{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7584fbf6",
   "metadata": {},
   "source": [
    "# File input and output with arrays\n",
    "\n",
    "NumPy is able to store data in some text or binary formats on disk and load it from there. However, in this section I only discuss NumPy's own binary format, as mostly pandas or other tools are used to load text or table data (see [Read, persist and provide data](../../data-processing/index.rst).\n",
    "\n",
    "`np.save` and `np.load` are the two most important functions for efficiently saving and loading array data to disk. Arrays are saved by default in an uncompressed raw binary format with the file extension `.npy`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d6c92106",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-17T15:00:29.001682Z",
     "iopub.status.busy": "2026-05-17T15:00:29.001384Z",
     "iopub.status.idle": "2026-05-17T15:00:29.046528Z",
     "shell.execute_reply": "2026-05-17T15:00:29.046233Z",
     "shell.execute_reply.started": "2026-05-17T15:00:29.001662Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "rng = np.random.default_rng()\n",
    "data = rng.random((7, 3))\n",
    "\n",
    "np.save(\"my_data\", data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9b70227",
   "metadata": {},
   "source": [
    "If the file path does not already end in `.npy`, the extension is appended. The array on the hard disk can then be loaded with `np.load`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ed098f7c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-17T15:00:29.047077Z",
     "iopub.status.busy": "2026-05-17T15:00:29.046981Z",
     "iopub.status.idle": "2026-05-17T15:00:29.050509Z",
     "shell.execute_reply": "2026-05-17T15:00:29.050283Z",
     "shell.execute_reply.started": "2026-05-17T15:00:29.047069Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.69327601, 0.82653332, 0.40657863],\n",
       "       [0.41500074, 0.08436697, 0.75347185],\n",
       "       [0.26657504, 0.65470011, 0.40720279],\n",
       "       [0.32698017, 0.11859362, 0.21459654],\n",
       "       [0.86305016, 0.81381353, 0.12060564],\n",
       "       [0.91046405, 0.58527599, 0.92125107],\n",
       "       [0.85952529, 0.82071675, 0.77440136]])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.load(\"my_data.npy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "467f7c50",
   "metadata": {},
   "source": [
    "You can save multiple arrays in an uncompressed archive by using `np.savez` and passing the arrays as keyword arguments:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3df6c190",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-17T15:00:29.050943Z",
     "iopub.status.busy": "2026-05-17T15:00:29.050842Z",
     "iopub.status.idle": "2026-05-17T15:00:29.053301Z",
     "shell.execute_reply": "2026-05-17T15:00:29.053082Z",
     "shell.execute_reply.started": "2026-05-17T15:00:29.050930Z"
    }
   },
   "outputs": [],
   "source": [
    "np.savez(\"data_archive.npz\", a=data, b=np.square(data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "14ea0da5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-17T15:00:29.053662Z",
     "iopub.status.busy": "2026-05-17T15:00:29.053593Z",
     "iopub.status.idle": "2026-05-17T15:00:29.056955Z",
     "shell.execute_reply": "2026-05-17T15:00:29.056680Z",
     "shell.execute_reply.started": "2026-05-17T15:00:29.053656Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.48063163, 0.68315733, 0.16530618],\n",
       "       [0.17222561, 0.00711779, 0.56771982],\n",
       "       [0.07106225, 0.42863223, 0.16581411],\n",
       "       [0.10691603, 0.01406445, 0.04605167],\n",
       "       [0.74485557, 0.66229247, 0.01454572],\n",
       "       [0.82894478, 0.34254799, 0.84870353],\n",
       "       [0.73878372, 0.67357598, 0.59969746]])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "archive = np.load(\"data_archive.npz\")\n",
    "\n",
    "archive[\"b\"]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.13 Kernel",
   "language": "python",
   "name": "python313"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}