diff --git a/Machine Learning/ch3.pandas/pandas_book_readers/book_readers.ipynb b/Machine Learning/ch3.pandas/pandas_book_readers/book_readers.ipynb
new file mode 100644
index 0000000..ab0ea03
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_book_readers/book_readers.ipynb	
@@ -0,0 +1,721 @@
+{
+ "cells": [
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ddU9LZWUEiAd"
+   },
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قشر کتاب‌خوان\n",
+    "</font>\n",
+    "</h1>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در این تمرین برای کمک به قشر کتابخوان می‌خواهیم به تحلیل داده‌های سایت <i>goodreads</i> بپردازیم و اطلاعات جالب و کاربردی‌ای را از آن استخراج کنیم.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "مجموعه‌داده\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "اطلاعات کتاب‌ها در فایل <code>books_db.csv</code>،‌ نویسندگان در فایل <code>authors.csv</code>، ناشران در فایل <code>publisher.csv</code> و زبان‌ها در فایل <code>language.csv</code> قرار گرفته است. به‌کمک پانداز هرکدام از این فایل‌ها را خوانده و در متغیر متناظر آن ذخیره کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>bookID</th>\n",
+       "      <th>title</th>\n",
+       "      <th>average_rating</th>\n",
+       "      <th>isbn</th>\n",
+       "      <th>isbn13</th>\n",
+       "      <th>num_pages</th>\n",
+       "      <th>ratings_count</th>\n",
+       "      <th>text_reviews_count</th>\n",
+       "      <th>publication_date</th>\n",
+       "      <th>author_id</th>\n",
+       "      <th>lang_id</th>\n",
+       "      <th>publisher_id</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1</td>\n",
+       "      <td>Harry Potter and the Half-Blood Prince (Harry ...</td>\n",
+       "      <td>4.57</td>\n",
+       "      <td>439785960</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>652</td>\n",
+       "      <td>2095690</td>\n",
+       "      <td>27591</td>\n",
+       "      <td>9/16/2006</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>2</td>\n",
+       "      <td>Harry Potter and the Order of the Phoenix (Har...</td>\n",
+       "      <td>4.49</td>\n",
+       "      <td>439358078</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>870</td>\n",
+       "      <td>2153167</td>\n",
+       "      <td>29221</td>\n",
+       "      <td>9/1/2004</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>4</td>\n",
+       "      <td>Harry Potter and the Chamber of Secrets (Harry...</td>\n",
+       "      <td>4.42</td>\n",
+       "      <td>439554896</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>352</td>\n",
+       "      <td>6333</td>\n",
+       "      <td>244</td>\n",
+       "      <td>11/1/2003</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>5</td>\n",
+       "      <td>Harry Potter and the Prisoner of Azkaban (Harr...</td>\n",
+       "      <td>4.56</td>\n",
+       "      <td>043965548X</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>435</td>\n",
+       "      <td>2339585</td>\n",
+       "      <td>36325</td>\n",
+       "      <td>5/1/2004</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>8</td>\n",
+       "      <td>Harry Potter Boxed Set  Books 1-5 (Harry Potte...</td>\n",
+       "      <td>4.78</td>\n",
+       "      <td>439682584</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>2690</td>\n",
+       "      <td>41428</td>\n",
+       "      <td>164</td>\n",
+       "      <td>9/13/2004</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   bookID                                              title  average_rating  \\\n",
+       "0       1  Harry Potter and the Half-Blood Prince (Harry ...            4.57   \n",
+       "1       2  Harry Potter and the Order of the Phoenix (Har...            4.49   \n",
