{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e3455885",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using backend: tensorflow\n",
      "Other supported backends: tensorflow.compat.v1, pytorch, jax, paddle.\n",
      "paddle supports more examples now and is recommended.\n"
     ]
    }
   ],
   "source": [
    "import deepxde as dde\n",
    "from deepxde.backend import tf\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fdb7abe3",
   "metadata": {},
   "outputs": [],
   "source": [
    "k = 0.4\n",
    "L = 1\n",
    "n = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "405e8d90",
   "metadata": {},
   "outputs": [],
   "source": [
    "geom = dde.geometry.Interval(0,L)\n",
    "timedomain = dde.geometry.TimeDomain(0,n)\n",
    "geomtime = dde.geometry.GeometryXTime(geom, timedomain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "655cc132",
   "metadata": {},
   "outputs": [],
   "source": [
    "ic = dde.icbc.IC(geomtime, lambda x: np.sin(n* np.pi * x[:,0:1]/L), lambda _, on_initial: on_initial)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "eae33ad3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 2]\n",
      " [ 8]\n",
      " [14]]\n"
     ]
    }
   ],
   "source": [
    "input_array = np.array([[1,2,3],\n",
    "                        [4,5,6],\n",
    "                        [7,8,9]])\n",
    "\n",
    "lambda_function = lambda x : 2*x[:,0:1]\n",
    "\n",
    "result_array = lambda_function(input_array)\n",
    "\n",
    "print(result_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "46afa2b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Even\n",
      "Odd\n"
     ]
    }
   ],
   "source": [
    "cond_lambda = lambda x : \"Even\" if x%2 == 0 else \"Odd\"\n",
    "print(cond_lambda(4))\n",
    "print(cond_lambda(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ab47feda",
   "metadata": {},
   "outputs": [],
   "source": [
    "def double_first_column(input_array):\n",
    "    print(\"input_array\", input_array)\n",
    "    print(\"input_array.shape\", input_array.shape)\n",
    "    print(\"input_array[:, 0:1]\", input_array[:, 0:1])\n",
    "    print(\"input_array[:, 0:1].shape\", input_array[:, 0:1].shape)\n",
    "    print(\"input_array[:, 0]\", input_array[:, 0])\n",
    "    print(\"input_array[:, 0].shape\", input_array[:, 0].shape)\n",
    "    return 2 * input_array[:, 0:1]\n",
    "\n",
    "bc = dde.icbc.DirichletBC(\n",
    "    geomtime, \n",
    "    lambda input_array: double_first_column(input_array),\n",
    "    lambda _,\n",
    "    on_boundary: on_boundary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f0e1f09d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pde(comp,u):\n",
    "    du_t = dde.grad.jacobian(u,comp, i=0,j=1)\n",
    "    du_xx = dde.grad.hessian(u,comp, i=0,j=0)\n",
    "    return du_t - k * du_xx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5604246a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warning: 2540 points required, but 2550 points sampled.\n"
     ]
    }
   ],
   "source": [
    "data = dde.data.TimePDE(geomtime,\n",
    "                       pde,\n",
    "                       [bc,ic],\n",
    "                       num_domain = 2540,\n",
    "                       num_boundary = 80,\n",
    "                       num_initial = 160,\n",
    "                       num_test = 2540,\n",
    "                       )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "372238b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "net = dde.nn.FNN([2] + [20]*3+ [1], \"tanh\", \"Glorot normal\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8844069f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10,8))\n",
