=x, t=t)\n",
    "        return model_mean, pred_xstart"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c115736",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003404,
     "end_time": "2026-06-13T02:38:38.895709+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.892305+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 6. Exponential Moving Average\n",
    "\n",
    "EMA maintains a shadow copy of model weights that is updated as a running\n",
    "average: $\\theta_{\\text{ema}} \\leftarrow \\beta \\theta_{\\text{ema}} + (1-\\beta) \\theta$.\n",
    "Sampling from the EMA model produces smoother, higher-quality outputs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "780d0dcf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.903147Z",
     "iopub.status.busy": "2026-06-13T02:38:38.903070Z",
     "iopub.status.idle": "2026-06-13T02:38:38.905577Z",
     "shell.execute_reply": "2026-06-13T02:38:38.905208Z"
    },
    "papermill": {
     "duration": 0.006718,
     "end_time": "2026-06-13T02:38:38.905785+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.899067+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class EMA:\n",
    "    \"\"\"Simple exponential moving average of model parameters.\"\"\"\n",
    "\n",
    "    def __init__(self, model, decay=0.995, update_every=10):\n",
    "        self.decay = decay\n",
    "        self.update_every = update_every\n",
    "        self.step = 0\n",
    "        self.ema_model = deepcopy(model)\n",
    "        self.ema_model.eval()\n",
    "        for p in self.ema_model.parameters():\n",
    "            p.requires_grad_(False)\n",
    "\n",
    "    def update(self, model):\n",
    "        self.step += 1\n",
    "        if self.step % self.update_every != 0:\n",
    "            return\n",
    "        with torch.no_grad():\n",
    "            for ema_p, model_p in zip(\n",
    "                self.ema_model.parameters(), model.parameters(), strict=False\n",
    "            ):\n",
    "                ema_p.data.mul_(self.decay).add_(model_p.data, alpha=1.0 - self.decay)\n",
    "\n",
    "    def to(self, device):\n",
    "        self.ema_model = self.ema_model.to(device)\n",
    "        return self"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2c202ec",
   "metadata": {
    "papermill": {
     "duration": 0.003355,
     "end_time": "2026-06-13T02:38:38.912584+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.909229+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 7. Training\n",
    "\n",
    "Training follows the standard diffusion objective: sample a timestep $t$,\n",
    "add noise to create $x_t$, predict $\\hat{x}_0$, and minimize the combined\n",
    "time-domain + Fourier loss. We use gradient clipping, warmup scheduler,\n",
    "gradient accumulation, and EMA for stable convergence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4316a048",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.920043Z",
     "iopub.status.busy": "2026-06-13T02:38:38.919973Z",
     "iopub.status.idle": "2026-06-13T02:38:38.930788Z",
     "shell.execute_reply": "2026-06-13T02:38:38.930391Z"
    },
    "papermill": {
     "duration": 0.015111,
     "end_time": "2026-06-13T02:38:38.931198+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.916087+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normalized: mean=-0.0046, std=1.0091\n"
     ]
    }
   ],
   "source": [
    "# Normalize data before training\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(returns)  # Fit on raw returns, not sequences\n",
    "\n",
    "seq_shape = sequences.shape\n",
    "sequences_flat = sequences.reshape(-1, seq_shape[-1])\n",
    "sequences_norm = scaler.transform(sequences_flat).reshape(seq_shape).astype(np.float32)\n",
    "\n",
    "print(f\"Normalized: mean={sequences_norm.mean():.4f}, std={sequences_norm.std():.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d3352d8d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:38.939104Z",
     "iopub.status.busy": "2026-06-13T02:38:38.939025Z",
     "iopub.status.idle": "2026-06-13T02:38:39.075466Z",
     "shell.execute_reply": "2026-06-13T02:38:39.074946Z"
    },
    "papermill": {
     "duration": 0.141062,
     "end_time": "2026-06-13T02:38:39.075909+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:38.934847+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model parameters: 517,716\n",
      "DDIM: 500 training steps → 50 sampling steps\n"
     ]
    }
   ],
   "source": [
    "# Initialize model\n",
    "diffusion_model = DiffusionTS(\n",
    "    seq_length=CONFIG[\"seq_length\"],\n",
    "    feature_size=n_assets,\n",
    "    n_layer_enc=CONFIG[\"n_layer_enc\"],\n",
    "    n_layer_dec=CONFIG[\"n_layer_dec\"],\n",
    "    d_model=CONFIG[\"d_model\"],\n",
    "    timesteps=CONFIG[\"timesteps\"],\n",
    "    sampling_timesteps=CONFIG[\"sampling_timesteps\"],\n",
    "    eta=CONFIG[\"eta\"],\n",
    "    loss_type=CONFIG[\"loss_type\"],\n",
    "    n_heads=CONFIG[\"n_heads\"],\n",
    ").to(device)\n",
    "\n",
    "n_params = sum(p.numel() for p in diffusion_model.parameters())\n",
    "print(f\"Model parameters: {n_params:,}\")\n",
    "print(f\"DDIM: {CONFIG['timesteps']} training steps → {CONFIG['sampling_timesteps']} sampling steps\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "364a7552",
   "metadata": {
    "papermill": {
     "duration": 0.003606,
     "end_time": "2026-06-13T02:38:39.083220+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.079614+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Checkpoint Loading\n",
    "\n",
    "If a trained model exists and `RETRAIN=False`, we skip training and load\n",
    "the saved weights. This allows iterating on evaluation/visualization\n",
    "without retraining."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "a5fa4089",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.090858Z",
     "iopub.status.busy": "2026-06-13T02:38:39.090782Z",
     "iopub.status.idle": "2026-06-13T02:38:39.125808Z",
     "shell.execute_reply": "2026-06-13T02:38:39.125401Z"
    },
    "papermill": {
     "duration": 0.039648,
     "end_time": "2026-06-13T02:38:39.126290+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.086642+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading checkpoint from 05_synthetic_data/output/diffusion_ts/checkpoints/diffusion_ts_model.pt\n",
      "Loaded model trained for 10000 epochs\n"
     ]
    }
   ],
   "source": [
    "# Check for existing checkpoint\n",
    "checkpoint_exists = CHECKPOINT_PATH.exists()\n",
    "training_losses = []\n",
    "\n",
    "if checkpoint_exists and not RETRAIN:\n",
    "    print(f\"Loading checkpoint from {CHECKPOINT_PATH}\")\n",
    "    checkpoint = torch.load(CHECKPOINT_PATH, map_location=device, weights_only=False)\n",
    "    diffusion_model.load_state_dict(checkpoint[\"model_state\"])\n",
    "    scaler = StandardScaler()\n",
    "    scaler.mean_ = np.array(checkpoint[\"scaler_mean\"])\n",
    "    scaler.scale_ = np.array(checkpoint[\"scaler_scale\"])\n",
    "    training_losses = checkpoint.get(\"training_losses\", [])\n",
    "    print(f\"Loaded model trained for {len(training_losses)} epochs\")\n",
    "\n",
    "    # Create EMA wrapper with loaded weights\n",
    "    ema = EMA(diffusion_model, decay=CONFIG[\"ema_decay\"]).to(device)\n",
    "    ema.ema_model.load_state_dict(checkpoint[\"ema_state\"])\n",
    "    SKIP_TRAINING = True\n",
    "else:\n",
    "    if checkpoint_exists:\n",
    "        print(\"RETRAIN=True: Ignoring existing checkpoint\")\n",
    "    else:\n",
    "        print(f\"No checkpoint found at {CHECKPOINT_PATH}\")\n",
    "    SKIP_TRAINING = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "cde2c7c1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.136411Z",
     "iopub.status.busy": "2026-06-13T02:38:39.136319Z",
     "iopub.status.idle": "2026-06-13T02:38:39.139666Z",
     "shell.execute_reply": "2026-06-13T02:38:39.139232Z"
    },
    "papermill": {
     "duration": 0.008084,
     "end_time": "2026-06-13T02:38:39.140174+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.132090+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Create tensor data and utilities (needed for visualization/classifier even when loading from checkpoint)\n",
    "tensor_data = torch.FloatTensor(sequences_norm).to(device)\n",
    "grad_accum = CONFIG[\"gradient_accumulate_every\"]\n",
    "\n",
    "\n",
    "def cycle(dl):\n",
    "    \"\"\"Infinite iterator over a dataloader.\"\"\"\n",
    "    while True:\n",
    "        yield from dl"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69341220",
   "metadata": {
    "papermill": {
     "duration": 0.003431,
     "end_time": "2026-06-13T02:38:39.147284+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.143853+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Training (or Loading from Checkpoint)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "e77355c6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.154841Z",
     "iopub.status.busy": "2026-06-13T02:38:39.154713Z",
     "iopub.status.idle": "2026-06-13T02:38:39.157762Z",
     "shell.execute_reply": "2026-06-13T02:38:39.157390Z"
    },
    "papermill": {
     "duration": 0.007371,
     "end_time": "2026-06-13T02:38:39.158080+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.150709+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Training setup: optimizer, scheduler, dataloader\n",
    "if not SKIP_TRAINING:\n",
    "    ema = EMA(diffusion_model, decay=CONFIG[\"ema_decay\"]).to(device)\n",
    "\n",
    "    optimizer = torch.optim.Adam(diffusion_model.parameters(), lr=CONFIG[\"lr\"], betas=(0.9, 0.96))\n",