+       "2       4  Harry Potter and the Chamber of Secrets (Harry...            4.42   \n",
+       "3       5  Harry Potter and the Prisoner of Azkaban (Harr...            4.56   \n",
+       "4       8  Harry Potter Boxed Set  Books 1-5 (Harry Potte...            4.78   \n",
+       "\n",
+       "         isbn        isbn13  num_pages  ratings_count  text_reviews_count  \\\n",
+       "0   439785960  9.780440e+12        652        2095690               27591   \n",
+       "1   439358078  9.780440e+12        870        2153167               29221   \n",
+       "2   439554896  9.780440e+12        352           6333                 244   \n",
+       "3  043965548X  9.780440e+12        435        2339585               36325   \n",
+       "4   439682584  9.780440e+12       2690          41428                 164   \n",
+       "\n",
+       "  publication_date  author_id  lang_id  publisher_id  \n",
+       "0        9/16/2006          0        0             0  \n",
+       "1         9/1/2004          0        0             0  \n",
+       "2        11/1/2003          0        0             1  \n",
+       "3         5/1/2004          0        0             0  \n",
+       "4        9/13/2004          0        0             1  "
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "\n",
+    "df = pd.read_csv(\"./books_db.csv\")\n",
+    "authors = pd.read_csv(\"./authors.csv\")\n",
+    "language = pd.read_csv(\"./language.csv\")\n",
+    "publisher = pd.read_csv(\"./publisher.csv\")\n",
+    "\n",
+    "df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت اول\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "همان‌طور که مشاهده می‌کنید در مجموعه‌داده‌ی اصلی (کتاب‌ها)، مقادیر سه ستون <code>lang_id</code>، <code>author_id</code> و <code>publisher_id</code> تنها به‌صورت آیدی (<code>id</code>) هستند. این آیدی‌ها با شماره‌ی نمایه‌ها در سایر دیتافریم‌ها متناظر هستند. در این قسمت قصد داریم تا طبق این نمایه‌ها،‌ اطلاعات کامل‌شان را از دیتافریم مربوطه استخراج کرده و به دیتافریم <code>df</code> اضافه کنیم. بنابراین سه دیتافریم <code>authors</code>، <code>language</code> و <code>publisher</code> را به‌شکل مناسبی با دیتافریم <code>df</code> ادغام (<code>merge</code>) کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>bookID</th>\n",
+       "      <th>title</th>\n",
+       "      <th>average_rating</th>\n",
+       "      <th>isbn</th>\n",
+       "      <th>isbn13</th>\n",
+       "      <th>num_pages</th>\n",
+       "      <th>ratings_count</th>\n",
+       "      <th>text_reviews_count</th>\n",
+       "      <th>publication_date</th>\n",
+       "      <th>author_id</th>\n",
+       "      <th>lang_id</th>\n",
+       "      <th>publisher_id</th>\n",
+       "      <th>name_x</th>\n",
+       "      <th>code</th>\n",
+       "      <th>name_y</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1</td>\n",
+       "      <td>Harry Potter and the Half-Blood Prince (Harry ...</td>\n",
+       "      <td>4.57</td>\n",
+       "      <td>439785960</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>652</td>\n",
+       "      <td>2095690</td>\n",
+       "      <td>27591</td>\n",
+       "      <td>9/16/2006</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>J.K. Rowling</td>\n",
+       "      <td>eng</td>\n",
+       "      <td>Scholastic Inc.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>2</td>\n",
+       "      <td>Harry Potter and the Order of the Phoenix (Har...</td>\n",
+       "      <td>4.49</td>\n",
+       "      <td>439358078</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>870</td>\n",
+       "      <td>2153167</td>\n",
+       "      <td>29221</td>\n",
+       "      <td>9/1/2004</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>J.K. Rowling</td>\n",
