TimeDIffussion_viz.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "322ef4d8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-11-21T16:07:52.183107Z",
     "iopub.status.busy": "2022-11-21T16:07:52.182112Z",
     "iopub.status.idle": "2022-11-21T16:07:57.243245Z",
     "shell.execute_reply": "2022-11-21T16:07:57.241811Z"
    },
    "papermill": {
     "duration": 5.067812,
     "end_time": "2022-11-21T16:07:57.246221",
     "exception": false,
     "start_time": "2022-11-21T16:07:52.178409",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from tqdm import tqdm\n",
    "from pathlib import Path\n",
    "\n",
    "from utils.data import get_dataset_iterator\n",
    "\n",
    "import torch\n",
    "from torch import nn\n",
    "\n",
    "from utils.dl import QuantGAN_TemporalBlock\n",
    "\n",
    "data_path = Path(\"data\")\n",
    "models_dir = Path(\"models\")\n",
    "\n",
    "hsm_dataset_path = data_path / \"huge_stock_market_dataset\"\n",
    "solar_energy_dataset_path = data_path / \"solar_energy\"\n",
    "fuel_prices_dataset_path = data_path / \"fuel_prices\"\n",
    "passengers_dataset_path = data_path / \"air_passengers\"\n",
    "\n",
    "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8e4e4afe",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-11-21T16:07:57.290905Z",
     "iopub.status.busy": "2022-11-21T16:07:57.290379Z",
     "iopub.status.idle": "2022-11-21T18:49:10.029398Z",
     "shell.execute_reply": "2022-11-21T18:49:10.028343Z"
    },
    "papermill": {
     "duration": 9672.746812,
     "end_time": "2022-11-21T18:49:10.032404",
     "exception": false,
     "start_time": "2022-11-21T16:07:57.285592",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# 3rd version (5, 256, 32) + high freq preprocessing + new gen\n",
    "def is_high_freq(time_series, threshold=0.85, rolling_parts=200):\n",
    "    orig_std = time_series.std().values[0]\n",
    "    ma_ts = time_series.rolling(len(time_series) // rolling_parts).mean()\n",
    "    ma_std = ma_ts.std().values[0]\n",
    "    return abs(ma_std - orig_std) / orig_std > threshold\n",
    "\n",
    "def ma(time_series, rolling_parts=200, window=None):\n",
    "    if window is None:\n",
    "        window = max(len(time_series) // rolling_parts, 2)\n",
    "    ts1 = time_series.rolling(window, closed=\"left\").mean()\n",
    "    ts2 = time_series[:: - 1].rolling(window).mean()[:: - 1]\n",
    "    ts1[ts1.isna()] = ts2[ts1.isna()]\n",
    "    ts2[ts2.isna()] = ts1[ts2.isna()]\n",
    "    ats = (ts1 + ts2) / 2\n",
    "    return ats\n",
    "\n",
    "\n",
    "class TimeDIffusion(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.tcn = nn.ModuleList([QuantGAN_TemporalBlock(1, 128, kernel_size=1, stride=1, dilation=1, padding=0, dropout=0.25),\n",
    "                                 *[QuantGAN_TemporalBlock(128, 128, kernel_size=2, stride=1, dilation=i, padding=i, dropout=0.0)\n",
    "                                        for i in [2 ** i for i in range(14)]]])\n",
    "        self.last = nn.Conv1d(128, 1, kernel_size=1, stride=1, dilation=1)\n",
    "\n",
    "    def forward(self, x):\n",
    "        skip_layers = []\n",
    "        for layer in self.tcn:\n",
    "            skip, x = layer(x)\n",
    "            skip_layers.append(skip)\n",
    "        x = self.last(x + sum(skip_layers))\n",
    "        return x\n",
    "\n",
    "\n",
    "epochs = 20\n",
    "steps_per_epoch = 32\n",
    "samples_to_gen = 20\n",
    "steps_to_gen = set(range(3, 3 + samples_to_gen))\n",
    "max_gen_step = max(steps_to_gen) + 1\n",
    "batch_size = 32"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ab172bc",
   "metadata": {},
   "source": [
    "5 min 1k ts"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "771b0586",
   "metadata": {
    "papermill": {
     "duration": 0.003351,
     "end_time": "2022-11-21T18:49:10.039617",
     "exception": false,
     "start_time": "2022-11-21T18:49:10.036266",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "noise thresholds (steps): 1.4 (25/256), 0.9 (41/256), 0.68 (15/256), 0.04 (98/256), 0.55 (65/256)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "57ca8fc3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = TimeDIffusion()\n",
    "model.load_state_dict(torch.load(\"models/hsm/model1.pt\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0ebb315f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TimeDIffusion(\n",
       "  (tcn): ModuleList(\n",
       "    (0): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(1, 128, kernel_size=(1,), stride=(1,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.25, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(1,), stride=(1,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.25, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(1, 128, kernel_size=(1,), stride=(1,))\n",
       "        (1): ReLU()\n",
       "        (2): Dropout(p=0.25, inplace=False)\n",
       "        (3): Conv1d(128, 128, kernel_size=(1,), stride=(1,))\n",
       "        (4): ReLU()\n",
       "        (5): Dropout(p=0.25, inplace=False)\n",
       "      )\n",