    "    scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(\n",
    "        optimizer, factor=0.5, patience=SCHEDULER_PATIENCE, min_lr=1e-5, threshold=0.1\n",
    "    )\n",
    "\n",
    "    warmup_steps = CONFIG[\"warmup_steps\"]\n",
    "    warmup_lr = CONFIG[\"warmup_lr\"]\n",
    "    base_lr = CONFIG[\"lr\"]\n",
    "\n",
    "    dataset = TensorDataset(tensor_data)\n",
    "    dataloader = DataLoader(dataset, batch_size=CONFIG[\"batch_size\"], shuffle=True, drop_last=True)\n",
    "    data_iter = cycle(dataloader)\n",
    "\n",
    "    print(f\"Training samples: {len(dataset):,}\")\n",
    "    print(f\"Effective batch size: {CONFIG['batch_size'] * grad_accum}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2763e02a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.165847Z",
     "iopub.status.busy": "2026-06-13T02:38:39.165735Z",
     "iopub.status.idle": "2026-06-13T02:38:39.168691Z",
     "shell.execute_reply": "2026-06-13T02:38:39.168421Z"
    },
    "papermill": {
     "duration": 0.007485,
     "end_time": "2026-06-13T02:38:39.169035+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.161550+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Training loop\n",
    "if not SKIP_TRAINING:\n",
    "    print(f\"Training Diffusion-TS for {CONFIG['epochs']} epochs...\")\n",
    "\n",
    "    training_losses = []\n",
    "    for step in range(CONFIG[\"epochs\"]):\n",
    "        diffusion_model.train()\n",
    "        total_loss = 0.0\n",
    "\n",
    "        # Warmup LR\n",
    "        if step < warmup_steps:\n",
    "            lr = base_lr + (warmup_lr - base_lr) * step / max(warmup_steps, 1)\n",
    "            for pg in optimizer.param_groups:\n",
    "                pg[\"lr\"] = lr\n",
    "\n",
    "        # Gradient accumulation\n",
    "        for _ in range(grad_accum):\n",
    "            (batch,) = next(data_iter)\n",
    "            loss = diffusion_model(batch, target=batch)\n",
    "            loss = loss / grad_accum\n",
    "            loss.backward()\n",
    "            total_loss += loss.item()\n",
    "\n",
    "        torch.nn.utils.clip_grad_norm_(diffusion_model.parameters(), 1.0)\n",
    "        optimizer.step()\n",
    "        if step >= warmup_steps:\n",
    "            scheduler.step(total_loss)\n",
    "        optimizer.zero_grad()\n",
    "        ema.update(diffusion_model)\n",
    "        training_losses.append(total_loss)\n",
    "\n",
    "        if (step + 1) % 100 == 0 or step == 0:\n",
    "            current_lr = optimizer.param_groups[0][\"lr\"]\n",
    "            print(\n",
    "                f\"  Epoch {step + 1}/{CONFIG['epochs']}: Loss = {total_loss:.6f}, LR = {current_lr:.2e}\",\n",
    "                flush=True,\n",
    "            )\n",
    "\n",
    "    print(f\"Training complete. Final loss: {training_losses[-1]:.6f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "16b60cb8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.176413Z",
     "iopub.status.busy": "2026-06-13T02:38:39.176351Z",
     "iopub.status.idle": "2026-06-13T02:38:39.178092Z",
     "shell.execute_reply": "2026-06-13T02:38:39.177814Z"
    },
    "papermill": {
     "duration": 0.005766,
     "end_time": "2026-06-13T02:38:39.178309+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.172543+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Save checkpoint\n",
    "if not SKIP_TRAINING:\n",
    "    checkpoint = {\n",
    "        \"model_state\": diffusion_model.state_dict(),\n",
    "        \"ema_state\": ema.ema_model.state_dict(),\n",
    "        \"scaler_mean\": scaler.mean_.tolist(),\n",
    "        \"scaler_scale\": scaler.scale_.tolist(),\n",
    "        \"training_losses\": training_losses,\n",
    "        \"config\": CONFIG,\n",
    "    }\n",
    "    torch.save(checkpoint, CHECKPOINT_PATH)\n",
    "    print(f\"Saved checkpoint to {CHECKPOINT_PATH}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "055641df",
   "metadata": {
    "papermill": {
     "duration": 0.003405,
     "end_time": "2026-06-13T02:38:39.185166+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.181761+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Training Progress"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "1e572e5e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.192642Z",
     "iopub.status.busy": "2026-06-13T02:38:39.192576Z",
     "iopub.status.idle": "2026-06-13T02:38:39.296279Z",
     "shell.execute_reply": "2026-06-13T02:38:39.295717Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.107846,
     "end_time": "2026-06-13T02:38:39.296576+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.188730+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if training_losses:\n",
    "    fig, ax = plt.subplots(figsize=(8, 4))\n",
    "    ax.plot(training_losses, linewidth=1)\n",
    "    ax.set_yscale(\"log\")\n",
    "    ax.set_xlabel(\"Epoch\")\n",
    "    ax.set_ylabel(\"Loss (L1 + Fourier)\")\n",
    "    ax.set_title(\"Diffusion-TS Training Progress\")\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "else:\n",
    "    print(\"No training losses available (loaded from checkpoint)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a7ac883",
   "metadata": {
    "papermill": {
     "duration": 0.003617,
     "end_time": "2026-06-13T02:38:39.304010+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.300393+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 8. Generate Synthetic Sequences\n",
    "\n",
    "We sample from the EMA model using DDIM fast sampling (50 steps instead\n",
    "of 500). The samples are generated in normalized space and then\n",
    "denormalized back to original return scale."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "67940aa5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:39.311891Z",
     "iopub.status.busy": "2026-06-13T02:38:39.311802Z",
     "iopub.status.idle": "2026-06-13T02:38:41.345528Z",
     "shell.execute_reply": "2026-06-13T02:38:41.344898Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 2.03847,
     "end_time": "2026-06-13T02:38:41.346048+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:39.307578+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Generating 500 synthetic sequences via DDIM (50 steps)...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Normalized space (before scaling):\n",
      "  Training std:  1.0091\n",
      "  Synthetic std: 0.6799\n",
      "  Ratio:         67.38%\n",
      "\n",
      "Applied variance scaling: 1.484x\n",
      "  Scaled std:    1.0091\n",
      "\n",
      "Denormalized (return space):\n",
      "  Synthetic: mean=0.000524, std=0.011658\n",
      "  Real:      mean=0.000195, std=0.011741\n"
     ]
    }
   ],
   "source": [
    "# N_SYNTHETIC is set in the parameters cell above\n",
    "\n",
    "print(\n",
    "    f\"Generating {N_SYNTHETIC} synthetic sequences via DDIM ({CONFIG['sampling_timesteps']} steps)...\"\n",
    ")\n",
    "synthetic_norm = ema.ema_model.generate_mts(batch_size=N_SYNTHETIC).detach().cpu().numpy()\n",
    "\n",
    "# Diagnostic: compare normalized variance\n",
    "print(\"\\nNormalized space (before scaling):\")\n",
    "print(f\"  Training std:  {sequences_norm.std():.4f}\")\n",
    "print(f\"  Synthetic std: {synthetic_norm.std():.4f}\")\n",
    "variance_ratio = synthetic_norm.std() / sequences_norm.std()\n",
    "print(f\"  Ratio:         {variance_ratio:.2%}\")\n",
    "\n",
    "# Variance scaling: Diffusion-TS with trend+seasonal decomposition tends to underestimate\n",
    "# variance. We scale outputs to match training distribution variance.\n",
    "# This scale_factor is also applied to regime-conditional samples below.\n",
    "if variance_ratio < 0.9:\n",
    "    VARIANCE_SCALE_FACTOR = sequences_norm.std() / synthetic_norm.std()\n",
    "    synthetic_norm = synthetic_norm * VARIANCE_SCALE_FACTOR\n",
    "    print(f\"\\nApplied variance scaling: {VARIANCE_SCALE_FACTOR:.3f}x\")\n",
    "    print(f\"  Scaled std:    {synthetic_norm.std():.4f}\")\n",
    "else:\n",
    "    VARIANCE_SCALE_FACTOR = 1.0\n",
    "\n",
    "# Denormalize\n",
    "syn_shape = synthetic_norm.shape\n",
    "synthetic_flat = synthetic_norm.reshape(-1, syn_shape[-1])\n",
    "synthetic_sequences = scaler.inverse_transform(synthetic_flat).reshape(syn_shape).astype(np.float32)\n",
    "\n",
    "print(\"\\nDenormalized (return space):\")\n",
    "print(f\"  Synthetic: mean={synthetic_sequences.mean():.6f}, std={synthetic_sequences.std():.6f}\")\n",
    "print(f\"  Real:      mean={sequences.mean():.6f}, std={sequences.std():.6f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c62d621",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.00362,
     "end_time": "2026-06-13T02:38:41.353652+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:41.350032+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 9. Unconditional Evaluation\n",
    "\n",
    "We evaluate the generated data on three axes:\n",
    "\n",
    "### Statistical Tests\n",
    "\n",
    "- **Kolmogorov-Smirnov (KS) test**: Measures the maximum distance between\n",
    "  two empirical CDFs. Values range from 0 (identical distributions) to 1\n",
    "  (completely different). For returns, KS < 0.1 indicates good marginal fit.\n",
    "\n",
    "- **Correlation error**: Mean absolute difference between real and synthetic\n",
    "  cross-asset correlation matrices. Captures whether the model learned\n",
    "  dependence structure (e.g., sector correlations).\n",
    "\n",
    "- **Autocorrelation error**: Compares lag-1 autocorrelation. Returns have\n",