+       "      <td>eng</td>\n",
+       "      <td>Scholastic Inc.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>5</td>\n",
+       "      <td>Harry Potter and the Prisoner of Azkaban (Harr...</td>\n",
+       "      <td>4.56</td>\n",
+       "      <td>043965548X</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>435</td>\n",
+       "      <td>2339585</td>\n",
+       "      <td>36325</td>\n",
+       "      <td>5/1/2004</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>J.K. Rowling</td>\n",
+       "      <td>eng</td>\n",
+       "      <td>Scholastic Inc.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2837</th>\n",
+       "      <td>10546</td>\n",
+       "      <td>The Long-Lost Map (Ulysses Moore #2)</td>\n",
+       "      <td>4.00</td>\n",
+       "      <td>043977439X</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>272</td>\n",
+       "      <td>1693</td>\n",
+       "      <td>76</td>\n",
+       "      <td>7/1/2006</td>\n",
+       "      <td>1280</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Pierdomenico Baccalario</td>\n",
+       "      <td>eng</td>\n",
+       "      <td>Scholastic Inc.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6190</th>\n",
+       "      <td>23300</td>\n",
+       "      <td>The Door to Time (Ulysses Moore  #1)</td>\n",
+       "      <td>3.93</td>\n",
+       "      <td>439774381</td>\n",
+       "      <td>9.780440e+12</td>\n",
+       "      <td>239</td>\n",
+       "      <td>3099</td>\n",
+       "      <td>241</td>\n",
+       "      <td>1/1/2006</td>\n",
+       "      <td>1280</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Pierdomenico Baccalario</td>\n",
+       "      <td>eng</td>\n",
+       "      <td>Scholastic Inc.</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      bookID                                              title  \\\n",
+       "0          1  Harry Potter and the Half-Blood Prince (Harry ...   \n",
+       "1          2  Harry Potter and the Order of the Phoenix (Har...   \n",
+       "3          5  Harry Potter and the Prisoner of Azkaban (Harr...   \n",
+       "2837   10546               The Long-Lost Map (Ulysses Moore #2)   \n",
+       "6190   23300               The Door to Time (Ulysses Moore  #1)   \n",
+       "\n",
+       "      average_rating        isbn        isbn13  num_pages  ratings_count  \\\n",
+       "0               4.57   439785960  9.780440e+12        652        2095690   \n",
+       "1               4.49   439358078  9.780440e+12        870        2153167   \n",
+       "3               4.56  043965548X  9.780440e+12        435        2339585   \n",
+       "2837            4.00  043977439X  9.780440e+12        272           1693   \n",
+       "6190            3.93   439774381  9.780440e+12        239           3099   \n",
+       "\n",
+       "      text_reviews_count publication_date  author_id  lang_id  publisher_id  \\\n",
+       "0                  27591        9/16/2006          0        0             0   \n",
+       "1                  29221         9/1/2004          0        0             0   \n",
+       "3                  36325         5/1/2004          0        0             0   \n",
+       "2837                  76         7/1/2006       1280        0             0   \n",
+       "6190                 241         1/1/2006       1280        0             0   \n",
+       "\n",
+       "                       name_x code           name_y  \n",
+       "0                J.K. Rowling  eng  Scholastic Inc.  \n",
+       "1                J.K. Rowling  eng  Scholastic Inc.  \n",
+       "3                J.K. Rowling  eng  Scholastic Inc.  \n",
+       "2837  Pierdomenico Baccalario  eng  Scholastic Inc.  \n",
