LSTMCell Output

lstm_cell_output.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c1333286",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4bd8e564",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_size = 5\n",
    "hidden_size = 8\n",
    "cell = nn.LSTMCell(input_size=input_size, hidden_size=hidden_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e4c9ff01",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[-1.4978, -1.0567, -1.9804, -0.0280,  0.4982],\n",
       "        [ 0.0765, -0.7857,  0.6896, -0.8095, -0.2884],\n",
       "        [-0.4736,  1.1645, -0.3312,  0.5085,  1.2523]])"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "input_tensor = torch.randn(3, input_size)\n",
    "input_tensor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d3ad7340",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[-0.0045,  0.2381, -0.0528, -0.0542,  0.1259, -0.1155, -0.0484,  0.0413],\n",
       "         [-0.0841,  0.0597,  0.0675, -0.1352, -0.0363, -0.0520,  0.0694, -0.0789],\n",
       "         [ 0.0259,  0.1760, -0.1397, -0.0058,  0.0424, -0.1639, -0.0007, -0.0492]],\n",
       "        grad_fn=<MulBackward0>),\n",
       " tensor([[-0.0072,  0.3069, -0.1259, -0.2533,  0.2088, -0.1996, -0.1047,  0.0992],\n",
       "         [-0.1583,  0.0992,  0.1276, -0.2523, -0.0539, -0.0919,  0.1993, -0.1501],\n",
       "         [ 0.0423,  0.3415, -0.2315, -0.0178,  0.0849, -0.3342, -0.0022, -0.1224]],\n",
       "        grad_fn=<AddBackward0>))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_tensor, state = cell(input_tensor)\n",
    "output_tensor, state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b7b5e77f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([3, 8]), torch.Size([3, 8]))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_tensor.shape, state.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7bd296a8",
   "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": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}