{
 "cells": [
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:22.530839Z",
     "start_time": "2025-06-12T02:34:20.159404Z"
    }
   },
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "\n",
    "torch.__version__"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.2.1'"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 1
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:24.139479Z",
     "start_time": "2025-06-12T02:34:24.116052Z"
    }
   },
   "cell_type": "code",
   "source": [
    "a = torch.tensor(1)\n",
    "b = a.item()\n",
    "print(a)\n",
    "print(b)"
   ],
   "id": "ec73dc2f6feeece4",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(1)\n",
      "1\n"
     ]
    }
   ],
   "execution_count": 2
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:26.619937Z",
     "start_time": "2025-06-12T02:34:26.608636Z"
    }
   },
   "cell_type": "code",
   "source": [
    "a = [1, 2, 3]\n",
    "b = torch.tensor(a)\n",
    "c = b.numpy().tolist()\n",
    "print(c)"
   ],
   "id": "6c3e0063d8fcc299",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 3]\n"
     ]
    }
   ],
   "execution_count": 3
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:28.340829Z",
     "start_time": "2025-06-12T02:34:28.333276Z"
    }
   },
   "cell_type": "code",
   "source": [
    "a = torch.zeros(2, 3, 5)\n",
    "print(a.shape)\n",
    "\n",
    "print(a.size())\n",
    "\n",
    "print(a.numel())\n"
   ],
   "id": "d04c60d3f01351c2",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([2, 3, 5])\n",
      "torch.Size([2, 3, 5])\n",
      "30\n"
     ]
    }
   ],
   "execution_count": 4
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:37.031549Z",
     "start_time": "2025-06-12T02:34:36.991394Z"
    }
   },
   "cell_type": "code",
   "source": [
    "x = torch.rand(2, 3, 5)\n",
    "print(x.shape)\n",
    "print(x)"
   ],
   "id": "774add1439f9aa94",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([2, 3, 5])\n",
      "tensor([[[0.1437, 0.3582, 0.4219, 0.4514, 0.6537],\n",
      "         [0.0089, 0.5737, 0.0201, 0.7728, 0.1827],\n",
      "         [0.6573, 0.1262, 0.0877, 0.2302, 0.0151]],\n",
      "\n",
      "        [[0.0757, 0.7126, 0.4238, 0.0535, 0.0578],\n",
      "         [0.4909, 0.5616, 0.7342, 0.7925, 0.8879],\n",
      "         [0.3011, 0.1606, 0.2856, 0.8165, 0.4100]]])\n"
     ]
    }
   ],
   "execution_count": 6
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:34:48.128975Z",
     "start_time": "2025-06-12T02:34:48.116080Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# 矩阵转秩\n",
    "x = x.permute(2, 1, 0)\n",
    "print(x.shape)\n",
    "print(x)"
   ],
   "id": "ceb1debce1c62ffd",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([5, 3, 2])\n",
      "tensor([[[0.1437, 0.0757],\n",
      "         [0.0089, 0.4909],\n",
      "         [0.6573, 0.3011]],\n",
      "\n",
      "        [[0.3582, 0.7126],\n",
      "         [0.5737, 0.5616],\n",
      "         [0.1262, 0.1606]],\n",
      "\n",
      "        [[0.4219, 0.4238],\n",
      "         [0.0201, 0.7342],\n",
      "         [0.0877, 0.2856]],\n",
      "\n",
      "        [[0.4514, 0.0535],\n",
      "         [0.7728, 0.7925],\n",
      "         [0.2302, 0.8165]],\n",
      "\n",
      "        [[0.6537, 0.0578],\n",
      "         [0.1827, 0.8879],\n",
      "         [0.0151, 0.4100]]])\n"
     ]
    }
   ],
   "execution_count": 7
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:36:03.393949Z",
     "start_time": "2025-06-12T02:36:03.381874Z"
    }
   },
   "cell_type": "code",
   "source": [
    "x = torch.rand(2, 3, 4)\n",
    "x = x.transpose(1, 0)\n",
    "print(x.shape)"
   ],
   "id": "e56528fc1753cf04",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([3, 2, 4])\n"
     ]
    }
   ],
   "execution_count": 8
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T02:49:57.093978Z",
     "start_time": "2025-06-12T02:49:57.088700Z"
    }
   },
   "cell_type": "code",
   "source": [
    "x = torch.rand(4, 4)\n",
    "x = x.view(2, 8)\n",
    "x = x.permute(1, 0)\n",
    "# x.view(4,4) # 不能直接用view，因为view需要连续的内存\n",
    "x.reshape(4, 4)"
   ],
   "id": "74df80d1396ec1d3",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([2, 8])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 13
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T03:34:30.576671Z",
     "start_time": "2025-06-12T03:34:30.569216Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# 增减维度\n",
    "x = torch.rand(2, 1, 3)\n",
    "print(x)\n",
    "x = x.squeeze(1)  # 去掉维度为1的维度\n",
    "\n",
    "print(x.shape)\n",
    "print(x)\n",
    "\n"
   ],
   "id": "f5009aa3b8b1335c",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[[0.0287, 0.7995, 0.4072]],\n",
      "\n",
      "        [[0.4378, 0.6384, 0.2777]]])\n",
      "torch.Size([2, 3])\n",
      "tensor([[0.0287, 0.7995, 0.4072],\n",
      "        [0.4378, 0.6384, 0.2777]])\n"
     ]
    }
   ],
   "execution_count": 29
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T03:42:20.284801Z",
     "start_time": "2025-06-12T03:42:20.271042Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# 增减维度\n",
    "x = torch.rand(2, 1, 3)\n",
    "print(x)\n",
    "x = x.unsqueeze()  # 去掉维度为1的维度\n",
    "\n",
    "print(x.shape)\n",
    "print(x)\n",
    "\n"
   ],
   "id": "dc138eb85bed2f3e",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[[0.4243, 0.1581, 0.4620]],\n",
      "\n",
      "        [[0.8510, 0.5490, 0.7694]]])\n",
      "torch.Size([2, 1, 1, 3])\n",
      "tensor([[[[0.4243, 0.1581, 0.4620]]],\n",
      "\n",
      "\n",
      "        [[[0.8510, 0.5490, 0.7694]]]])\n"
     ]
    }
   ],
   "execution_count": 30
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}