{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "530feebe",
   "metadata": {},
   "source": [
    "`NeuralODE.reference.ipynb` \n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bd0a3fa6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Version: 0.3.2\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 1;\n",
       "                var nbb_unformatted_code = \"%load_ext nb_black\\n# pip install neural-diffeqs\\n\\nimport neural_diffeqs\\n\\nprint(f\\\"Version: {neural_diffeqs.__version__}\\\")\\nimport torch\";\n",
       "                var nbb_formatted_code = \"%load_ext nb_black\\n# pip install neural-diffeqs\\n\\nimport neural_diffeqs\\n\\nprint(f\\\"Version: {neural_diffeqs.__version__}\\\")\\nimport torch\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %load_ext nb_black\n",
    "# pip install neural-diffeqs\n",
    "\n",
    "import neural_diffeqs\n",
    "\n",
    "print(f\"Version: {neural_diffeqs.__version__}\")\n",
    "import torch"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "763caa74",
   "metadata": {},
   "source": [
    "### Default `NeuralODE`\n",
    "\n",
    "The only required parameter is:\n",
    "\n",
    "* `state_size`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5b650fd1",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NeuralODE(\n",
      "  (mu): TorchNet(\n",
      "    (hidden_1): Sequential(\n",
      "      (linear): Linear(in_features=50, out_features=512, bias=True)\n",
      "      (activation): LeakyReLU(negative_slope=0.01)\n",
      "    )\n",
      "    (hidden_2): Sequential(\n",
      "      (linear): Linear(in_features=512, out_features=512, bias=True)\n",
      "      (activation): LeakyReLU(negative_slope=0.01)\n",
      "    )\n",
      "    (output): Sequential(\n",
      "      (linear): Linear(in_features=512, out_features=50, bias=True)\n",
      "    )\n",
      "  )\n",
      ")\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 2;\n",
       "                var nbb_unformatted_code = \"ODE = neural_diffeqs.NeuralODE(state_size=50)\\nprint(ODE)\";\n",
       "                var nbb_formatted_code = \"ODE = neural_diffeqs.NeuralODE(state_size=50)\\nprint(ODE)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ODE = neural_diffeqs.NeuralODE(state_size=50)\n",
    "print(ODE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49615c08",
   "metadata": {},
   "source": [
    "### Changing some parameters\n",
    "\n",
    "For example, specify the hidden state size for each network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ed36e1dd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NeuralODE(\n",
      "  (mu): TorchNet(\n",
      "    (hidden_1): Sequential(\n",
      "      (linear): Linear(in_features=50, out_features=64, bias=True)\n",
      "      (activation): LeakyReLU(negative_slope=0.01)\n",
      "    )\n",
      "    (hidden_2): Sequential(\n",
      "      (linear): Linear(in_features=64, out_features=128, bias=True)\n",
      "      (activation): LeakyReLU(negative_slope=0.01)\n",
      "    )\n",
      "    (hidden_3): Sequential(\n",
      "      (linear): Linear(in_features=128, out_features=64, bias=True)\n",
      "      (activation): LeakyReLU(negative_slope=0.01)\n",
      "    )\n",
      "    (output): Sequential(\n",
      "      (linear): Linear(in_features=64, out_features=50, bias=True)\n",
      "    )\n",
      "  )\n",
      ")\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 3;\n",
       "                var nbb_unformatted_code = \"ODE = neural_diffeqs.NeuralODE(\\n    state_size=50, mu_hidden=[64, 128, 64]\\n)\\nprint(ODE)\";\n",
       "                var nbb_formatted_code = \"ODE = neural_diffeqs.NeuralODE(state_size=50, mu_hidden=[64, 128, 64])\\nprint(ODE)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ODE = neural_diffeqs.NeuralODE(state_size=50, mu_hidden=[64, 128, 64])\n",
    "print(ODE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66dcbecc",
   "metadata": {},
   "source": [
    "### Activation functions, dropout, and bias"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1c822988",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NeuralODE(\n",
      "  (mu): TorchNet(\n",
      "    (hidden_1): Sequential(\n",
      "      (linear): Linear(in_features=50, out_features=64, bias=True)\n",
      "      (activation): Softmax(dim=None)\n",
      "    )\n",
      "    (hidden_2): Sequential(\n",
      "      (linear): Linear(in_features=64, out_features=128, bias=True)\n",
      "      (dropout): Dropout(p=0.2, inplace=False)\n",
      "      (activation): Tanh()\n",
      "    )\n",
      "    (hidden_3): Sequential(\n",
      "      (linear): Linear(in_features=128, out_features=128, bias=True)\n",
      "      (activation): Tanh()\n",
      "    )\n",
      "    (hidden_4): Sequential(\n",
      "      (linear): Linear(in_features=128, out_features=64, bias=True)\n",
      "      (activation): Tanh()\n",
      "    )\n",
      "    (output): Sequential(\n",
      "      (linear): Linear(in_features=64, out_features=50, bias=True)\n",
      "    )\n",
      "  )\n",
      ")\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 4;\n",
       "                var nbb_unformatted_code = \"ODE = neural_diffeqs.NeuralODE(\\n    state_size=50,\\n    mu_hidden=[64, 128, 128, 64],\\n    mu_activation=[torch.nn.Softmax, torch.nn.Tanh],\\n    mu_dropout=[0, 0.2, 0]\\n)\\nprint(ODE)\";\n",
       "                var nbb_formatted_code = \"ODE = neural_diffeqs.NeuralODE(\\n    state_size=50,\\n    mu_hidden=[64, 128, 128, 64],\\n    mu_activation=[torch.nn.Softmax, torch.nn.Tanh],\\n    mu_dropout=[0, 0.2, 0],\\n)\\nprint(ODE)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ODE = neural_diffeqs.NeuralODE(\n",
    "    state_size=50,\n",
    "    mu_hidden=[64, 128, 128, 64],\n",
    "    mu_activation=[torch.nn.Softmax, torch.nn.Tanh],\n",
    "    mu_dropout=[0, 0.2, 0],\n",
    ")\n",
    "print(ODE)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:sdq]",
   "language": "python",
   "name": "conda-env-sdq-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}