+       "6190  Pierdomenico Baccalario  eng  Scholastic Inc.  "
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df = pd.merge(df,authors,left_on=\"author_id\",right_index=True)\n",
+    "df = pd.merge(df,language,left_on=\"lang_id\",right_index=True)\n",
+    "df = pd.merge(df,publisher,left_on=\"publisher_id\",right_index=True)\n",
+    "df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "حال که تمام مقادیر اضافه شده‌اند، نیازی به سه ستون <code>id</code> که به‌عنوان کلید استفاده کردیم نداریم. پس آن‌ها را با استفاده از دستور <code>drop</code> حذف می‌کنیم.\n",
+    "        در درسنامه‌های بعدی نحوه‌ی کار با این دستور را یاد خواهید گرفت.    \n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df.drop(columns = ['author_id', 'lang_id', 'publisher_id'], inplace = True)"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت دوم\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "حال نام ستون‌های دیتافریم را به‌گونه‌ای تغییر دهید که به‌ترتیب از چپ به راست برابر نام‌های زیر باشد:\n",
+    "<ul>\n",
+    "<li><code>bookID</code></li>\n",
+    "<li><code>title</code></li>\n",
+    "<li><code>average_rating</code></li>\n",
+    "<li><code>isbn</code></li>\n",
+    "<li><code>isbn13</code></li>\n",
+    "<li><code>num_pages</code></li>\n",
+    "<li><code>ratings_count</code></li>\n",
+    "<li><code>text_reviews_count</code></li>\n",
+    "<li><code>publication_date</code></li>\n",
+    "<li><code>author</code></li>\n",
+    "<li><code>lang</code></li>\n",
+    "<li><code>publisher</code></li>\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df.columns=[\"bookID\",\"title\",\"average_rating\",\"isbn\",\"isbn13\",\"num_pages\",\"ratings_count\",\"text_reviews_count\",\"publication_date\",\"author\",\"lang\",\"publisher\"]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت سوم\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "اکنون میانگین امتیاز کتاب‌های هر انتشاراتی را محاسبه کرده و حاصل را در متغیر <code>publisher_rating</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "publisher\n",
+       "18-Oct                           3.650000\n",
+       "1st Book Library                 3.580000\n",
+       "1st World Library                4.180000\n",
+       "A & C Black (Childrens books)    4.400000\n",
+       "A Harvest Book/Harcourt  Inc.    4.070000\n",
+       "                                   ...   \n",
+       "聯經出版事業股份有限公司                     4.400000\n",
+       "英文漢聲出版股份有限公司                     4.220000\n",
+       "角川書店 (Kadokawa Shoten)           3.815000\n",
+       "講談社                              4.283333\n",
+       "集英社                              4.237500\n",
+       "Name: average_rating, Length: 2290, dtype: float64"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "publisher_rating = df.groupby(\"publisher\").average_rating.mean()\n",
+    "publisher_rating"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت چهارم\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در این قسمت ابتدا ستونی با نام <code>num_pages_grp</code> به دیتافریم اضافه کنید که در آن تعداد صفحات کتاب را به‌صورت بازه‌های زیر نشان دهد:\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<center>\n",
+    "<div dir=rtl style=\"direction: rtl;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "\n",
+    "| <b>مقدار</b> | <b>توضیح</b> |\n",
+    "| :---: | :---: | \n",
+    "| <code dir=ltr><250</code> | کمتر از ۲۵۰ صفحه |\n",
+    "| <code dir=ltr>250-499</code> | از ۲۵۰ تا ۴۹۹ صفحه |\n",
+    "| <code dir=ltr>500-749</code> | از ۵۰۰ تا ۷۴۹ صفحه |\n",
+    "| <code dir=ltr>750-999</code> | از ۷۵۰ تا ۹۹۹ صفحه |\n",
+    "| <code dir=ltr>>=1000</code> | بیشتر یا مساوی ۱۰۰۰ صفحه |\n",
+    "\n",
+    "</font>\n",
+    "</div>\n",