    "plt.scatter(data.train_x_all[:,0],data.train_x_all[:,1])\n",
    "plt.xlabel('x')\n",
    "plt.ylabel('y')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a6eb15a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = dde.Model(data, net)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f263a5c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Compiling model...\n",
      "'compile' took 0.004281 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model.compile(\"adam\", lr=1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b1382fec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model...\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mecha\\anaconda3\\Lib\\site-packages\\keras\\src\\initializers\\initializers.py:120: UserWarning: The initializer GlorotNormal is unseeded and being called multiple times, which will return identical values each time (even if the initializer is unseeded). Please update your code to provide a seed to the initializer, or avoid using the same initializer instance more than once.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:AutoGraph could not transform <function <lambda> at 0x000001C31AD34FE0> and will run it as-is.\n",
      "Cause: could not parse the source code of <function <lambda> at 0x000001C31AD34FE0>: no matching AST found among candidates:\n",
      "# coding=utf-8\n",
      "lambda x, on: np.array([on_boundary(x[i], on[i]) for i in range(len(x))])\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <function <lambda> at 0x000001C31AD34FE0> and will run it as-is.\n",
      "Cause: could not parse the source code of <function <lambda> at 0x000001C31AD34FE0>: no matching AST found among candidates:\n",
      "# coding=utf-8\n",
      "lambda x, on: np.array([on_boundary(x[i], on[i]) for i in range(len(x))])\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "input_array [[1.        0.796875 ]\n",
      " [0.        0.140625 ]\n",
      " [0.        0.71875  ]\n",
      " [0.        0.96875  ]\n",
      " [1.        0.84375  ]\n",
      " [1.        0.2578125]\n",
      " [1.        0.015625 ]\n",
      " [0.        0.765625 ]\n",
      " [0.        0.859375 ]\n",
      " [1.        0.5078125]\n",
      " [1.        0.5625   ]\n",
      " [0.        0.3828125]\n",
      " [1.        0.7578125]\n",
      " [1.        0.0390625]\n",
      " [1.        0.640625 ]\n",
      " [0.        0.484375 ]\n",
      " [1.        0.125    ]\n",
      " [1.        0.28125  ]\n",
      " [1.        0.90625  ]\n",
      " [0.        0.40625  ]\n",
      " [0.        0.046875 ]\n",
      " [0.        0.1953125]\n",
      " [1.        0.171875 ]\n",
      " [1.        0.109375 ]\n",
      " [1.        0.0078125]\n",
      " [1.        0.203125 ]\n",
      " [0.        0.953125 ]\n",
      " [1.        0.75     ]\n",
      " [1.        0.78125  ]\n",
      " [0.        0.703125 ]\n",
      " [1.        0.6328125]\n",
      " [1.        0.828125 ]\n",
      " [0.        0.234375 ]\n",
      " [0.        0.875    ]\n",
      " [0.        0.21875  ]\n",
      " [0.        0.4453125]\n",
      " [1.        0.9375   ]\n",
      " [0.        0.9453125]\n",
      " [0.        0.65625  ]\n",
      " [1.        0.078125 ]\n",
      " [1.        0.1875   ]\n",
      " [0.        0.890625 ]\n",
      " [1.        0.390625 ]\n",
      " [1.        0.5      ]\n",
      " [1.        0.546875 ]\n",
      " [0.        0.734375 ]\n",
      " [1.        0.359375 ]\n",
      " [0.        0.375    ]\n",
      " [1.        0.15625  ]\n",
      " [0.        0.453125 ]\n",
      " [0.        0.578125 ]\n",
      " [0.        0.984375 ]\n",
      " [0.        0.3203125]\n",
      " [1.        0.4375   ]\n",
      " [1.        0.625    ]\n",
      " [0.        0.265625 ]\n",