    "  near-zero AC (weak form efficiency) but volatility clusters (squared returns\n",
    "  have positive AC). Good generators preserve these stylized facts.\n",
    "\n",
    "### Visual Comparison\n",
    "\n",
    "- **PCA/t-SNE**: Project high-dimensional sequences to 2D. Real and synthetic\n",
    "  distributions should overlap if the generator captures the data manifold.\n",
    "\n",
    "### Utility (TSTR)\n",
    "\n",
    "- **Train-Synthetic-Test-Real**: Train a classifier on synthetic data, test\n",
    "  on real data. High accuracy means synthetic data is useful for downstream\n",
    "  ML tasks -- the ultimate practical validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "17b3ff27",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:41.361571Z",
     "iopub.status.busy": "2026-06-13T02:38:41.361467Z",
     "iopub.status.idle": "2026-06-13T02:38:41.647286Z",
     "shell.execute_reply": "2026-06-13T02:38:41.646741Z"
    },
    "papermill": {
     "duration": 0.290712,
     "end_time": "2026-06-13T02:38:41.647975+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:41.357263+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Statistical Evaluation ===\n",
      "  mean_ks_statistic: 0.0642\n",
      "  mean_error: 0.0003\n",
      "  std_error: 0.0008\n",
      "  correlation_error: 0.0368\n",
      "  autocorrelation_error: 0.0464\n"
     ]
    }
   ],
   "source": [
    "def evaluate_statistics(real_data: np.ndarray, synthetic_data: np.ndarray) -> dict:\n",
    "    \"\"\"Compare distributional properties of real and synthetic data.\"\"\"\n",
    "    n_assets = real_data.shape[2]\n",
    "    real_flat = real_data.reshape(-1, n_assets)\n",
    "    syn_flat = synthetic_data.reshape(-1, n_assets)\n",
    "\n",
    "    # KS test per asset\n",
    "    ks_stats = [stats.ks_2samp(real_flat[:, i], syn_flat[:, i])[0] for i in range(n_assets)]\n",
    "\n",
    "    # Correlation matrix comparison\n",
    "    real_corr = np.corrcoef(real_flat.T)\n",
    "    syn_corr = np.corrcoef(syn_flat.T)\n",
    "    corr_error = np.mean(np.abs(real_corr - syn_corr))\n",
    "\n",
    "    # Autocorrelation (lag-1) per asset\n",
    "    def autocorr(x, lag=1):\n",
    "        return np.corrcoef(x[:-lag], x[lag:])[0, 1]\n",
    "\n",
    "    real_ac = [autocorr(real_flat[:, i]) for i in range(n_assets)]\n",
    "    syn_ac = [autocorr(syn_flat[:, i]) for i in range(n_assets)]\n",
    "    ac_error = np.mean(np.abs(np.array(real_ac) - np.array(syn_ac)))\n",
    "\n",
    "    return {\n",
    "        \"mean_ks_statistic\": np.mean(ks_stats),\n",
    "        \"mean_error\": np.mean(np.abs(real_flat.mean(0) - syn_flat.mean(0))),\n",
    "        \"std_error\": np.mean(np.abs(real_flat.std(0) - syn_flat.std(0))),\n",
    "        \"correlation_error\": corr_error,\n",
    "        \"autocorrelation_error\": ac_error,\n",
    "    }\n",
    "\n",
    "\n",
    "stats_results = evaluate_statistics(sequences, synthetic_sequences)\n",
    "\n",
    "print(\"\\n=== Statistical Evaluation ===\")\n",
    "for key, value in stats_results.items():\n",
    "    print(f\"  {key}: {value:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1dcf60f",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003682,
     "end_time": "2026-06-13T02:38:41.655699+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:41.652017+00:00",
     "status": "completed"
    }
   },
   "source": [
    "**Interpretation**: Low KS statistics indicate that marginal distributions\n",
    "per asset are well-matched. Correlation error tests cross-asset dependence\n",
    "preservation. Autocorrelation error verifies that the model reproduces\n",
    "the weak serial dependence and volatility clustering of daily returns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "e4638685",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:41.664097Z",
     "iopub.status.busy": "2026-06-13T02:38:41.663995Z",
     "iopub.status.idle": "2026-06-13T02:38:43.171125Z",
     "shell.execute_reply": "2026-06-13T02:38:43.170571Z"
    },
    "papermill": {
     "duration": 1.512477,
     "end_time": "2026-06-13T02:38:43.171984+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:41.659507+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "findfont: Failed to find font weight semibold, now using 700.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_fidelity_comparison(\n",
    "    sequences,\n",
    "    synthetic_sequences,\n",
    "    title=\"Diffusion-TS: Real vs Synthetic Distribution\",\n",
    "    n_samples=500,\n",
    "    flatten_method=\"flatten\",  # Flatten all timesteps for full sequence comparison\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8cbe50d7",
   "metadata": {
    "papermill": {
     "duration": 0.004185,
     "end_time": "2026-06-13T02:38:43.180696+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:43.176511+00:00",
     "status": "completed"
    }
   },
   "source": [
    "**Interpretation**: Overlapping PCA/t-SNE point clouds confirm that synthetic\n",
    "sequences occupy the same region of feature space as real data. Gaps or\n",
    "isolated clusters would indicate missing regimes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b393d22",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.003936,
     "end_time": "2026-06-13T02:38:43.188559+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:43.184623+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### TSTR Evaluation\n",
    "\n",
    "We evaluate downstream utility via **extreme-move classification**: predict\n",
    "whether the next-day absolute return exceeds the 90th percentile.\n",
    "\n",
    "**Important context**: This is a highly imbalanced task -- by definition only\n",
    "10% of samples are positive (extreme moves). A naive \"always predict normal\"\n",
    "classifier achieves **90% accuracy** as baseline. The meaningful metric is\n",
    "**TSTR Ratio** (synthetic/real), not raw accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "4de5e693",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:43.197278Z",
     "iopub.status.busy": "2026-06-13T02:38:43.197173Z",
     "iopub.status.idle": "2026-06-13T02:38:44.200812Z",
     "shell.execute_reply": "2026-06-13T02:38:44.200545Z"
    },
    "papermill": {
     "duration": 1.009848,
     "end_time": "2026-06-13T02:38:44.202348+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:43.192500+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== TSTR Evaluation: Extreme-Move Classification ===\n",
      "  Task: Predict |return| > 90th percentile\n",
      "  Test samples: 443 (4.1% positive)\n",
      "  Naive baseline: 95.9% (always predict 'normal')\n",
      "\n",
      "  Metric         Real-Trained  Synth-Trained\n",
      "  ------------------------------------------\n",
      "  Accuracy              93.7%          93.5%\n",
      "  Precision              0.0%           7.7%\n",
      "  Recall                 0.0%           5.6%\n",
      "  F1                     0.0%           6.5%\n",
      "\n",
      "  TSTR Ratio: 0.998\n"
     ]
    }
   ],
   "source": [
    "def tstr_evaluation(train_data, holdout_data, synthetic_data):\n",
    "    \"\"\"TSTR on extreme-move classification (90th percentile threshold).\"\"\"\n",
    "    from sklearn.linear_model import LogisticRegression\n",
    "    from sklearn.metrics import precision_recall_fscore_support\n",
    "\n",
    "    train_returns = train_data[:, -1, 0]\n",
    "    threshold = np.percentile(np.abs(train_returns), 90)\n",
    "\n",
    "    X_train_real = train_data[:, :-1, :].reshape(len(train_data), -1)\n",
    "    y_train_real = (np.abs(train_returns) > threshold).astype(int)\n",
    "\n",
    "    X_train_syn = synthetic_data[:, :-1, :].reshape(len(synthetic_data), -1)\n",
    "    y_train_syn = (np.abs(synthetic_data[:, -1, 0]) > threshold).astype(int)\n",
    "\n",
    "    X_test = holdout_data[:, :-1, :].reshape(len(holdout_data), -1)\n",
    "    y_test = (np.abs(holdout_data[:, -1, 0]) > threshold).astype(int)\n",
    "\n",
    "    scaler_r = StandardScaler()\n",
    "    X_tr_s = scaler_r.fit_transform(X_train_real)\n",
    "    X_te_r = scaler_r.transform(X_test)\n",
    "\n",
    "    scaler_s = StandardScaler()\n",
    "    X_ts_s = scaler_s.fit_transform(X_train_syn)\n",
    "    X_te_s = scaler_s.transform(X_test)\n",
    "\n",
    "    model_r = LogisticRegression(max_iter=1000)\n",
    "    model_r.fit(X_tr_s, y_train_real)\n",
    "    acc_real = model_r.score(X_te_r, y_test)\n",
    "    y_pred_real = model_r.predict(X_te_r)\n",
    "    prec_r, rec_r, f1_r, _ = precision_recall_fscore_support(\n",
    "        y_test, y_pred_real, average=\"binary\", zero_division=0\n",
    "    )\n",
    "\n",
    "    if len(np.unique(y_train_syn)) < 2:\n",
    "        acc_syn = (y_test == int(y_train_syn.mean() > 0.5)).mean()\n",
    "        prec_s, rec_s, f1_s = 0, 0, 0\n",
    "    else:\n",
    "        model_s = LogisticRegression(max_iter=1000)\n",
    "        model_s.fit(X_ts_s, y_train_syn)\n",
    "        acc_syn = model_s.score(X_te_s, y_test)\n",
    "        y_pred_syn = model_s.predict(X_te_s)\n",
    "        prec_s, rec_s, f1_s, _ = precision_recall_fscore_support(\n",
    "            y_test, y_pred_syn, average=\"binary\", zero_division=0\n",
    "        )\n",
    "\n",
    "    return {\n",
    "        \"accuracy_real\": acc_real,\n",
    "        \"accuracy_synthetic\": acc_syn,\n",
    "        \"precision_real\": prec_r,\n",
    "        \"precision_synthetic\": prec_s,\n",
    "        \"recall_real\": rec_r,\n",
    "        \"recall_synthetic\": rec_s,\n",
    "        \"f1_real\": f1_r,\n",