+    "</center>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:green\"><b>راهنمایی :</b></span>\n",
+    " برای این کار می‌توانید از تابع <code>np.select</code> استفاده کنید. \n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['num_pages_grp'] = np.select([df.num_pages>=1000,df.num_pages>=750,df.num_pages>=500,df.num_pages>=250,df.num_pages<250],[\">=1000\",\"750-999\",\"500-749\",\"250-499\",\"<250\"])"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "حال می‌خواهیم بدانیم هر نویسنده چند کتاب در هرکدام از این بازه‌ها نوشته است. یعنی حاصل باید چیزی شبیه به جدول زیر باشد:\n",
+    "\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "raw",
+   "metadata": {},
+   "source": [
+    "A.B. Yehoshua          250-499          1\n",
+    "                       500-749          1\n",
+    "A.D.P. Briggs          250-499          1\n",
+    "A.E. Cunningham        <250             1\n",
+    "A.J. Jacobs            250-499          1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "author                 num_pages_grp\n",
+       "A.B. Yehoshua          250-499          1\n",
+       "                       500-749          1\n",
+       "A.D.P. Briggs          250-499          1\n",
+       "A.E. Cunningham        <250             1\n",
+       "A.J. Jacobs            250-499          1\n",
+       "                                       ..\n",
+       "bell hooks             <250             5\n",
+       "Åsne Seierstad         250-499          3\n",
+       "Émile Zola             250-499          6\n",
+       "                       500-749          3\n",
+       "Éric-Emmanuel Schmitt  <250             1\n",
+       "Length: 5547, dtype: int64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "author_page = df.groupby([\"author\",\"num_pages_grp\"]).size()\n",
+    "author_page"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl; text-align: justify; line-height:200%; font-family:vazir; font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برای ساخته‌شدن فایل <code>result.zip</code> سلول زیر را اجرا کنید. توجه داشته باشید که پیش از اجرای سلول زیر تغییرات اعمال شده در نت‌بوک را ذخیره کرده باشید (<code>ctrl+s</code>) تا در صورت نیاز به پشتیبانی امکان بررسی کد شما وجود داشته باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['books_result.csv', 'publisher_rating.csv', 'author_page.csv', 'book_readers.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zlib\n",
+    "import zipfile\n",
+    "\n",
+    "df.to_csv('books_result.csv', index=False)\n",
+    "publisher_rating.to_csv('publisher_rating.csv', index=True)\n",
+    "author_page.to_csv(\"author_page.csv\", index=True)\n",
+    "\n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"books_result.csv\", \"publisher_rating.csv\", \"author_page.csv\", \"book_readers.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "YWBTRJvqdG_u"
+   },
+   "source": [
+    "\n",
+    "<br />\n",
+    "<div dir=rtl id=\"start\" style=\"direction:rtl;line-height:200%;\">\n",
+    "\t<font face=\"vazir\" size\n",
+    "        سلاموری ارسال کنید.\n",
+    "  </p>\n",
+    "</div>"
+   ]
+  }
+ ],
+ "metadata": {
+  "colab": {
+   "collapsed_sections": [],
+   "name": "exercise1.ipynb",
+   "provenance": []
+  },
+  "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Machine Learning/ch3.pandas/pandas_diamond/diamond.ipynb b/Machine Learning/ch3.pandas/pandas_diamond/diamond.ipynb
new file mode 100644
index 0000000..60b1990
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_diamond/diamond.ipynb	
@@ -0,0 +1,410 @@
+{
+ "cells": [
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "الماس\n",
+    "\n",
+    "</font>\n",
+    "</h1>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در این مسئله از داده‌های یک فروشگاه الماس که شامل اطلاعات مختلف الماس‌ها (وزن به قیراط،‌ طول، عرض، ارتفاع و غیره) است استفاده خواهیم کرد و ساخت دیتاسری و دیتافریم پانداز را با همدیگر تمرین خواهیم کرد.\n",