      " [0.        0.03125  ]\n",
      " [1.        0.5703125]\n",
      " [1.        0.6953125]\n",
      " [1.        0.921875 ]\n",
      " [0.        0.0625   ]\n",
      " [1.        0.609375 ]\n",
      " [1.        0.53125  ]\n",
      " [1.        0.8125   ]\n",
      " [1.        0.671875 ]\n",
      " [0.        0.8203125]\n",
      " [1.        0.46875  ]\n",
      " [1.        0.3125   ]\n",
      " [1.        0.296875 ]\n",
      " [0.        0.421875 ]\n",
      " [0.        0.25     ]\n",
      " [0.        0.8828125]\n",
      " [1.        0.0703125]\n",
      " [1.        0.59375  ]\n",
      " [1.        0.34375  ]\n",
      " [0.        0.1328125]\n",
      " [1.        0.515625 ]\n",
      " [0.        0.09375  ]\n",
      " [0.        0.6875   ]\n",
      " [1.        0.328125 ]]\n",
      "input_array.shape (80, 2)\n",
      "input_array[:, 0:1] [[1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]]\n",
      "input_array[:, 0:1].shape (80, 1)\n",
      "input_array[:, 0] [1. 0. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0. 1. 1.\n",
      " 1. 1. 0. 1. 1. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 0.\n",
      " 1. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0.\n",
      " 1. 1. 1. 0. 1. 0. 0. 1.]\n",
      "input_array[:, 0].shape (80,)\n",
      "WARNING:tensorflow:AutoGraph could not transform <function <lambda> at 0x000001C31ACE3CE0> and will run it as-is.\n",
      "Cause: could not parse the source code of <function <lambda> at 0x000001C31ACE3CE0>: no matching AST found among candidates:\n",
      "# coding=utf-8\n",
      "lambda x, on: np.array([on_boundary(x[i], on[i]) for i in range(len(x))])\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <function <lambda> at 0x000001C31ACE3CE0> and will run it as-is.\n",
      "Cause: could not parse the source code of <function <lambda> at 0x000001C31ACE3CE0>: no matching AST found among candidates:\n",
      "# coding=utf-8\n",
      "lambda x, on: np.array([on_boundary(x[i], on[i]) for i in range(len(x))])\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "input_array [[1.        0.796875 ]\n",
      " [0.        0.140625 ]\n",
      " [0.        0.71875  ]\n",
      " [0.        0.96875  ]\n",
      " [1.        0.84375  ]\n",
      " [1.        0.2578125]\n",
      " [1.        0.015625 ]\n",
      " [0.        0.765625 ]\n",
      " [0.        0.859375 ]\n",
      " [1.        0.5078125]\n",
      " [1.        0.5625   ]\n",
      " [0.        0.3828125]\n",
      " [1.        0.7578125]\n",
      " [1.        0.0390625]\n",
      " [1.        0.640625 ]\n",
      " [0.        0.484375 ]\n",
      " [1.        0.125    ]\n",
      " [1.        0.28125  ]\n",
      " [1.        0.90625  ]\n",
      " [0.        0.40625  ]\n",
      " [0.        0.046875 ]\n",
      " [0.        0.1953125]\n",
      " [1.        0.171875 ]\n",
      " [1.        0.109375 ]\n",
      " [1.        0.0078125]\n",
      " [1.        0.203125 ]\n",
      " [0.        0.953125 ]\n",
      " [1.        0.75     ]\n",
      " [1.        0.78125  ]\n",
      " [0.        0.703125 ]\n",
      " [1.        0.6328125]\n",
      " [1.        0.828125 ]\n",
      " [0.        0.234375 ]\n",
      " [0.        0.875    ]\n",
      " [0.        0.21875  ]\n",
      " [0.        0.4453125]\n",
      " [1.        0.9375   ]\n",
      " [0.        0.9453125]\n",
      " [0.        0.65625  ]\n",
      " [1.        0.078125 ]\n",
      " [1.        0.1875   ]\n",
      " [0.        0.890625 ]\n",
      " [1.        0.390625 ]\n",
      " [1.        0.5      ]\n",
      " [1.        0.546875 ]\n",
      " [0.        0.734375 ]\n",
      " [1.        0.359375 ]\n",
      " [0.        0.375    ]\n",