    "        \"f1_synthetic\": f1_s,\n",
    "        \"tstr_ratio\": acc_syn / acc_real if acc_real > 0 else 0,\n",
    "        \"positive_rate\": y_test.mean(),\n",
    "        \"baseline_accuracy\": 1 - y_test.mean(),\n",
    "        \"n_test_samples\": len(y_test),\n",
    "    }\n",
    "\n",
    "\n",
    "tstr_results = tstr_evaluation(sequences, holdout_sequences, synthetic_sequences)\n",
    "\n",
    "print(\"\\n=== TSTR Evaluation: Extreme-Move Classification ===\")\n",
    "print(\"  Task: Predict |return| > 90th percentile\")\n",
    "print(\n",
    "    f\"  Test samples: {tstr_results['n_test_samples']:,} ({tstr_results['positive_rate']:.1%} positive)\"\n",
    ")\n",
    "print(f\"  Naive baseline: {tstr_results['baseline_accuracy']:.1%} (always predict 'normal')\")\n",
    "print()\n",
    "print(f\"  {'Metric':<12} {'Real-Trained':>14} {'Synth-Trained':>14}\")\n",
    "print(f\"  {'-' * 42}\")\n",
    "print(\n",
    "    f\"  {'Accuracy':<12} {tstr_results['accuracy_real']:>14.1%} {tstr_results['accuracy_synthetic']:>14.1%}\"\n",
    ")\n",
    "print(\n",
    "    f\"  {'Precision':<12} {tstr_results['precision_real']:>14.1%} {tstr_results['precision_synthetic']:>14.1%}\"\n",
    ")\n",
    "print(\n",
    "    f\"  {'Recall':<12} {tstr_results['recall_real']:>14.1%} {tstr_results['recall_synthetic']:>14.1%}\"\n",
    ")\n",
    "print(f\"  {'F1':<12} {tstr_results['f1_real']:>14.1%} {tstr_results['f1_synthetic']:>14.1%}\")\n",
    "print()\n",
    "print(f\"  TSTR Ratio: {tstr_results['tstr_ratio']:.3f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72113f97",
   "metadata": {
    "papermill": {
     "duration": 0.016411,
     "end_time": "2026-06-13T02:38:44.231707+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.215296+00:00",
     "status": "completed"
    }
   },
   "source": [
    "**Interpretation**: Raw accuracy is misleading here (90% baseline). The **TSTR\n",
    "Ratio** is the key metric: a ratio near 1.0 means synthetic-trained models\n",
    "perform comparably to real-trained ones. Precision/recall for the extreme-move\n",
    "class reveal how well each model identifies rare large moves.\n",
    "\n",
    "**Trading context**: Extreme-move prediction exploits volatility clustering.\n",
    "High recall means we catch most extreme moves; high precision means fewer\n",
    "false alarms. TSTR ≥ 0.9 indicates synthetic data preserves this structure."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "684b6084",
   "metadata": {
    "papermill": {
     "duration": 0.016401,
     "end_time": "2026-06-13T02:38:44.261607+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.245206+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Sample Sequences and Decomposition\n",
    "\n",
    "Visualize example synthetic paths and the model's internal trend+seasonal\n",
    "decomposition. The decomposition shows what the model attributes to slow\n",
    "drift versus periodic patterns at a given noise level."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "88022575",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:44.282918Z",
     "iopub.status.busy": "2026-06-13T02:38:44.282816Z",
     "iopub.status.idle": "2026-06-13T02:38:44.502506Z",
     "shell.execute_reply": "2026-06-13T02:38:44.502075Z"
    },
    "papermill": {
     "duration": 0.22861,
     "end_time": "2026-06-13T02:38:44.503515+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.274905+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sample paths: Daily returns (left) and Cumulative returns (right)\n",
    "# Daily returns show moment-to-moment behavior; cumulative reveals drift patterns\n",
    "n_sample_paths = 10\n",
    "\n",
    "fig, axes = plt.subplots(2, 2, figsize=(12, 6))\n",
    "\n",
    "# Top row: Real data\n",
    "for i in range(n_sample_paths):\n",
    "    axes[0, 0].plot(sequences[i, :, 0], color=COLORS[\"blue\"], alpha=0.4, linewidth=0.8)\n",
    "    axes[0, 1].plot(sequences[i, :, 0].cumsum(), color=COLORS[\"blue\"], alpha=0.4, linewidth=0.8)\n",
    "axes[0, 0].set_ylabel(\"Daily Return\")\n",
    "axes[0, 0].set_title(\"Real - Daily Returns\")\n",
    "axes[0, 1].set_ylabel(\"Cumulative Return\")\n",
    "axes[0, 1].set_title(\"Real - Cumulative\")\n",
    "for ax in axes[0]:\n",
    "    ax.axhline(0, color=COLORS[\"neutral\"], linestyle=\"--\", linewidth=0.5, alpha=0.5)\n",
    "\n",
    "# Bottom row: Synthetic data\n",
    "for i in range(n_sample_paths):\n",
    "    axes[1, 0].plot(synthetic_sequences[i, :, 0], color=COLORS[\"copper\"], alpha=0.4, linewidth=0.8)\n",
    "    axes[1, 1].plot(\n",
    "        synthetic_sequences[i, :, 0].cumsum(), color=COLORS[\"copper\"], alpha=0.4, linewidth=0.8\n",
    "    )\n",
    "axes[1, 0].set_ylabel(\"Daily Return\")\n",
    "axes[1, 0].set_xlabel(\"Time Step\")\n",
    "axes[1, 0].set_title(\"Synthetic - Daily Returns\")\n",
    "axes[1, 1].set_ylabel(\"Cumulative Return\")\n",
    "axes[1, 1].set_xlabel(\"Time Step\")\n",
    "axes[1, 1].set_title(\"Synthetic - Cumulative\")\n",
    "for ax in axes[1]:\n",
    "    ax.axhline(0, color=COLORS[\"neutral\"], linestyle=\"--\", linewidth=0.5, alpha=0.5)\n",
    "\n",
    "# Match y-axis for fair comparison\n",
    "for col in range(2):\n",
    "    ymin = min(axes[0, col].get_ylim()[0], axes[1, col].get_ylim()[0])\n",
    "    ymax = max(axes[0, col].get_ylim()[1], axes[1, col].get_ylim()[1])\n",
    "    axes[0, col].set_ylim(ymin, ymax)\n",
    "    axes[1, col].set_ylim(ymin, ymax)\n",
    "\n",
    "fig.suptitle(\"Real vs Synthetic Sample Paths (Asset 0)\", fontsize=12, y=1.02)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "860520e3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:44.515039Z",
     "iopub.status.busy": "2026-06-13T02:38:44.514932Z",
     "iopub.status.idle": "2026-06-13T02:38:44.654199Z",
     "shell.execute_reply": "2026-06-13T02:38:44.653651Z"
    },
    "papermill": {
     "duration": 0.145309,
     "end_time": "2026-06-13T02:38:44.654535+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.509226+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Trend + Seasonal decomposition visualization (Matplotlib, vertically stacked)\n",
    "with torch.no_grad():\n",
    "    sample_batch = tensor_data[:4]\n",
    "    # Show decomposition at moderate noise level (t = T/4)\n",
    "    t_vis = CONFIG[\"timesteps\"] // 4\n",
    "    trend, season, residual, x_noised = ema.ema_model.return_components(sample_batch, t_vis)\n",
    "\n",
    "asset_idx = 0\n",
    "components = [\n",
    "    (trend[0, :, asset_idx].cpu().numpy(), \"Trend\", COLORS[\"blue\"]),\n",
    "    (season[0, :, asset_idx].cpu().numpy(), \"Seasonal\", COLORS[\"amber\"]),\n",
    "    (residual[0, :, asset_idx].cpu().numpy(), \"Residual\", COLORS[\"copper\"]),\n",
    "]\n",
    "\n",
    "fig, axes = plt.subplots(3, 1, figsize=(8, 6), sharex=True)\n",
    "for ax, (data, label, color) in zip(axes, components, strict=False):\n",
    "    ax.plot(data, color=color, linewidth=1)\n",
    "    ax.set_ylabel(label)\n",
    "    ax.axhline(0, color=COLORS[\"neutral\"], linestyle=\"--\", linewidth=0.5, alpha=0.5)\n",
    "\n",
    "axes[-1].set_xlabel(\"Time Step\")\n",
    "fig.suptitle(\n",
    "    f\"Diffusion-TS Learned Decomposition at Intermediate Noise (t={t_vis}/{CONFIG['timesteps']})\",\n",
    "    fontsize=12,\n",
    "    y=1.02,\n",
    ")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13b5a782",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.005419,
     "end_time": "2026-06-13T02:38:44.665315+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.659896+00:00",
     "status": "completed"
    }
   },
   "source": [
    "**Interpretation**: The trend component captures slow drift (market direction),\n",
    "the seasonal component captures periodic oscillations (day-of-week effects,\n",
    "monthly cycles), and the residual captures high-frequency noise. This\n",
    "decomposition is analogous to classical STL but learned end-to-end within\n",
    "the diffusion framework. At higher noise levels (larger $t$), the\n",
    "decomposition becomes noisier as the model has less signal to work with."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2e8a716",
   "metadata": {
    "papermill": {
     "duration": 0.00493,
     "end_time": "2026-06-13T02:38:44.675343+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.670413+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 10. Regime-Conditional Generation\n",
    "\n",
    "This section implements **classifier guidance** for conditional diffusion:\n",
    "\n",
    "1. Fit a Gaussian HMM on SPY returns to identify market regimes\n",
    "2. Train a classifier on noised sequences labeled by regime\n",
    "3. At sampling time, compute $\\nabla_x \\log p(\\text{regime}|x_t)$ and\n",
    "   shift the denoising trajectory toward the target regime\n",
    "\n",
    "### Hidden Markov Models for Regime Detection\n",
    "\n",
    "A **Hidden Markov Model (HMM)** assumes the market cycles through unobserved\n",
    "\"states\" (regimes), each with its own return distribution. The Gaussian HMM\n",
    "assumes each regime has a distinct mean and variance. Given observed returns,\n",
    "the Viterbi algorithm infers the most likely regime sequence.\n",
    "\n",
    "Here we use a simple 2-state HMM as a label source for conditional generation.\n",
    "**Chapter 9 (Model-Based Feature Extraction)** covers HMMs in depth, including:\n",
    "- Multi-state regime models for volatility forecasting\n",
    "- Regime-switching strategies\n",
