DAY35_Revisiting_RNN_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ba372692",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import torch.optim as optim\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b863e2ba",
   "metadata": {},
   "source": [
    "### RNN vs RNNCell in pytorch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "60a7e73f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RNNCell(10, 20)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = nn.RNNCell(10, 20)\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e9e6dbd3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([20, 10]), torch.Size([20, 20]))"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.weight_ih.shape, model.weight_hh.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "98f0a33e",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = torch.randn(5, 10)\n",
    "h_0 = torch.randn(1,20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "08d0bdc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([1, 10]), torch.Size([1, 20]))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x[0].reshape(1,-1).shape, h_0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c5ffc909",
   "metadata": {},
   "outputs": [],
   "source": [
    "output = []\n",
    "for i in range(x.size(0)):\n",
    "    h_0 = model(x[i].reshape(1,-1), h_0)\n",
    "    output.append(h_0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "971cb755",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([5, 1, 20])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.stack(output).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "857f8252",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'list' object has no attribute 'shape'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-8-961ed67232b2>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0moutput\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m: 'list' object has no attribute 'shape'"
     ]
    }
   ],
   "source": [
    "output.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9c63607b",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = nn.RNN(10, 20)\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "18c8a94a",
   "metadata": {},
   "outputs": [],
   "source": [
    "h_0 = torch.randn(5,20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "246df7cf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([1, 5, 10]), torch.Size([5, 20]))"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.unsqueeze(0).shape, h_0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "af79a14e",
   "metadata": {},
   "outputs": [
    {
     "ename": "RuntimeError",
     "evalue": "input has inconsistent input_size: got 5 expected 10",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-11-e7f1f77fde4e>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mout\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mh_out\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0munsqueeze\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mh_0\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0munsqueeze\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32m~\\anaconda3\\envs\\boston_house_prices_prediction\\lib\\site-packages\\torch\\nn\\modules\\module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[1;34m(self, *input, **kwargs)\u001b[0m\n\u001b[0;32m   1049\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[0;32m   1050\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[1;32m-> 1051\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1052\u001b[0m         \u001b[1;31m# Do not call functions when jit is used\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1053\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\envs\\boston_house_prices_prediction\\lib\\site-packages\\torch\\nn\\modules\\rnn.py\u001b[0m in \u001b[0;36mforward\u001b[1;34m(self, input, hx)\u001b[0m\n\u001b[0;32m    964\u001b[0m             \u001b[0mhx\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mzeros\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msize\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mhidden_size\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdtype\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0minput\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdevice\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    965\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnonlinearity\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;34m\"tanh\"\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 966\u001b[1;33m             ret = _VF.rnn_tanh_cell(\n\u001b[0m\u001b[0;32m    967\u001b[0m                 \u001b[0minput\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhx\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    968\u001b[0m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mweight_ih\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mweight_hh\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mRuntimeError\u001b[0m: input has inconsistent input_size: got 5 expected 10"
     ]
    }
   ],
   "source": [
    "out, h_out = model(x.unsqueeze(0), h_0.unsqueeze(0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "4c16e9e9",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'out' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-12-f181230b10d0>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mout\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mh_out\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'out' is not defined"
     ]
    }
   ],
   "source": [
    "out.shape, h_out.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3958dc0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_time_sequence(T=20, L=1000, N=100):\n",
    "    np.random.seed(2)\n",
    "    x = np.empty((N, L), 'int64')\n",
    "    x[:] = np.array(range(L)) + np.random.randint(-4 * T, 4 * T, N).reshape(N, 1)\n",
    "    data = np.sin(x / 1.0 / T).astype('float64')\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "2154de6c",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = generate_time_sequence()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d1ef33ed",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([100, 1000])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.tensor(data).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a20887fd",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x21c951613d0>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data[3][:-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a6a88ffe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x21c95298f10>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data[3][1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c7f90ca7",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Sequence(nn.Module):\n",
    "    def __init__(self, input_size=1, hidden_size = 100):\n",
    "        super().__init__()\n",
    "        self.hidden_size = hidden_size\n",
    "        self.rnn = nn.RNNCell(input_size, hidden_size)\n",
    "        self.Linear = nn.Linear(hidden_size, input_size)\n",
    "    \n",
    "    def forward(self, x):\n",
    "        h_0 = torch.zeros(x.size(0), self.hidden_size)\n",
    "        out = []\n",
    "        for i in x.split(1,dim=1):\n",
    "            h_0 = self.rnn(i, h_0)\n",
    "            out_ = self.Linear(h_0)\n",
    "            out += [out_]\n",
    "        return torch.stack(out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "10b7fea8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequence(\n",
       "  (rnn): RNNCell(1, 100)\n",
       "  (Linear): Linear(in_features=100, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = Sequence()\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b93e9b2c",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = torch.from_numpy(data[3:, :-1]).type(torch.float32)\n",
    "target = torch.from_numpy(data[3:, 1:]).type(torch.float32)\n",
    "x_test = torch.from_numpy(data[:3, :-1]).type(torch.float32)\n",
    "target_test = torch.from_numpy(data[:3, 1:]).type(torch.float32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "16ef0ddb",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(x.shape, target.shape)\n",
    "print(x_test.shape, target_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "863106da",
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.MSELoss()\n",
    "optimizer = optim.LBFGS(model.parameters(), lr=0.8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ed72ec19",
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs=100\n",
    "for i in tqdm(range(epochs)):\n",
    "    print(f\"Step : {i}\")\n",
    "    def closure():\n",
    "        optimizer.zero_grad()\n",
    "        out = model(x)\n",
    "        loss = criterion(out.reshape(97, 999), target)\n",
    "#         print('loss:', loss.item())\n",
    "        loss.backward()\n",
    "        return loss\n",
    "    optimizer.step(closure)\n",
    "    with torch.no_grad():\n",
    "        pred = model(x_test)\n",
    "        loss = criterion(pred.reshape(3, 999), target_test)\n",
    "        print('test loss:', loss.item())\n",
    "        y = pred.detach().numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c9f925a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "37503b3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "output = model(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b7632854",
   "metadata": {},
   "outputs": [],
   "source": [
    "output = output.reshape(97, 999)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ea85368",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = output.detach().numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "22747b9b",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_sample = x.detach().numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1076ddf8",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(input_sample[0][1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e3ef2219",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(x[0][1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dea64a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Sequence2(nn.Module):\n",
    "    def __init__(self, input_size=1, hidden_size = 100):\n",
    "        super().__init__()\n",
    "        self.hidden_size = hidden_size\n",
    "        self.rnn = nn.RNN(input_size, hidden_size)\n",
    "        self.Linear = nn.Linear(hidden_size, input_size)\n",
    "    \n",
    "    def forward(self, x):\n",
    "        h_0 = torch.zeros(x.size(0), self.hidden_size)\n",
    "        out = []\n",
    "        for i in x.split(1,dim=1):\n",
    "            h_0 = self.rnn(i, h_0)\n",
    "            out_ = self.Linear(h_0)\n",
    "            out += [out_]\n",
    "        return torch.stack(out)"
   ]
  }
 ],
 "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": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}