{
 "cells": [
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": "import torch",
   "id": "d3cf166c4c1b5e28"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "# 连接操作\n",
    "A = torch.ones(3, 3)\n",
    "B = 2 * torch.ones(3, 3)\n",
    "\n",
    "C = torch.cat((A, B), 1)\n",
    "C"
   ],
   "id": "8995ad1fc7997846"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "# 堆叠操作\n",
    "A = torch.arange(0, 4)\n",
    "B = torch.arange(4, 8)\n",
    "\n",
    "C = torch.stack((A, B), 1)\n",
    "C"
   ],
   "id": "3bc82f18d81a5906"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "# 堆叠操作\n",
    "A = torch.arange(6).reshape(2, 3)\n",
    "B = torch.arange(7, 13).reshape(2, 3)\n",
    "C = torch.stack((A, B), 2)\n",
    "print(C.shape)\n",
    "C"
   ],
   "id": "e33ad7cbb371544b"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "A = torch.tensor([[1, 2], [3, 4]])  # shape = (2, 2)\n",
    "B = torch.tensor([[5, 6], [7, 8]])  # shape = (2, 2)\n",
    "\n",
    "S = torch.stack((A, B), dim=0)  # 在最外面添加一个新维度 → shape = (2, 2, 2)\n",
    "print(S)\n",
    "S = torch.stack((A, B), dim=1)  # 插入第1维 → shape = (2, 2, 2)\n",
    "print(S)\n",
    "S = torch.stack((A, B), dim=2)  # 插入第2维 → shape = (2, 2, 2)\n",
    "print(S)"
   ],
   "id": "3ae1ac55280d699"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "# 切分操作\n",
    "A = torch.arange(10) + 1\n",
    "B = torch.chunk(A, 3, 0)\n",
    "B"
   ],
   "id": "d05ece3cb5c8b356"
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "# 切分操作\n",
    "A = torch.arange(10) + 1\n",
    "B = torch.split(A, 3, 0)\n",
    "B"
   ],
   "id": "e6b4083b682fa47f"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T11:53:25.664652Z",
     "start_time": "2025-06-12T11:53:25.653976Z"
    }
   },
   "cell_type": "code",
   "source": [
    "x = torch.arange(10)\n",
    "torch.chunk(x, 3)\n",
    "# 输出：3个 tensor，形状为 [4], [3], [3]"
   ],
   "id": "b59b47ca78552e94",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([0, 1, 2, 3]), tensor([4, 5, 6, 7]), tensor([8, 9]))"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 36
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T11:58:17.374940Z",
     "start_time": "2025-06-12T11:58:17.367564Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# 索引操作\n",
    "A = torch.arange(16).view(4, 4)\n",
    "torch.index_select(A, 1, torch.tensor([1, 3]))"
   ],
   "id": "66be63810446009e",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 1,  3],\n",
       "        [ 5,  7],\n",
       "        [ 9, 11],\n",
       "        [13, 15]])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 53
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T12:01:09.454904Z",
     "start_time": "2025-06-12T12:01:09.448698Z"
    }
   },
   "cell_type": "code",
   "source": [
    "A = torch.rand(5)\n",
    "torch.masked_select(A, A > 0.3)"
   ],
   "id": "75ba3061d9102d3f",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0.5685, 0.4049, 0.4695])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 61
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-12T12:08:23.692196Z",
     "start_time": "2025-06-12T12:08:23.683368Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# 提取出其中第一行的第一个，第二行的第一、第二个，第三行的最后一个\n",
    "A = torch.tensor([[4, 5, 7], [3, 9, 8], [2, 3, 4]])\n",
    "torch.diagonal(A)"
   ],
   "id": "ae81a2f0efa79d83",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([4, 9, 4])"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 62
  }
 ],
 "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
}