    "- Comparison with Markov-switching GARCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "a7c41098",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:44.685857Z",
     "iopub.status.busy": "2026-06-13T02:38:44.685754Z",
     "iopub.status.idle": "2026-06-13T02:38:45.072126Z",
     "shell.execute_reply": "2026-06-13T02:38:45.069159Z"
    },
    "papermill": {
     "duration": 0.393494,
     "end_time": "2026-06-13T02:38:45.073667+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:44.680173+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "HMM Regime Detection (SPY):\n",
      "  Low-Vol: 1,237 days, mean=0.0007, vol=0.0091\n",
      "  High-Vol: 155 days, mean=-0.0022, vol=0.0283\n"
     ]
    }
   ],
   "source": [
    "# Fit HMM on SPY returns (first asset) from training period\n",
    "spy_returns = returns[:, 0].reshape(-1, 1)\n",
    "\n",
    "hmm = GaussianHMM(\n",
    "    n_components=CONFIG[\"n_regimes\"], covariance_type=\"full\", n_iter=200, random_state=42\n",
    ")\n",
    "hmm.fit(spy_returns)\n",
    "\n",
    "regime_labels = hmm.predict(spy_returns)\n",
    "regime_means = hmm.means_.flatten()\n",
    "regime_vols = np.sqrt(hmm.covars_.flatten())\n",
    "\n",
    "# Sort regimes by volatility (0=low-vol, 1=normal, 2=crisis)\n",
    "sort_idx = np.argsort(regime_vols)\n",
    "label_map = {old: new for new, old in enumerate(sort_idx)}\n",
    "regime_labels = np.array([label_map[r] for r in regime_labels])\n",
    "\n",
    "regime_names = [\"Low-Vol\", \"High-Vol\"]  # 2 regimes for cleaner separation\n",
    "print(\"\\nHMM Regime Detection (SPY):\")\n",
    "for i in range(CONFIG[\"n_regimes\"]):\n",
    "    mask = regime_labels == i\n",
    "    print(\n",
    "        f\"  {regime_names[i]}: {mask.sum():,} days, \"\n",
    "        f\"mean={regime_means[sort_idx[i]]:.4f}, vol={regime_vols[sort_idx[i]]:.4f}\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8eb4f49c",
   "metadata": {
    "papermill": {
     "duration": 0.010227,
     "end_time": "2026-06-13T02:38:45.094147+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.083920+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Label Training Sequences by Majority Regime\n",
    "\n",
    "Each training sequence of length $L$ spans multiple days. We assign the\n",
    "regime label that covers the majority of days in the window."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "26364b57",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.114892Z",
     "iopub.status.busy": "2026-06-13T02:38:45.114766Z",
     "iopub.status.idle": "2026-06-13T02:38:45.120782Z",
     "shell.execute_reply": "2026-06-13T02:38:45.120428Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.019363,
     "end_time": "2026-06-13T02:38:45.123196+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.103833+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Sequence regime distribution (2 active regimes):\n",
      "  Low-Vol: 1178 sequences (88.4%)\n",
      "  High-Vol: 155 sequences (11.6%)\n"
     ]
    }
   ],
   "source": [
    "seq_len = CONFIG[\"seq_length\"]\n",
    "n_seq = len(returns) - seq_len + 1\n",
    "\n",
    "# Label by last day in window (preserves rare regime labels better than majority vote)\n",
    "seq_regime_labels = np.array([regime_labels[i + seq_len - 1] for i in range(n_seq)], dtype=np.int64)\n",
    "\n",
    "# Check for underrepresented regimes and merge if needed\n",
    "MIN_SEQUENCES = 20\n",
    "regime_counts = {i: (seq_regime_labels == i).sum() for i in range(CONFIG[\"n_regimes\"])}\n",
    "active_regimes = [i for i, c in regime_counts.items() if c >= MIN_SEQUENCES]\n",
    "\n",
    "if len(active_regimes) < CONFIG[\"n_regimes\"]:\n",
    "    print(f\"\\nWarning: Only {len(active_regimes)} regimes have >= {MIN_SEQUENCES} sequences\")\n",
    "    # Merge underrepresented regimes into the nearest active regime (by volatility)\n",
    "    for i in range(CONFIG[\"n_regimes\"]):\n",
    "        if i not in active_regimes:\n",
    "            # Merge into nearest active regime by volatility\n",
    "            nearest = min(\n",
    "                active_regimes,\n",
    "                key=lambda a: abs(regime_vols[sort_idx[a]] - regime_vols[sort_idx[i]]),\n",
    "            )\n",
    "            seq_regime_labels[seq_regime_labels == i] = nearest\n",
    "            print(\n",
    "                f\"  Merged {regime_names[i]} ({regime_counts[i]} seq) into {regime_names[nearest]}\"\n",
    "            )\n",
    "\n",
    "    # Remap labels to contiguous range\n",
    "    unique_labels = sorted(set(seq_regime_labels))\n",
    "    label_remap = {old: new for new, old in enumerate(unique_labels)}\n",
    "    seq_regime_labels = np.array([label_remap[l] for l in seq_regime_labels], dtype=np.int64)\n",
    "    n_active_regimes = len(unique_labels)\n",
    "    active_regime_names = [regime_names[l] for l in unique_labels]\n",
    "else:\n",
    "    n_active_regimes = CONFIG[\"n_regimes\"]\n",
    "    active_regime_names = regime_names\n",
    "\n",
    "print(f\"\\nSequence regime distribution ({n_active_regimes} active regimes):\")\n",
    "for i in range(n_active_regimes):\n",
    "    count = (seq_regime_labels == i).sum()\n",
    "    print(f\"  {active_regime_names[i]}: {count} sequences ({count / len(seq_regime_labels):.1%})\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60757d6d",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.00976,
     "end_time": "2026-06-13T02:38:45.152667+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.142907+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Regime Classifier on Noised Data\n",
    "\n",
    "The classifier must operate on noised sequences $x_t$ (not clean data),\n",
    "because during sampling we compute gradients w.r.t. partially-denoised\n",
    "samples. We reuse the Encoder architecture from Diffusion-TS with an\n",
    "attention pooling head for classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "2b11499d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.163130Z",
     "iopub.status.busy": "2026-06-13T02:38:45.163017Z",
     "iopub.status.idle": "2026-06-13T02:38:45.165011Z",
     "shell.execute_reply": "2026-06-13T02:38:45.164661Z"
    },
    "papermill": {
     "duration": 0.007775,
     "end_time": "2026-06-13T02:38:45.165408+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.157633+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class GroupNorm32(nn.GroupNorm):\n",
    "    \"\"\"GroupNorm cast to float32 for numerical stability.\"\"\"\n",
    "\n",
    "    def forward(self, input: torch.Tensor) -> torch.Tensor:  # type: ignore[override]\n",
    "        return super().forward(input.float()).type(input.dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "59b17fbf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.176053Z",
     "iopub.status.busy": "2026-06-13T02:38:45.175959Z",
     "iopub.status.idle": "2026-06-13T02:38:45.178418Z",
     "shell.execute_reply": "2026-06-13T02:38:45.178132Z"
    },
    "papermill": {
     "duration": 0.008431,
     "end_time": "2026-06-13T02:38:45.178786+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.170355+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class QKVAttention(nn.Module):\n",
    "    \"\"\"QKV attention for attention pooling.\"\"\"\n",
    "\n",
    "    def __init__(self, n_heads):\n",
    "        super().__init__()\n",
    "        self.n_heads = n_heads\n",
    "\n",
    "    def forward(self, qkv):\n",
    "        bs, width, length = qkv.shape\n",
    "        ch = width // (3 * self.n_heads)\n",
    "        q, k, v = qkv.chunk(3, dim=1)\n",
    "        scale = 1 / math.sqrt(math.sqrt(ch))\n",
    "        weight = torch.einsum(\n",
    "            \"bct,bcs->bts\",\n",
    "            (q * scale).view(bs * self.n_heads, ch, length),\n",
    "            (k * scale).view(bs * self.n_heads, ch, length),\n",
    "        )\n",
    "        weight = torch.softmax(weight.float(), dim=-1).type(weight.dtype)\n",
    "        a = torch.einsum(\"bts,bcs->bct\", weight, v.reshape(bs * self.n_heads, ch, length))\n",
    "        return a.reshape(bs, -1, length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "fc9306cd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.189346Z",
     "iopub.status.busy": "2026-06-13T02:38:45.189272Z",
     "iopub.status.idle": "2026-06-13T02:38:45.191722Z",
     "shell.execute_reply": "2026-06-13T02:38:45.191444Z"
    },
    "papermill": {
     "duration": 0.008227,
     "end_time": "2026-06-13T02:38:45.191955+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.183728+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class AttentionPool(nn.Module):\n",
    "    \"\"\"Attention pooling: sequence → single vector.\"\"\"\n",
    "\n",
    "    def __init__(self, embed_dim, num_heads_channels, output_dim=None):\n",
    "        super().__init__()\n",
    "        self.qkv_proj = nn.Conv1d(embed_dim, 3 * embed_dim, 1)\n",
    "        self.c_proj = nn.Conv1d(embed_dim, output_dim or embed_dim, 1)\n",
    "        self.num_heads = embed_dim // num_heads_channels\n",
    "        self.attention = QKVAttention(self.num_heads)\n",
    "\n",
    "    def forward(self, x):\n",
    "        b, c, *_ = x.shape\n",
    "        x = x.reshape(b, c, -1)\n",
    "        x = torch.cat([x.mean(dim=-1, keepdim=True), x], dim=-1)\n",
    "        x = self.qkv_proj(x)\n",
    "        x = self.attention(x)\n",
    "        x = self.c_proj(x)\n",
    "        return x[:, :, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9d100571",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.202641Z",
     "iopub.status.busy": "2026-06-13T02:38:45.202544Z",