       "      (downsample): Conv1d(1, 128, kernel_size=(1,), stride=(1,))\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (1): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (2): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2,), dilation=(2,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2,), dilation=(2,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2,), dilation=(2,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2,), dilation=(2,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (3): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4,), dilation=(4,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4,), dilation=(4,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4,), dilation=(4,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4,), dilation=(4,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (4): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8,), dilation=(8,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8,), dilation=(8,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8,), dilation=(8,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8,), dilation=(8,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (5): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(16,), dilation=(16,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(16,), dilation=(16,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(16,), dilation=(16,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(16,), dilation=(16,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (6): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(32,), dilation=(32,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(32,), dilation=(32,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(32,), dilation=(32,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(32,), dilation=(32,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (7): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(64,), dilation=(64,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(64,), dilation=(64,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(64,), dilation=(64,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(64,), dilation=(64,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (8): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(128,), dilation=(128,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(128,), dilation=(128,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(128,), dilation=(128,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(128,), dilation=(128,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (9): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(256,), dilation=(256,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(256,), dilation=(256,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(256,), dilation=(256,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(256,), dilation=(256,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (10): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(512,), dilation=(512,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(512,), dilation=(512,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(512,), dilation=(512,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(512,), dilation=(512,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (11): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1024,), dilation=(1024,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1024,), dilation=(1024,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1024,), dilation=(1024,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(1024,), dilation=(1024,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (12): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2048,), dilation=(2048,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2048,), dilation=(2048,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2048,), dilation=(2048,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(2048,), dilation=(2048,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (13): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4096,), dilation=(4096,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4096,), dilation=(4096,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4096,), dilation=(4096,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(4096,), dilation=(4096,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "    (14): QuantGAN_TemporalBlock(\n",
       "      (conv1): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8192,), dilation=(8192,))\n",
       "      (chomp1): Chomp1d()\n",
       "      (relu1): ReLU()\n",
       "      (dropout1): Dropout(p=0.0, inplace=False)\n",
       "      (conv2): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8192,), dilation=(8192,))\n",
       "      (chomp2): Chomp1d()\n",
       "      (relu2): ReLU()\n",
       "      (dropout2): Dropout(p=0.0, inplace=False)\n",
       "      (net): Sequential(\n",
       "        (0): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8192,), dilation=(8192,))\n",
       "        (1): Chomp1d()\n",
       "        (2): ReLU()\n",
       "        (3): Dropout(p=0.0, inplace=False)\n",
       "        (4): Conv1d(128, 128, kernel_size=(2,), stride=(1,), padding=(8192,), dilation=(8192,))\n",
       "        (5): Chomp1d()\n",
       "        (6): ReLU()\n",
       "        (7): Dropout(p=0.0, inplace=False)\n",
       "      )\n",
       "      (relu): ReLU()\n",
       "    )\n",
       "  )\n",
       "  (last): Conv1d(128, 1, kernel_size=(1,), stride=(1,))\n",
       ")"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "917ecf04",
   "metadata": {},
   "outputs": [],
   "source": [
    "iterator = get_dataset_iterator(\"hsm\", hsm_dataset_path)\n",
    "next(iterator)\n",
    "time_series = next(iterator)\n",
    "train = time_series.values.flatten()\n",
