ch02-基本概念/01-张量初体验.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 张量"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 0阶张量"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "mammal = tf.Variable('Elephant', tf.string)\n",
    "ignition = tf.Variable(342, tf.int16)\n",
    "floating = tf.Variable(3.14159265, tf.float64)\n",
    "its_complicated = tf.Variable(12.3 - 4.56j, tf.complex64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tf.Variable 'Variable_1:0' shape=() dtype=string_ref>,\n",
       " <tf.Variable 'Variable_2:0' shape=() dtype=int32_ref>,\n",
       " <tf.Variable 'Variable_3:0' shape=() dtype=float32_ref>,\n",
       " <tf.Variable 'Variable_4:0' shape=() dtype=complex128_ref>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[mammal, ignition, floating, its_complicated]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'Variable_8:0' shape=() dtype=float32_ref>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.Variable(3.14159265)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1阶张量"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "mystr = tf.Variable(['Hello'], tf.string)\n",
    "cool_numbers = tf.Variable([3.14, 4.5], tf.float32)\n",
    "first_prime = tf.Variable([1, 3, 5, 7, 11], tf.int16)\n",
    "its_very_complicated = tf.Variable([12.3-2.3j, 7.8-6.3j], tf.complex64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tf.Variable 'Variable_9:0' shape=(2,) dtype=string_ref>,\n",
       " <tf.Variable 'Variable_10:0' shape=(2,) dtype=float32_ref>,\n",
       " <tf.Variable 'Variable_11:0' shape=(5,) dtype=int32_ref>,\n",
       " <tf.Variable 'Variable_12:0' shape=(2,) dtype=complex128_ref>]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[mystr, cool_numbers, first_prime, its_very_complicated]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'Variable_7:0' shape=(2,) dtype=int32_ref>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.Variable([12, 233])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2阶张量"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'Variable_19:0' shape=(2, 2) dtype=int32_ref>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.Variable([[2,3],[45,33]], tf.int16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'Variable_21:0' shape=(2, 1) dtype=int32_ref>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.Variable([[2],[17]], tf.int16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'Variable_23:0' shape=(2, 2) dtype=bool_ref>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.Variable([[False, True],[True, False]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor 'Rank:0' shape=() dtype=int32>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 计算形状？？阶数\n",
    "tf.rank(tf.Variable([[2],[3]], tf.int16))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4阶张量"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor 'zeros:0' shape=(19, 3, 4, 5) dtype=float32>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.zeros([19,3,4,5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.4rc1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}