     "iopub.status.idle": "2026-06-13T02:38:45.205042Z",
     "shell.execute_reply": "2026-06-13T02:38:45.204755Z"
    },
    "papermill": {
     "duration": 0.008357,
     "end_time": "2026-06-13T02:38:45.205314+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.196957+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "class RegimeClassifier(nn.Module):\n",
    "    \"\"\"Transformer classifier for noised time series sequences.\"\"\"\n",
    "\n",
    "    def __init__(\n",
    "        self,\n",
    "        feature_size,\n",
    "        seq_length,\n",
    "        num_classes=3,\n",
    "        n_layer_enc=2,\n",
    "        n_embd=64,\n",
    "        n_heads=4,\n",
    "        num_head_channels=8,\n",
    "    ):\n",
    "        super().__init__()\n",
    "        self.emb = Conv_MLP(feature_size, n_embd)\n",
    "        self.encoder = Encoder(n_layer_enc, n_embd, n_heads)\n",
    "        self.pos_enc = LearnablePositionalEncoding(n_embd, max_len=seq_length)\n",
    "\n",
    "        # Encoder output is (B, seq, n_embd); transpose to (B, n_embd, seq) for Conv1d ops\n",
    "        self.out = nn.Sequential(\n",
    "            Transpose((1, 2)),\n",
    "            GroupNorm32(32, n_embd),\n",
    "            nn.SiLU(),\n",
    "            AttentionPool(n_embd, num_head_channels, num_classes),\n",
    "        )\n",
    "\n",
    "    def forward(self, x, t):\n",
    "        emb = self.emb(x)\n",
    "        inp_enc = self.pos_enc(emb)\n",
    "        output = self.encoder(inp_enc, t)\n",
    "        return self.out(output)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75a1c85e",
   "metadata": {
    "papermill": {
     "duration": 0.004919,
     "end_time": "2026-06-13T02:38:45.215370+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.210451+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Train the Classifier\n",
    "\n",
    "Training samples noised sequences at random timesteps and classifies them.\n",
    "The classifier must learn to identify regime characteristics even through\n",
    "heavy noise -- this is what makes the gradient signal useful during sampling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "8768d88f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:45.226120Z",
     "iopub.status.busy": "2026-06-13T02:38:45.226020Z",
     "iopub.status.idle": "2026-06-13T02:38:56.062842Z",
     "shell.execute_reply": "2026-06-13T02:38:56.062502Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 10.842861,
     "end_time": "2026-06-13T02:38:56.063284+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:45.220423+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Training regime classifier for 1500 epochs...\n",
      "  Epoch 1/1500: Loss = 0.6947\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 100/1500: Loss = 0.2522\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 200/1500: Loss = 0.1712\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 300/1500: Loss = 0.0979\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 400/1500: Loss = 0.1303\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 500/1500: Loss = 0.1310\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 600/1500: Loss = 0.0492\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 700/1500: Loss = 0.1118\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 800/1500: Loss = 0.1381\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 900/1500: Loss = 0.1042\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1000/1500: Loss = 0.0971\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1100/1500: Loss = 0.0532\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1200/1500: Loss = 0.0726\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1300/1500: Loss = 0.0579\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1400/1500: Loss = 0.0900\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Epoch 1500/1500: Loss = 0.1137\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifier training complete.\n"
     ]
    }
   ],
   "source": [
    "classifier = RegimeClassifier(\n",
    "    feature_size=n_assets,\n",
    "    seq_length=CONFIG[\"seq_length\"],\n",
    "    num_classes=n_active_regimes,\n",
    "    n_layer_enc=CONFIG[\"n_layer_enc\"],\n",
    "    n_embd=CONFIG[\"d_model\"],\n",
    "    n_heads=CONFIG[\"n_heads\"],\n",
    ").to(device)\n",
    "\n",
    "cls_optimizer = torch.optim.Adam(classifier.parameters(), lr=CONFIG[\"classifier_lr\"])\n",
    "\n",
    "# Prepare labeled dataset\n",
    "labels_tensor = torch.LongTensor(seq_regime_labels).to(device)\n",
    "cls_dataset = TensorDataset(tensor_data, labels_tensor)\n",
    "cls_loader = DataLoader(cls_dataset, batch_size=CONFIG[\"batch_size\"], shuffle=True, drop_last=True)\n",
    "cls_iter = cycle(cls_loader)\n",
    "\n",
    "print(f\"\\nTraining regime classifier for {CONFIG['classifier_epochs']} epochs...\")\n",
    "\n",
    "for step in range(CONFIG[\"classifier_epochs\"]):\n",
    "    classifier.train()\n",
    "    total_loss = 0.0\n",
    "\n",
    "    for _ in range(grad_accum):\n",
    "        batch_x, batch_y = next(cls_iter)\n",
    "\n",
    "        # Forward-noise the batch\n",
    "        t = torch.randint(\n",
    "            0, diffusion_model.num_timesteps, (batch_x.shape[0],), device=device\n",
    "        ).long()\n",
    "        noise = torch.randn_like(batch_x)\n",
    "        x_noised = diffusion_model.q_sample(batch_x, t, noise)\n",
    "\n",
    "        logits = classifier(x_noised, t)\n",
    "        loss = F.cross_entropy(logits, batch_y) / grad_accum\n",
    "        loss.backward()\n",
    "        total_loss += loss.item()\n",
    "\n",
    "    cls_optimizer.step()\n",
    "    cls_optimizer.zero_grad()\n",
    "\n",
    "    if (step + 1) % 100 == 0 or step == 0:\n",
    "        print(\n",
    "            f\"  Epoch {step + 1}/{CONFIG['classifier_epochs']}: Loss = {total_loss:.4f}\", flush=True\n",
    "        )\n",
    "\n",
    "print(\"Classifier training complete.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08d64ef8",
   "metadata": {
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.00509,
     "end_time": "2026-06-13T02:38:56.073829+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:56.068739+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Classifier Guidance Function\n",
    "\n",
    "The `cond_fn` computes $\\nabla_x \\log p(y|x_t)$ -- the gradient of the\n",
    "classifier's log-probability for the target regime with respect to the\n",
    "noised input. This gradient steers the diffusion sampling toward\n",
    "sequences that the classifier recognizes as belonging to the target regime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ac96e98b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:56.085317Z",
     "iopub.status.busy": "2026-06-13T02:38:56.085117Z",
     "iopub.status.idle": "2026-06-13T02:38:56.087666Z",
     "shell.execute_reply": "2026-06-13T02:38:56.087387Z"
    },
    "papermill": {
     "duration": 0.009036,
     "end_time": "2026-06-13T02:38:56.088072+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:56.079036+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "def make_cond_fn(classifier, target_regime, scale=1.0, temperature=1.0):\n",
    "    \"\"\"Create a conditioning function for classifier guidance.\n",
    "\n",
    "    Args:\n",
    "        classifier: Trained regime classifier\n",
    "        target_regime: Target regime index\n",
    "        scale: Gradient scale - higher = stronger steering toward target regime\n",
    "        temperature: Softmax temperature - higher = softer classifier, more diverse samples\n",
    "    \"\"\"\n",
    "\n",
    "    def cond_fn(x, t, **kwargs):\n",
    "        with torch.enable_grad():\n",
    "            x_in = x.detach().requires_grad_(True)\n",
    "            logits = classifier(x_in, t)\n",
    "            # Temperature scaling: higher temp = softer distribution, less confident\n",
    "            log_probs = F.log_softmax(logits / temperature, dim=-1)\n",
    "            target = torch.full((x.shape[0],), target_regime, device=x.device, dtype=torch.long)\n",
    "            selected = log_probs[range(len(logits)), target]\n",
    "            return torch.autograd.grad(selected.sum(), x_in)[0] * scale\n",
    "\n",
    "    return cond_fn"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "797790ac",
   "metadata": {
    "papermill": {
     "duration": 0.005276,
     "end_time": "2026-06-13T02:38:56.098673+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:56.093397+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Generate Regime-Conditional Samples\n",
    "\n",
    "For each regime, we generate synthetic sequences steered by the classifier.\n",
    "Then we compare the volatility characteristics of generated samples against\n",
    "historical data from that regime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "11008390",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:56.110011Z",
     "iopub.status.busy": "2026-06-13T02:38:56.109904Z",
     "iopub.status.idle": "2026-06-13T02:38:57.843717Z",
     "shell.execute_reply": "2026-06-13T02:38:57.842927Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 1.740143,
     "end_time": "2026-06-13T02:38:57.844167+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:56.104024+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Generating 100 samples for regime: Low-Vol...\n",
      "  Guidance: scale=0.75, temp=1.0, eta=0.5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Low-Vol: generated vol = 0.0104\n",
      "\n",
      "Generating 100 samples for regime: High-Vol...\n",