      " [1.        0.15625  ]\n",
      " [0.        0.453125 ]\n",
      " [0.        0.578125 ]\n",
      " [0.        0.984375 ]\n",
      " [0.        0.3203125]\n",
      " [1.        0.4375   ]\n",
      " [1.        0.625    ]\n",
      " [0.        0.265625 ]\n",
      " [0.        0.03125  ]\n",
      " [1.        0.5703125]\n",
      " [1.        0.6953125]\n",
      " [1.        0.921875 ]\n",
      " [0.        0.0625   ]\n",
      " [1.        0.609375 ]\n",
      " [1.        0.53125  ]\n",
      " [1.        0.8125   ]\n",
      " [1.        0.671875 ]\n",
      " [0.        0.8203125]\n",
      " [1.        0.46875  ]\n",
      " [1.        0.3125   ]\n",
      " [1.        0.296875 ]\n",
      " [0.        0.421875 ]\n",
      " [0.        0.25     ]\n",
      " [0.        0.8828125]\n",
      " [1.        0.0703125]\n",
      " [1.        0.59375  ]\n",
      " [1.        0.34375  ]\n",
      " [0.        0.1328125]\n",
      " [1.        0.515625 ]\n",
      " [0.        0.09375  ]\n",
      " [0.        0.6875   ]\n",
      " [1.        0.328125 ]]\n",
      "input_array.shape (80, 2)\n",
      "input_array[:, 0:1] [[1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]]\n",
      "input_array[:, 0:1].shape (80, 1)\n",
      "input_array[:, 0] [1. 0. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0. 1. 1.\n",
      " 1. 1. 0. 1. 1. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 0.\n",
      " 1. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0.\n",
      " 1. 1. 1. 0. 1. 0. 0. 1.]\n",
      "input_array[:, 0].shape (80,)\n",
      "input_array [[1.        0.796875 ]\n",
      " [0.        0.140625 ]\n",
      " [0.        0.71875  ]\n",
      " [0.        0.96875  ]\n",
      " [1.        0.84375  ]\n",
      " [1.        0.2578125]\n",
      " [1.        0.015625 ]\n",
      " [0.        0.765625 ]\n",
      " [0.        0.859375 ]\n",
      " [1.        0.5078125]\n",
      " [1.        0.5625   ]\n",
      " [0.        0.3828125]\n",
      " [1.        0.7578125]\n",
      " [1.        0.0390625]\n",
      " [1.        0.640625 ]\n",
      " [0.        0.484375 ]\n",
      " [1.        0.125    ]\n",
      " [1.        0.28125  ]\n",
      " [1.        0.90625  ]\n",
      " [0.        0.40625  ]\n",
      " [0.        0.046875 ]\n",
      " [0.        0.1953125]\n",
      " [1.        0.171875 ]\n",
      " [1.        0.109375 ]\n",
      " [1.        0.0078125]\n",
      " [1.        0.203125 ]\n",
      " [0.        0.953125 ]\n",
      " [1.        0.75     ]\n",
      " [1.        0.78125  ]\n",
      " [0.        0.703125 ]\n",
      " [1.        0.6328125]\n",
      " [1.        0.828125 ]\n",
      " [0.        0.234375 ]\n",
      " [0.        0.875    ]\n",
      " [0.        0.21875  ]\n",
      " [0.        0.4453125]\n",
      " [1.        0.9375   ]\n",
      " [0.        0.9453125]\n",
      " [0.        0.65625  ]\n",
      " [1.        0.078125 ]\n",
      " [1.        0.1875   ]\n",
      " [0.        0.890625 ]\n",
      " [1.        0.390625 ]\n",
      " [1.        0.5      ]\n",
      " [1.        0.546875 ]\n",
      " [0.        0.734375 ]\n",
      " [1.        0.359375 ]\n",
      " [0.        0.375    ]\n",
      " [1.        0.15625  ]\n",
      " [0.        0.453125 ]\n",
      " [0.        0.578125 ]\n",
      " [0.        0.984375 ]\n",
      " [0.        0.3203125]\n",
      " [1.        0.4375   ]\n",
      " [1.        0.625    ]\n",
      " [0.        0.265625 ]\n",
      " [0.        0.03125  ]\n",
      " [1.        0.5703125]\n",
      " [1.        0.6953125]\n",
      " [1.        0.921875 ]\n",
      " [0.        0.0625   ]\n",
      " [1.        0.609375 ]\n",
      " [1.        0.53125  ]\n",