+    "\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت اول:\n",
+    "برش ایده‌آل\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "برای بررسی کیفیت برش الماس‌ها از یک کارشناس کمک خواستیم و او کیفیت هر الماس را در یک لیست نوشته و لیست نهایی زیر را به ما تحویل داده است:\n",
+    "  \n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cut = ['Ideal', 'Premium', 'Premium', 'Premium', 'Premium', 'Very Good',\n",
+    "       'Premium', 'Premium', 'Ideal', 'Premium', 'Good', 'Good', 'Premium', \n",
+    "       'Very Good', 'Very Good', 'Good', 'Good','Very Good', 'Good', 'Good', \n",
+    "       'Premium', 'Premium', 'Good', 'Good']"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "با استفاده از این اطلاعات، یک دیتاسری به نام <code>cut_series</code> بسازید. \n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "cut_series = pd.Series(cut)"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در دیتاسری <code>cut_series</code>، کیفیت برش هر الماس ذخیره شده است. حال تعداد هر کیفیت برش را به‌شکل <b>صعودی</b> در یک دیتاسری جدید به نام <code>cut_sorted</code> ذخیره کنید. یعنی عنوان و تعداد کم‌ تعداد‌ترین نوع برش، باید در ابتدای این دیتاسری قرار بگیرد.\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:green\"><b>راهنمایی:</b></span>\n",
+    "برای ساخت این دیتاسری، باید از یکی از توابع معرفی‌شده در درسنامه‌ی «بررسی دیتافریم» استفاده کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Ideal         2\n",
+       "Very Good     4\n",
+       "Good          8\n",
+       "Premium      10\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "cut_sorted = pd.value_counts(cut_series,True,True)\n",
+    "cut_sorted"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت دوم:\n",
+    "قیراط الماس‌ها\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: right;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    " با استفاده از ترازو الماس‌ها را وزن کرده و حاصل را در لیست زیر قرار داده‌ایم.\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "carat = [1.07, 1.19, 1.19, 1.2, 1.2, 1.25, 1.2, 1.2, 1.2,\n",
+    "         1.14, 1.01, 1.01, 1.01, 1.02, 1.01, 1.01, 1.01, \n",
+    "         1.01, 1.01, 1.01, 1.01, 1.01, 1.0, 1.0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "با استفاده از این اطلاعات، یک دیتاسری به نام <code>carat_series</code> بسازید.\n",
+    "\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "carat_series = pd.Series(carat)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در فصل نامپای با صدک‌ها آشنا شدید. صدک هفتاد و پنجم این دیتاسری را محاسبه کنید و در متغیر <code>carat_75</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "carat_75 = np.percentile( np.array(carat_series.values),75)\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت سوم:\n",
+    "ابعاد الماس\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در این مرحله از یک کارشناس دیگر خواستیم که الماس‌ها را اندازه‌گیری کند و او یک آرایه‌ی نامپای به ما تحویل داده است.\n",
+    "<br>\n",
+    "یکی از راه‌های ذخیره‌سازی آرایه‌ی نامپای، استفاده از فرمت ذخیره‌سازی <code>npz</code> است. در این فایل می‌توان هر تعداد آرایه‌ی نامپای را ذخیره کرد و بعد با استفاده از نام آرایه‌ها، آرایه‌ی مورد نظر را از فایل بیرون کشید. برای آشنایی بیشتر با این فرمت می‌توانید\n",
+    "<a href=\"https://stackoverflow.com/questions/54238670/what-is-the-advantage-of-saving-npz-files-instead-of-npy-in-python-regard\"> \n",
+    " این لینک\n",
+    "</a>\n",
+    " را مطالعه کنید. \n",
+    "<br>\n",
+    "چون نحوه‌ی استفاده این فایل را برای شما توضیح نداده‌ایم، کد خواندن این فایل و سپس بیرون کشیدن آرایه‌ی <code>sizes</code> از آن را در سلول زیر برای شما نوشته‌ایم. آرایه‌ی حاصل در متغیری به نام <code>sizes_array</code> ذخیره می‌شود. \n",