    "tmean = train.mean()\n",
    "tstd = train.std()\n",
    "train = (train - tmean) / tstd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "23c59351",
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils.synth_eval import kl_div"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "06484c91",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-11-21T18:49:10.048086Z",
     "iopub.status.busy": "2022-11-21T18:49:10.047771Z",
     "iopub.status.idle": "2022-11-21T18:49:10.052898Z",
     "shell.execute_reply": "2022-11-21T18:49:10.051868Z"
    },
    "papermill": {
     "duration": 0.011853,
     "end_time": "2022-11-21T18:49:10.055022",
     "exception": false,
     "start_time": "2022-11-21T18:49:10.043169",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "inf 132.115966796875\n"
     ]
    }
   ],
   "source": [
    "steps = steps_per_epoch // 2\n",
    "plot_rate = steps // 5\n",
    "with torch.no_grad():\n",
    "    model.eval()\n",
    "    noise = torch.rand(1, 1, len(train)).to(device)\n",
    "    generated = noise\n",
    "    plt.plot(train)\n",
    "    plt.plot(generated.detach().cpu().squeeze().numpy())\n",
    "    plt.legend([\"ground truth\", \"noise\"])\n",
    "    plt.title(f\"Step {0}\")\n",
    "    plt.show()\n",
    "    kl_divs = []\n",
    "    pred_noises = []\n",
    "    for step in range(1, steps + 1):\n",
    "        pred_noise = model(generated)\n",
    "        generated -= pred_noise\n",
    "        \n",
    "        pred_noises.append(pred_noise.sum().item())\n",
    "        kl_divs.append(kl_div(generated.detach().cpu().squeeze().numpy(), train).mean())\n",
    "        \n",
    "        if step  in range(5) or step == 100:\n",
    "\n",
    "            result = generated.detach().cpu().squeeze().numpy()\n",
    "            plt.plot(train)\n",
    "            plt.plot(result)\n",
    "            plt.legend([\"ground truth\", \"synthetic\"])\n",
    "            # plt.title(f\"Step #{step} pred_noise: {pred_noises[- 1]:0.4f} kl_div: {kl_divs[- 1]: 0.4f}\")\n",
    "            plt.title(f\"Step {step}\")\n",
    "            plt.show()\n",
    "    \n",
    "    plt.plot(range(len(kl_divs)), kl_divs)\n",
    "    ind = np.argmin(kl_divs)\n",
    "    plt.title(\"kl_div \" + str(ind))\n",
    "    plt.show()\n",
    "    plt.plot(range(len(pred_noises)), pred_noises)\n",
    "    plt.title(\"pred noises\")\n",
    "    plt.show()\n",
    "    print(kl_divs[ind], pred_noises[ind])\n",
    "# plt.plot(noise.cpu().numpy().squeeze())\n",
    "# plt.plot(pred_noise.cpu().numpy().squeeze())\n",
    "# plt.legend([\"noise\", \"pred_noise\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "469ed094",
   "metadata": {
    "papermill": {
     "duration": 0.00338,
     "end_time": "2022-11-21T18:49:10.061929",
     "exception": false,
     "start_time": "2022-11-21T18:49:10.058549",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dde4d527",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = None\n",
    "for time_series in dataset:\n",
    "    ts_index += 1\n",
    "    \n",
    "    # high freq check\n",
    "    if is_high_freq(time_series):\n",
    "        time_series = ma(time_series)\n",
    "    \n",
    "    train = time_series.values.flatten()\n",
    "    tmean = train.mean()\n",
    "    tstd = train.std()\n",
    "    train = (train - tmean) / tstd\n",
    "    train_tensor = torch.from_numpy(train).float().to(device)\n",
    "\n",
    "    torch.random.manual_seed(0)\n",
    "    model = TimeDIffusion().to(device)\n",
    "    optim = torch.optim.AdamW(model.parameters(), lr=4e-4)\n",
    "    generate_noise = lambda x: x\n",
    "    num_epochs = epochs\n",
    "\n",
    "    for epoch in range(num_epochs):\n",
    "        X = train_tensor.repeat(batch_size, 1).unsqueeze(1)\n",
    "        noise = generate_noise(X.shape).to(device)\n",
    "        noise_level = torch.rand(X.shape).to(device)\n",
    "        noise *= noise_level\n",
    "\n",
    "        for step in range(steps_per_epoch):\n",
    "            optim.zero_grad()\n",
    "\n",
    "            predicted_noise = model(X + noise)\n",
    "            loss = (noise - predicted_noise).abs().mean()\n",
    "            loss.backward()\n",
    "            optim.step()\n",
    "            with torch.no_grad():\n",
    "                noise -= predicted_noise\n",
    "\n",
    "    model.eval()\n",
    "    result = []\n",
    "    steps_to_generate = steps_per_epoch // 2\n",
    "\n",
    "    with torch.no_grad():\n",
    "\n",
    "        new_samples = generate_noise(X.shape)\n",
    "        for step in range(steps_per_epoch):\n",
    "            predicted_noise = model(new_samples)\n",
    "            new_samples -= predicted_noise\n",
    "            if step in steps_to_gen:\n",
    "                result.append(new_samples)\n",
    "\n",
    "    result = np.row_stack(result)\n",
    "    \n",
    "#         break\n",
    "#     break"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.9 64-bit",
   "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": "3.8.9"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 9690.933553,
   "end_time": "2022-11-21T18:49:12.391501",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2022-11-21T16:07:41.457948",
   "version": "2.3.4"
  },
  "vscode": {
   "interpreter": {
    "hash": "1289e797c8b2364a1b561fc46768e8fcf8446b2e18e77ab0795c8743ff6ac10a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}