      " [1.        0.8125   ]\n",
      " [1.        0.671875 ]\n",
      " [0.        0.8203125]\n",
      " [1.        0.46875  ]\n",
      " [1.        0.3125   ]\n",
      " [1.        0.296875 ]\n",
      " [0.        0.421875 ]\n",
      " [0.        0.25     ]\n",
      " [0.        0.8828125]\n",
      " [1.        0.0703125]\n",
      " [1.        0.59375  ]\n",
      " [1.        0.34375  ]\n",
      " [0.        0.1328125]\n",
      " [1.        0.515625 ]\n",
      " [0.        0.09375  ]\n",
      " [0.        0.6875   ]\n",
      " [1.        0.328125 ]]\n",
      "input_array.shape (80, 2)\n",
      "input_array[:, 0:1] [[1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]]\n",
      "input_array[:, 0:1].shape (80, 1)\n",
      "input_array[:, 0] [1. 0. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0. 1. 1.\n",
      " 1. 1. 0. 1. 1. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 0.\n",
      " 1. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0.\n",
      " 1. 1. 1. 0. 1. 0. 0. 1.]\n",
      "input_array[:, 0].shape (80,)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Step      Train loss                        Test loss                         Test metric\n",
      "0         [1.74e-02, 7.05e-01, 1.65e-01]    [1.56e-02, 7.05e-01, 1.65e-01]    []  \n",
      "1000      [7.26e-03, 5.43e-02, 4.69e-02]    [3.29e-03, 5.43e-02, 4.69e-02]    []  \n",
      "2000      [5.93e-03, 3.12e-02, 1.96e-02]    [3.72e-03, 3.12e-02, 1.96e-02]    []  \n",
      "3000      [4.67e-03, 2.43e-02, 1.59e-02]    [3.25e-03, 2.43e-02, 1.59e-02]    []  \n",
      "4000      [3.34e-03, 2.01e-02, 1.36e-02]    [2.21e-03, 2.01e-02, 1.36e-02]    []  \n",
      "5000      [6.41e-03, 1.62e-02, 1.36e-02]    [3.08e-03, 1.62e-02, 1.36e-02]    []  \n",
      "6000      [2.63e-03, 1.48e-02, 1.13e-02]    [1.74e-03, 1.48e-02, 1.13e-02]    []  \n",
      "7000      [2.30e-03, 1.28e-02, 1.05e-02]    [1.47e-03, 1.28e-02, 1.05e-02]    []  \n",
      "8000      [2.10e-03, 1.13e-02, 9.54e-03]    [1.38e-03, 1.13e-02, 9.54e-03]    []  \n",
      "9000      [1.93e-03, 9.90e-03, 8.87e-03]    [1.32e-03, 9.90e-03, 8.87e-03]    []  \n",
      "10000     [1.75e-03, 8.64e-03, 8.43e-03]    [1.16e-03, 8.64e-03, 8.43e-03]    []  \n",
      "11000     [1.62e-03, 7.69e-03, 7.91e-03]    [1.01e-03, 7.69e-03, 7.91e-03]    []  \n",
      "12000     [1.55e-03, 6.99e-03, 7.41e-03]    [8.88e-04, 6.99e-03, 7.41e-03]    []  \n",
      "13000     [1.56e-03, 6.41e-03, 6.96e-03]    [8.31e-04, 6.41e-03, 6.96e-03]    []  \n",
      "14000     [3.30e-03, 5.84e-03, 6.76e-03]    [9.30e-04, 5.84e-03, 6.76e-03]    []  \n",
      "15000     [3.44e-03, 5.98e-03, 5.82e-03]    [2.27e-03, 5.98e-03, 5.82e-03]    []  \n",
      "\n",
      "Best model at step 13000:\n",
      "  train loss: 1.49e-02\n",
      "  test loss: 1.42e-02\n",
      "  test metric: []\n",
      "\n",
      "'train' took 40.792252 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "losshistory, train_state = model.train(iterations = 15000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ec29f303",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Compiling model...\n",
      "'compile' took 0.022056 s\n",
      "\n",
      "Training model...\n",
      "\n",
      "input_array [[1.        0.796875 ]\n",
      " [0.        0.140625 ]\n",
      " [0.        0.71875  ]\n",
      " [0.        0.96875  ]\n",
      " [1.        0.84375  ]\n",
      " [1.        0.2578125]\n",
      " [1.        0.015625 ]\n",
      " [0.        0.765625 ]\n",
      " [0.        0.859375 ]\n",
      " [1.        0.5078125]\n",
      " [1.        0.5625   ]\n",
      " [0.        0.3828125]\n",
      " [1.        0.7578125]\n",
      " [1.        0.0390625]\n",
