{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.optim as optim\n",
    "import torch.nn as nn\n",
    "\n",
    "torch.set_printoptions(edgeitems=2, linewidth=75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequential(\n",
       "  (0): Linear(in_features=1, out_features=11, bias=True)\n",
       "  (1): Tanh()\n",
       "  (2): Linear(in_features=11, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "seq_model = nn.Sequential(\n",
    "            nn.Linear(1, 11), # <1>\n",
    "            nn.Tanh(),\n",
    "            nn.Linear(11, 1)) # <1>\n",
    "seq_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequential(\n",
       "  (hidden_linear): Linear(in_features=1, out_features=12, bias=True)\n",
       "  (hidden_activation): Tanh()\n",
       "  (output_linear): Linear(in_features=12, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from collections import OrderedDict\n",
    "\n",
    "namedseq_model = nn.Sequential(OrderedDict([\n",
    "    ('hidden_linear', nn.Linear(1, 12)),\n",
    "    ('hidden_activation', nn.Tanh()),\n",
    "    ('output_linear', nn.Linear(12 , 1))\n",
    "]))\n",
    "\n",
    "namedseq_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SubclassModel(\n",
       "  (hidden_linear): Linear(in_features=1, out_features=13, bias=True)\n",
       "  (hidden_activation): Tanh()\n",
       "  (output_linear): Linear(in_features=13, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class SubclassModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()  # <1>\n",
    "        \n",
    "        self.hidden_linear = nn.Linear(1, 13)\n",
    "        self.hidden_activation = nn.Tanh()\n",
    "        self.output_linear = nn.Linear(13, 1)\n",
    "        \n",
    "    def forward(self, input):\n",
    "        hidden_t = self.hidden_linear(input)\n",
    "        activated_t = self.hidden_activation(hidden_t)\n",
    "        output_t = self.output_linear(activated_t)\n",
    "        \n",
    "        return output_t\n",
    "    \n",
    "subclass_model = SubclassModel()\n",
    "subclass_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "seq\n"
     ]
    },
    {
     "ename": "TypeError",
     "evalue": "unsupported format string passed to torch.Size.__format__",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-8-4f1be40ed447>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtype_str\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mname_str\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparam\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnamed_parameters\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m         \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"{name_str:21} {param.shape:19} {param.numel()}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mTypeError\u001b[0m: unsupported format string passed to torch.Size.__format__"
     ]
    }
   ],
   "source": [
    "for type_str, model in [('seq', seq_model),\n",
    "                        ('namedseq', namedseq_model),\n",
    "                        ('subclass', subclass_model)]:\n",
    "    print(type_str)\n",
    "    for name_str, param in model.named_parameters():\n",
    "        print(\"{:21} {:19} {}\".format(\n",
    "            name_str, str(param.shape), param.numel()))\n",
    "        \n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SubclassFunctionalModel(\n",
       "  (hidden_linear): Linear(in_features=1, out_features=14, bias=True)\n",
       "  (output_linear): Linear(in_features=14, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class SubclassFunctionalModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        \n",
    "        self.hidden_linear = nn.Linear(1, 14)  \n",
    "                                                # <1>\n",
    "        self.output_linear = nn.Linear(14, 1)\n",
    "        \n",
    "    def forward(self, input):\n",
    "        hidden_t = self.hidden_linear(input)\n",
    "        activated_t = torch.tanh(hidden_t) # <2>\n",
    "        output_t = self.output_linear(activated_t)\n",
    "        \n",
    "        return output_t\n",
    "    \n",
    "func_model = SubclassFunctionalModel()\n",
    "func_model"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}