      "  Guidance: scale=0.3, temp=2.0, eta=0.7\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  High-Vol: generated vol = 0.0219\n"
     ]
    }
   ],
   "source": [
    "# N_COND is set in the parameters cell above\n",
    "\n",
    "regime_samples = {}\n",
    "for regime_id in range(n_active_regimes):\n",
    "    # Regime-specific guidance settings\n",
    "    settings = CONFIG[\"guidance_settings\"].get(\n",
    "        regime_id, {\"scale\": 0.5, \"temperature\": 1.0, \"eta\": 0.5}\n",
    "    )\n",
    "    print(f\"\\nGenerating {N_COND} samples for regime: {active_regime_names[regime_id]}...\")\n",
    "    print(\n",
    "        f\"  Guidance: scale={settings['scale']}, temp={settings['temperature']}, eta={settings['eta']}\"\n",
    "    )\n",
    "\n",
    "    cond_fn = make_cond_fn(\n",
    "        classifier, regime_id, scale=settings[\"scale\"], temperature=settings[\"temperature\"]\n",
    "    )\n",
    "\n",
    "    with torch.no_grad():\n",
    "        samples = (\n",
    "            ema.ema_model.generate_mts(\n",
    "                batch_size=N_COND,\n",
    "                cond_fn=cond_fn,\n",
    "                eta=settings[\"eta\"],\n",
    "            )\n",
    "            .detach()\n",
    "            .cpu()\n",
    "            .numpy()\n",
    "        )\n",
    "\n",
    "    # Note: Do NOT apply VARIANCE_SCALE_FACTOR here. Classifier guidance already\n",
    "    # adjusts variance (and tends to inflate it for minority regimes). The variance\n",
    "    # scaling is only for unconditional samples where the raw model output is low.\n",
    "\n",
    "    # Denormalize\n",
    "    samples_flat = samples.reshape(-1, n_assets)\n",
    "    samples_denorm = scaler.inverse_transform(samples_flat).reshape(samples.shape)\n",
    "    regime_samples[regime_id] = samples_denorm.astype(np.float32)\n",
    "\n",
    "    vol = np.std(samples_denorm[:, :, 0])\n",
    "    print(f\"  {active_regime_names[regime_id]}: generated vol = {vol:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab2eafe3",
   "metadata": {
    "papermill": {
     "duration": 0.005434,
     "end_time": "2026-06-13T02:38:57.856581+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:57.851147+00:00",
     "status": "completed"
    }
   },
   "source": [
    "### Compare Regime-Conditional Statistics\n",
    "\n",
    "For each regime, we compare the volatility and mean return of the generated\n",
    "samples against the historical sequences labeled with that regime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "7ca91525",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:57.867942Z",
     "iopub.status.busy": "2026-06-13T02:38:57.867847Z",
     "iopub.status.idle": "2026-06-13T02:38:57.875029Z",
     "shell.execute_reply": "2026-06-13T02:38:57.874415Z"
    },
    "papermill": {
     "duration": 0.013597,
     "end_time": "2026-06-13T02:38:57.875404+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:57.861807+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Regime-Conditional Evaluation ===\n",
      "Reference: Real vol=0.0126, Unconditional synth vol=0.0118\n",
      "Guidance settings: regime-specific (see above)\n",
      "\n",
      "Regime         Hist Vol    Gen Vol  Vol Ratio  Hist Mean   Gen Mean\n",
      "------------------------------------------------------------------\n",
      "Low-Vol          0.0111     0.0104       0.94x   0.000626   0.001613\n",
      "High-Vol         0.0203     0.0219       1.08x  -0.001667  -0.002721\n"
     ]
    }
   ],
   "source": [
    "# Reference: unconditional synthetic volatility\n",
    "uncond_vol = np.std(synthetic_sequences[:, :, 0])\n",
    "real_vol = np.std(sequences[:, :, 0])\n",
    "\n",
    "print(\"\\n=== Regime-Conditional Evaluation ===\")\n",
    "print(f\"Reference: Real vol={real_vol:.4f}, Unconditional synth vol={uncond_vol:.4f}\")\n",
    "print(\"Guidance settings: regime-specific (see above)\")\n",
    "print()\n",
    "print(\n",
    "    f\"{'Regime':<12} {'Hist Vol':>10} {'Gen Vol':>10} {'Vol Ratio':>10} {'Hist Mean':>10} {'Gen Mean':>10}\"\n",
    ")\n",
    "print(\"-\" * 66)\n",
    "\n",
    "for regime_id in range(n_active_regimes):\n",
    "    mask = seq_regime_labels == regime_id\n",
    "    hist_seqs = sequences[mask]\n",
    "    gen_seqs = regime_samples[regime_id]\n",
    "\n",
    "    hist_vol = np.std(hist_seqs[:, :, 0])\n",
    "    gen_vol = np.std(gen_seqs[:, :, 0])\n",
    "    vol_ratio = gen_vol / hist_vol\n",
    "    hist_mean = np.mean(hist_seqs[:, :, 0])\n",
    "    gen_mean = np.mean(gen_seqs[:, :, 0])\n",
    "\n",
    "    print(\n",
    "        f\"{active_regime_names[regime_id]:<12} {hist_vol:>10.4f} {gen_vol:>10.4f} {vol_ratio:>10.2f}x \"\n",
    "        f\"{hist_mean:>10.6f} {gen_mean:>10.6f}\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "3444b935",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:57.887515Z",
     "iopub.status.busy": "2026-06-13T02:38:57.887394Z",
     "iopub.status.idle": "2026-06-13T02:38:57.893568Z",
     "shell.execute_reply": "2026-06-13T02:38:57.893093Z"
    },
    "papermill": {
     "duration": 0.012919,
     "end_time": "2026-06-13T02:38:57.893935+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:57.881016+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Per-Sequence Volatility Distribution ===\n",
      "Regime       Source          Min      25%      50%      75%      Max   Kurtosis\n",
      "--------------------------------------------------------------------------------\n",
      "Low-Vol      Historical   0.0042   0.0071   0.0087   0.0111   0.0368      14.22\n",
      "Low-Vol      Generated    0.0040   0.0057   0.0068   0.0095   0.0305       5.06\n",
      "High-Vol     Historical   0.0077   0.0129   0.0148   0.0168   0.0363      -0.19\n",
      "High-Vol     Generated    0.0062   0.0142   0.0185   0.0266   0.0386      -0.82\n"
     ]
    }
   ],
   "source": [
    "# Per-sequence volatility distribution diagnostics\n",
    "print(\"\\n=== Per-Sequence Volatility Distribution ===\")\n",
    "print(\n",
    "    f\"{'Regime':<12} {'Source':<10} {'Min':>8} {'25%':>8} {'50%':>8} {'75%':>8} {'Max':>8} {'Kurtosis':>10}\"\n",
    ")\n",
    "print(\"-\" * 80)\n",
    "\n",
    "for regime_id in range(n_active_regimes):\n",
    "    mask = seq_regime_labels == regime_id\n",
    "    hist_vols = np.std(sequences[mask][:, :, 0], axis=1)\n",
    "    gen_vols = np.std(regime_samples[regime_id][:, :, 0], axis=1)\n",
    "\n",
    "    for name, vols in [(\"Historical\", hist_vols), (\"Generated\", gen_vols)]:\n",
    "        pcts = np.percentile(vols, [0, 25, 50, 75, 100])\n",
    "        kurt = calc_kurtosis(vols)\n",
    "        print(\n",
    "            f\"{active_regime_names[regime_id]:<12} {name:<10} {pcts[0]:>8.4f} {pcts[1]:>8.4f} \"\n",
    "            f\"{pcts[2]:>8.4f} {pcts[3]:>8.4f} {pcts[4]:>8.4f} {kurt:>10.2f}\"\n",
    "        )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c79d8052",
   "metadata": {
    "papermill": {
     "duration": 0.005288,
     "end_time": "2026-06-13T02:38:57.904702+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:57.899414+00:00",
     "status": "completed"
    }
   },
   "source": [
    "**Interpretation**: Classifier guidance produces regime separation. Key observations:\n",
    "\n",
    "1. **Variance scaling**: The model generates ~66% of real variance in normalized space.\n",
    "   This is a known issue with diffusion models using trend+seasonal decomposition --\n",
    "   the smooth components underestimate high-frequency variation. We apply post-hoc\n",
    "   variance scaling to match training distribution.\n",
    "\n",
    "2. **Low-Vol (majority, 88%)**: Unconditional samples are already Low-Vol-like, so\n",
    "   gentle guidance suffices. With variance scaling, generated volatility matches\n",
    "   historical (~1.0x ratio).\n",
    "\n",
    "3. **High-Vol (minority, 12%)**: Classifier gradients must push harder to shift\n",
    "   samples toward this regime, which can add variance. Some inflation (~1.3-1.5x)\n",
    "   is expected for minority regimes.\n",
    "\n",
    "**Scale tuning**: Higher scale = more regime separation but more variance. Scale 0.75\n",
    "balances accuracy for the majority class with reasonable minority-class separation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "5157d5d9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:57.916351Z",
     "iopub.status.busy": "2026-06-13T02:38:57.916226Z",
     "iopub.status.idle": "2026-06-13T02:38:58.102640Z",
     "shell.execute_reply": "2026-06-13T02:38:58.102070Z"
    },
    "papermill": {
     "duration": 0.193011,
     "end_time": "2026-06-13T02:38:58.102977+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:57.909966+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Regime comparison: Historical vs Generated side-by-side\n",
    "# Shared y-axis across ALL panels for fair volatility comparison\n",
    "\n",
    "sns.set_style(\"whitegrid\")\n",
    "\n",
    "fig, axes = plt.subplots(\n",
    "    n_active_regimes, 2, figsize=(10, 3 * n_active_regimes), sharex=True, sharey=True\n",
    ")\n",
    "if n_active_regimes == 1:\n",
    "    axes = axes.reshape(1, 2)\n",
    "\n",
    "regime_colors = {\"hist\": COLORS[\"blue\"], \"gen\": COLORS[\"copper\"]}\n",
    "n_paths = 10\n",
    "\n",
    "for regime_id in range(n_active_regimes):\n",
    "    mask = seq_regime_labels == regime_id\n",
    "    hist_seqs = sequences[mask]\n",
    "    gen_seqs = regime_samples[regime_id]\n",
    "\n",
    "    # Left: Historical paths\n",
    "    ax_hist = axes[regime_id, 0]\n",