      " [1.        0.640625 ]\n",
      " [0.        0.484375 ]\n",
      " [1.        0.125    ]\n",
      " [1.        0.28125  ]\n",
      " [1.        0.90625  ]\n",
      " [0.        0.40625  ]\n",
      " [0.        0.046875 ]\n",
      " [0.        0.1953125]\n",
      " [1.        0.171875 ]\n",
      " [1.        0.109375 ]\n",
      " [1.        0.0078125]\n",
      " [1.        0.203125 ]\n",
      " [0.        0.953125 ]\n",
      " [1.        0.75     ]\n",
      " [1.        0.78125  ]\n",
      " [0.        0.703125 ]\n",
      " [1.        0.6328125]\n",
      " [1.        0.828125 ]\n",
      " [0.        0.234375 ]\n",
      " [0.        0.875    ]\n",
      " [0.        0.21875  ]\n",
      " [0.        0.4453125]\n",
      " [1.        0.9375   ]\n",
      " [0.        0.9453125]\n",
      " [0.        0.65625  ]\n",
      " [1.        0.078125 ]\n",
      " [1.        0.1875   ]\n",
      " [0.        0.890625 ]\n",
      " [1.        0.390625 ]\n",
      " [1.        0.5      ]\n",
      " [1.        0.546875 ]\n",
      " [0.        0.734375 ]\n",
      " [1.        0.359375 ]\n",
      " [0.        0.375    ]\n",
      " [1.        0.15625  ]\n",
      " [0.        0.453125 ]\n",
      " [0.        0.578125 ]\n",
      " [0.        0.984375 ]\n",
      " [0.        0.3203125]\n",
      " [1.        0.4375   ]\n",
      " [1.        0.625    ]\n",
      " [0.        0.265625 ]\n",
      " [0.        0.03125  ]\n",
      " [1.        0.5703125]\n",
      " [1.        0.6953125]\n",
      " [1.        0.921875 ]\n",
      " [0.        0.0625   ]\n",
      " [1.        0.609375 ]\n",
      " [1.        0.53125  ]\n",
      " [1.        0.8125   ]\n",
      " [1.        0.671875 ]\n",
      " [0.        0.8203125]\n",
      " [1.        0.46875  ]\n",
      " [1.        0.3125   ]\n",
      " [1.        0.296875 ]\n",
      " [0.        0.421875 ]\n",
      " [0.        0.25     ]\n",
      " [0.        0.8828125]\n",
      " [1.        0.0703125]\n",
      " [1.        0.59375  ]\n",
      " [1.        0.34375  ]\n",
      " [0.        0.1328125]\n",
      " [1.        0.515625 ]\n",
      " [0.        0.09375  ]\n",
      " [0.        0.6875   ]\n",
      " [1.        0.328125 ]]\n",
      "input_array.shape (80, 2)\n",
      "input_array[:, 0:1] [[1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]]\n",
      "input_array[:, 0:1].shape (80, 1)\n",
      "input_array[:, 0] [1. 0. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0. 1. 1.\n",
      " 1. 1. 0. 1. 1. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 0.\n",
      " 1. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0.\n",
      " 1. 1. 1. 0. 1. 0. 0. 1.]\n",
      "input_array[:, 0].shape (80,)\n",
      "input_array [[1.        0.796875 ]\n",
      " [0.        0.140625 ]\n",
      " [0.        0.71875  ]\n",
      " [0.        0.96875  ]\n",
      " [1.        0.84375  ]\n",
      " [1.        0.2578125]\n",
      " [1.        0.015625 ]\n",
      " [0.        0.765625 ]\n",
      " [0.        0.859375 ]\n",
      " [1.        0.5078125]\n",
      " [1.        0.5625   ]\n",
      " [0.        0.3828125]\n",
      " [1.        0.7578125]\n",
      " [1.        0.0390625]\n",
      " [1.        0.640625 ]\n",
      " [0.        0.484375 ]\n",
      " [1.        0.125    ]\n",
      " [1.        0.28125  ]\n",
      " [1.        0.90625  ]\n",
      " [0.        0.40625  ]\n",
      " [0.        0.046875 ]\n",
      " [0.        0.1953125]\n",
      " [1.        0.171875 ]\n",
      " [1.        0.109375 ]\n",
      " [1.        0.0078125]\n",
      " [1.        0.203125 ]\n",
      " [0.        0.953125 ]\n",
      " [1.        0.75     ]\n",
      " [1.        0.78125  ]\n",
      " [0.        0.703125 ]\n",