+    " \n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[6.46, 6.42, 4.09],\n",
+       "       [6.87, 6.84, 4.17],\n",
+       "       [6.73, 6.66, 4.2],\n",
+       "       [6.77, 6.72, 4.22],\n",
+       "       [6.77, 6.72, 4.22],\n",
+       "       [6.86, 6.9, 4.27],\n",
+       "       [6.93, 6.8, 4.24],\n",
+       "       [6.86, 6.74, 4.21],\n",
+       "       [6.73, 6.68, 4.24],\n",
+       "       [6.76, 6.72, 4.13],\n",
+       "       [6.19, 6.23, 4.05],\n",
+       "       [6.48, 6.5, 3.79],\n",
+       "       [6.46, 6.41, 3.94],\n",
+       "       [6.34, 6.41, 4.06],\n",
+       "       [6.35, 6.3, 4.0],\n",
+       "       [6.35, 6.3, 4.02],\n",
+       "       [6.37, 6.31, 4.05],\n",
+       "       [6.39, 6.34, 4.02],\n",
+       "       [6.4, 6.32, 4.05],\n",
+       "       [6.58, 6.53, 3.77],\n",
+       "       [6.46, 6.37, 3.92],\n",
+       "       [6.44, 6.38, 3.95],\n",
+       "       [6.33, 6.38, 4.01],\n",
+       "       [6.27, 6.31, 3.99]], dtype=object)"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "\n",
+    "sizes = np.load('sizes.npz', allow_pickle=True)\n",
+    "sizes_array = sizes['sizes']\n",
+    "sizes_array"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: right;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    " آرایه‌ی <code>sizes_array</code> دارای سه ستون است. در ستون اول طول، در ستون دوم عرض و در ستون سوم ارتفاع الماس‌ها ذخیره شده است. \n",
+    " <br>\n",
+    "با استفاده از این آرایه، یک دیتافریم به نام <code>sizes_df</code> بسازید و نام ستون‌های آن را برابر <code>x</code> , <code>y</code> , <code>z</code> قرار دهید. (اولین ستون <code>x</code> نام دارد).\n",
+    " </font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sizes_df = pd.DataFrame(sizes_array,columns=['x','y','z'])"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت چهارم:\n",
+    "نمایه‌سازی\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برخلاف دنیای برنامه‌نویسان در کسب‌و‌کار الماس مرسوم نیست که شماره‌گذاری از <code>0</code> شروع شود. بنابراین نمایه‌ها را طوری تغییر دهید که از <code>1</code> شروع شوند.  \n",
+    "دیتافریم جدید را در <code>sizes_new_index</code> ذخیره کنید.\n",
+    "<br>\n",
+    "\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sizes_new_index = sizes_df.copy()\n",
+    "sizes_new_index = sizes_new_index.set_index(pd.RangeIndex(start=1, stop=25, step=1))"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl; text-align: justify; line-height:200%; font-family:vazir; font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برای ساخته‌شدن فایل <code>result.zip</code> سلول زیر را اجرا کنید. توجه داشته باشید که پیش از اجرای سلول زیر تغییرات اعمال شده در نت‌بوک را ذخیره کرده باشید (<code>ctrl+s</code>) تا در صورت نیاز به پشتیبانی امکان بررسی کد شما وجود داشته باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['cut_series.csv', 'cut_sorted.csv', 'carat_series.csv', 'carat_75', 'sizes_df.csv', 'sizes_new_index.csv', 'diamond.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zipfile\n",
+    "import joblib\n",
+    "\n",
+    "carat_series.to_csv(\"carat_series.csv\")\n",
+    "cut_sorted.to_csv(\"cut_sorted.csv\")\n",
+    "cut_sorted.to_csv(\"cut_series.csv\")\n",
+    "joblib.dump(carat_75,\"carat_75\")\n",
+    "sizes_df.to_csv(\"sizes_df.csv\",index = True)\n",
+    "sizes_df.to_csv(\"sizes_new_index.csv\",index = True)\n",
+    "                \n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"cut_series.csv\", \"cut_sorted.csv\", \"carat_series.csv\",\n",
+    "              \"carat_75\", \"sizes_df.csv\", \"sizes_new_index.csv\", \"diamond.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  }
+ ],