    "    for j in range(min(n_paths, len(hist_seqs))):\n",
    "        ax_hist.plot(hist_seqs[j, :, 0], color=regime_colors[\"hist\"], alpha=0.5, linewidth=0.8)\n",
    "    ax_hist.set_ylabel(f\"{active_regime_names[regime_id]}\\nDaily Return\")\n",
    "    ax_hist.axhline(0, color=COLORS[\"neutral\"], linestyle=\"--\", linewidth=0.5, alpha=0.5)\n",
    "    sns.despine(ax=ax_hist)\n",
    "\n",
    "    # Right: Generated paths\n",
    "    ax_gen = axes[regime_id, 1]\n",
    "    for j in range(min(n_paths, len(gen_seqs))):\n",
    "        ax_gen.plot(gen_seqs[j, :, 0], color=regime_colors[\"gen\"], alpha=0.5, linewidth=0.8)\n",
    "    ax_gen.axhline(0, color=COLORS[\"neutral\"], linestyle=\"--\", linewidth=0.5, alpha=0.5)\n",
    "    sns.despine(ax=ax_gen)\n",
    "\n",
    "# Column titles\n",
    "axes[0, 0].set_title(\n",
    "    f\"Historical (vol: Low={np.std(sequences[seq_regime_labels == 0][:, :, 0]):.4f}, High={np.std(sequences[seq_regime_labels == 1][:, :, 0]):.4f})\"\n",
    ")\n",
    "axes[0, 1].set_title(\n",
    "    f\"Generated (vol: Low={np.std(regime_samples[0][:, :, 0]):.4f}, High={np.std(regime_samples[1][:, :, 0]):.4f})\"\n",
    ")\n",
    "axes[-1, 0].set_xlabel(\"Time Step\")\n",
    "axes[-1, 1].set_xlabel(\"Time Step\")\n",
    "fig.suptitle(\"Conditional Generation Separates Volatility Regimes\", fontsize=12, y=1.02)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "748881a4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:58.117197Z",
     "iopub.status.busy": "2026-06-13T02:38:58.117036Z",
     "iopub.status.idle": "2026-06-13T02:38:58.122706Z",
     "shell.execute_reply": "2026-06-13T02:38:58.122288Z"
    },
    "papermill": {
     "duration": 0.013396,
     "end_time": "2026-06-13T02:38:58.123145+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:58.109749+00:00",
     "status": "completed"
    }
   },
   "outputs": [],
   "source": [
    "# Collect all volatilities to set shared bins\n",
    "all_hist_vols = []\n",
    "all_gen_vols = []\n",
    "for regime_id in range(n_active_regimes):\n",
    "    mask = seq_regime_labels == regime_id\n",
    "    all_hist_vols.extend(np.std(sequences[mask][:, :, 0], axis=1))\n",
    "    all_gen_vols.extend(np.std(regime_samples[regime_id][:, :, 0], axis=1))\n",
    "\n",
    "all_vols = np.concatenate([all_hist_vols, all_gen_vols])\n",
    "bins = np.linspace(all_vols.min(), all_vols.max(), 31)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "c6c96ec7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:58.136569Z",
     "iopub.status.busy": "2026-06-13T02:38:58.136455Z",
     "iopub.status.idle": "2026-06-13T02:38:58.322503Z",
     "shell.execute_reply": "2026-06-13T02:38:58.322032Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.193458,
     "end_time": "2026-06-13T02:38:58.322944+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:58.129486+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Volatility distribution comparison - shared x-axis across regimes\n",
    "fig, axes = plt.subplots(\n",
    "    1, n_active_regimes, figsize=(5 * n_active_regimes, 4), sharex=True, sharey=True\n",
    ")\n",
    "if n_active_regimes == 1:\n",
    "    axes = [axes]\n",
    "\n",
    "for regime_id, ax in enumerate(axes):\n",
    "    mask = seq_regime_labels == regime_id\n",
    "    hist_seqs = sequences[mask]\n",
    "    gen_seqs = regime_samples[regime_id]\n",
    "\n",
    "    hist_vols = np.std(hist_seqs[:, :, 0], axis=1)\n",
    "    gen_vols = np.std(gen_seqs[:, :, 0], axis=1)\n",
    "\n",
    "    ax.hist(\n",
    "        hist_vols,\n",
    "        bins=bins,\n",
    "        alpha=0.6,\n",
    "        label=f\"Historical (μ={hist_vols.mean():.4f})\",\n",
    "        color=COLORS[\"blue\"],\n",
    "        density=True,\n",
    "    )\n",
    "    ax.hist(\n",
    "        gen_vols,\n",
    "        bins=bins,\n",
    "        alpha=0.6,\n",
    "        label=f\"Generated (μ={gen_vols.mean():.4f})\",\n",
    "        color=COLORS[\"copper\"],\n",
    "        density=True,\n",
    "    )\n",
    "    ax.set_xlabel(\"Sequence Volatility (std)\")\n",
    "    ax.set_title(f\"{active_regime_names[regime_id]}\")\n",
    "    ax.legend(fontsize=9)\n",
    "    sns.despine(ax=ax)\n",
    "\n",
    "axes[0].set_ylabel(\"Density\")\n",
    "fig.suptitle(\n",
    "    \"Generated Volatilities Track the Historical Distribution per Regime\", fontsize=12, y=1.02\n",
    ")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f070c115",
   "metadata": {
    "papermill": {
     "duration": 0.006461,
     "end_time": "2026-06-13T02:38:58.335865+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:58.329404+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## 11. Save Samples (Model checkpoint saved after training)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "61c0a908",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-13T02:38:58.349309Z",
     "iopub.status.busy": "2026-06-13T02:38:58.349189Z",
     "iopub.status.idle": "2026-06-13T02:38:58.359360Z",
     "shell.execute_reply": "2026-06-13T02:38:58.358941Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 0.017684,
     "end_time": "2026-06-13T02:38:58.359838+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:58.342154+00:00",
     "status": "completed"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples to: 05_synthetic_data/output/diffusion_ts/samples/\n"
     ]
    }
   ],
   "source": [
    "# Save samples and metadata (model checkpoint already saved in training section)\n",
    "samples_dir = OUTPUT_DIR / \"samples\"\n",
    "samples_dir.mkdir(parents=True, exist_ok=True)\n",
    "\n",
    "np.save(samples_dir / \"synthetic_sequences.npy\", synthetic_sequences)\n",
    "np.save(samples_dir / \"holdout_returns.npy\", holdout_returns)\n",
    "np.save(samples_dir / \"train_returns.npy\", returns)\n",
    "np.save(samples_dir / \"historical_sequences.npy\", sequences)\n",
    "np.save(samples_dir / \"seq_regime_labels.npy\", seq_regime_labels)\n",
    "\n",
    "# Save regime-conditional samples\n",
    "for regime_id, samples in regime_samples.items():\n",
    "    np.save(samples_dir / f\"regime_{regime_id}_samples.npy\", samples)\n",
    "np.save(samples_dir / \"active_regime_names.npy\", np.array(active_regime_names))\n",
    "\n",
    "# Save evaluation metadata\n",
    "metadata = {\n",
    "    \"created_at\": datetime.now(UTC).isoformat(),\n",
    "    \"config\": CONFIG,\n",
    "    \"n_samples\": N_SYNTHETIC,\n",
    "    \"evaluation\": {\n",
    "        \"mean_ks_statistic\": stats_results[\"mean_ks_statistic\"],\n",
    "        \"correlation_error\": stats_results[\"correlation_error\"],\n",
    "        \"autocorrelation_error\": stats_results[\"autocorrelation_error\"],\n",
    "        \"tstr_ratio\": tstr_results[\"tstr_ratio\"],\n",
    "    },\n",
    "}\n",
    "with open(samples_dir / \"metadata.json\", \"w\") as f:\n",
    "    json.dump(metadata, f, indent=2, default=str)\n",
    "\n",
    "print(f\"Saved samples to: {samples_dir}/\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76f9a806",
   "metadata": {
    "papermill": {
     "duration": 0.006639,
     "end_time": "2026-06-13T02:38:58.376726+00:00",
     "exception": false,
     "start_time": "2026-06-13T02:38:58.370087+00:00",
     "status": "completed"
    }
   },
   "source": [
    "## Key Takeaways\n",
    "\n",
    "1. **Interpretable decomposition**: Diffusion-TS separates denoising predictions\n",
    "   into trend (polynomial), seasonal (Fourier), and residual components, making\n",
    "   the generator's behavior inspectable.\n",
    "\n",
    "2. **Fourier loss preserves spectral structure**: The frequency-domain regularizer\n",
    "   encourages the model to match autocorrelation and periodicity patterns, which\n",
    "   is critical for realistic financial time series.\n",
    "\n",
    "3. **Classifier guidance enables conditional generation**: By training a separate\n",
    "   classifier on noised data, we can steer sampling toward specific regimes\n",
    "   (low-volatility, normal, crisis) -- enabling regime-specific stress testing\n",
    "   with synthetic data.\n",
    "\n",
    "4. **DDIM fast sampling**: Reducing reverse steps from 500 to 50 provides ~10x\n",
    "   speedup with minimal quality loss, making generation practical.\n",
    "\n",
    "**Limitations**:\n",
    "- Classifier guidance quality depends on regime label accuracy (HMM is simple)\n",
    "- The guidance scale requires tuning per application\n",
    "- High-dimensional generation (50 assets) is computationally intensive\n",
    "\n",
    "**Next**: Section 5.5 discusses the Fidelity-Utility-Privacy framework for\n",
    "systematic evaluation of any synthetic generator."
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "formats": "ipynb,py:percent",
   "cell_metadata_filter": "tags,-all"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.14.5"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 25.950464,
   "end_time": "2026-06-13T02:38:59.700192+00:00",
   "environment_variables": {},
   "exception": null,
   "input_path": "05_synthetic_data/05_diffusion_ts.ipynb",
   "output_path": "05_synthetic_data/05_diffusion_ts.ipynb",
   "parameters": {},
   "start_time": "2026-06-13T02:38:33.749728+00:00",
   "version": "2.7.0"
  },
  "ml4t_provenance": {
   "source_py_blob": "c8887310a029ae0c5b4192f1656c623d9cf0145b",
   "executed_at": "2026-06-13T02:39:00.067461+00:00",
   "executor": "gpu-docker",
   "production": true,
   "parameters": {},
   "notes": "executed-state finalization batch (2026-06-13 run)"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}