      " [1.        0.6328125]\n",
      " [1.        0.828125 ]\n",
      " [0.        0.234375 ]\n",
      " [0.        0.875    ]\n",
      " [0.        0.21875  ]\n",
      " [0.        0.4453125]\n",
      " [1.        0.9375   ]\n",
      " [0.        0.9453125]\n",
      " [0.        0.65625  ]\n",
      " [1.        0.078125 ]\n",
      " [1.        0.1875   ]\n",
      " [0.        0.890625 ]\n",
      " [1.        0.390625 ]\n",
      " [1.        0.5      ]\n",
      " [1.        0.546875 ]\n",
      " [0.        0.734375 ]\n",
      " [1.        0.359375 ]\n",
      " [0.        0.375    ]\n",
      " [1.        0.15625  ]\n",
      " [0.        0.453125 ]\n",
      " [0.        0.578125 ]\n",
      " [0.        0.984375 ]\n",
      " [0.        0.3203125]\n",
      " [1.        0.4375   ]\n",
      " [1.        0.625    ]\n",
      " [0.        0.265625 ]\n",
      " [0.        0.03125  ]\n",
      " [1.        0.5703125]\n",
      " [1.        0.6953125]\n",
      " [1.        0.921875 ]\n",
      " [0.        0.0625   ]\n",
      " [1.        0.609375 ]\n",
      " [1.        0.53125  ]\n",
      " [1.        0.8125   ]\n",
      " [1.        0.671875 ]\n",
      " [0.        0.8203125]\n",
      " [1.        0.46875  ]\n",
      " [1.        0.3125   ]\n",
      " [1.        0.296875 ]\n",
      " [0.        0.421875 ]\n",
      " [0.        0.25     ]\n",
      " [0.        0.8828125]\n",
      " [1.        0.0703125]\n",
      " [1.        0.59375  ]\n",
      " [1.        0.34375  ]\n",
      " [0.        0.1328125]\n",
      " [1.        0.515625 ]\n",
      " [0.        0.09375  ]\n",
      " [0.        0.6875   ]\n",
      " [1.        0.328125 ]]\n",
      "input_array.shape (80, 2)\n",
      "input_array[:, 0:1] [[1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]\n",
      " [1.]\n",
      " [1.]\n",
      " [0.]\n",
      " [1.]\n",
      " [0.]\n",
      " [0.]\n",
      " [1.]]\n",
      "input_array[:, 0:1].shape (80, 1)\n",
      "input_array[:, 0] [1. 0. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0. 1. 1.\n",
      " 1. 1. 0. 1. 1. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 1. 1. 1. 0. 1. 0.\n",
      " 1. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 0. 0.\n",
      " 1. 1. 1. 0. 1. 0. 0. 1.]\n",
      "input_array[:, 0].shape (80,)\n",
      "Step      Train loss                        Test loss                         Test metric\n",
      "15000     [3.44e-03, 5.98e-03, 5.82e-03]    [2.27e-03, 5.98e-03, 5.82e-03]    []  \n",
      "16790     [6.86e-04, 2.32e-04, 3.34e-04]    [8.84e-04, 2.32e-04, 3.34e-04]    []  \n",
      "\n",
      "Best model at step 16790:\n",
      "  train loss: 1.25e-03\n",
      "  test loss: 1.45e-03\n",
      "  test metric: []\n",
      "\n",
      "'train' took 211.147752 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Limited-memory Broyden-Fletcher-Goldfarb-Shanno\n",
    "model.compile(\"L-BFGS-B\")\n",
    "losshistory, train_state = model.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f152dd46",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving loss history to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Course_5_pinns\\05_deepxde_1D_pde_heat\\loss.dat ...\n",
      "Saving training data to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Course_5_pinns\\05_deepxde_1D_pde_heat\\train.dat ...\n",
      "Saving test data to C:\\Users\\mecha\\OneDrive\\デスクトップ\\main\\Course_5_pinns\\05_deepxde_1D_pde_heat\\test.dat ...\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dde.saveplot(losshistory,train_state, issave= True, isplot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9371a47c",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}