+ "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/Machine Learning/ch3.pandas/pandas_hindustan/hindustan.ipynb b/Machine Learning/ch3.pandas/pandas_hindustan/hindustan.ipynb
new file mode 100644
index 0000000..4116b8b
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_hindustan/hindustan.ipynb	
@@ -0,0 +1,224 @@
+{
+ "cells": [
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "ff54573a",
+   "metadata": {},
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "هندوستان\n",
+    "</font>\n",
+    "</h1>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در این تمرین به‌سراغ یک مجموعه‌داده از غذاهای هندوستان می‌رویم و بررسی می‌کنیم که هندوها تا چه حد ذائقه‌ی تندی دارند. همچنین چند مورد از دیرپزترین غذاهای‌شان را می‌یابیم.\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "f7cd7632",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "مجموعه‌داده\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "داده‌های مرتبط با این تمرین در فایل <code>food.csv</code> قرار گرفته و شامل ستون‌های زیر است:\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<center>\n",
+    "<div dir=rtl style=\"direction: rtl;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "\n",
+    "    \n",
+    "| ستون | توضیح |\n",
+    "| :---: | :---: |\n",
+    "| <code>name</code> | نام غذا |\n",
+    "| <code>ingredients</code>| مواد تشکیل‌دهنده‌ی غذا|\n",
+    "| <code>diet</code>|دارای دو مقدار <code>vegetarian</code> و <code>non vegetarian</code> است که نشان می‌دهد آیا غذا مناسب گیاه‌خواران هست یا خیر.|\n",
+    "| <code>prep_time</code>| زمانی که طول می‌کشد تا غذا آماده شود. این زمان شامل زمان پخت نمی‌شود. مثلاً زمان پوست کندن یا شستن مواد اولیه در این ستون لحاظ شده‌اند.|\n",
+    "| <code>cook_time</code>| زمان پخت یک غذا| \n",
+    "| <code>flavor_profile</code>| مزه‌ی غذا را نشان می‌دهد. دارای ۴ مقدار <code>spicy</code> برای غذاهای تند، <code>sweet</code> برای غذاهای شیرین، <code>bitter</code> برای غذاهای تلخ و <code>sour</code> برای غذاهای ترش است.|\n",
+    "| <code>course</code>| وعده‌ی اصلی، میان‌وعده، دسر یا پیش‌غذا بودن یک غذا را نشان می‌دهد.|\n",
+    "| <code>state</code>| ایالتی که زادگاه غذاست یا حداقل این غذا در آن‌جا معروف است.|\n",
+    "| <code>region</code>| منطقه‌ای که ایالت (<code>state</code>) در آن واقع شده است.|\n",
+    "\n",
+    "\n",
+    "</font>\n",
+    "</div>\n",
+    "</center>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "1d2bc674",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd \n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "cf5dff88",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(\"./food.csv\")"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "067c1145",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت اول\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "بسیار شنیده‌ایم که هندوها عاشق غذاهای تند هستند! حال می‌خواهیم با توجه به مجموعه‌داده‌ی موجود صحت این گزاره را بررسی کرده و ببینیم که چه نسبتی از کل غذاهای این مجموعه، دارای مزه‌ی تندی هستند؟\n",
+    "برای این کار کافیست حاصل تقسیم تعداد غذاهای تند مزه را بر تعداد کل غذاها حساب کنید و در متغیر <code>spicy_ratio</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "d6461854",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spicy_ratio = df[df.flavor_profile==\"spicy\"].shape[0]/df.shape[0]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "5137d5bc",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت دوم\n",
+    "</font>\n",
+    "</h2>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "اکنون می‌خواهیم بدانیم کدام غذاها از لحظه‌ی شروع آماده‌سازی مخلفات تا لحظه‌ی آماده‌ی سرو شدن، بیشترین زمان را مصرف می‌کنند.\n",
+    "برای این‌کار باید جمع مدت زمان آماده‌سازی و زمان لازم برای طبخ غذا را در نظر گرفته و سپس <b>۱۰ غذایی</b> که بیشترین زمان را مصرف می‌کنند <b>به‌ترتیب نزولی</b> در قالب یک دیتافریم در متغیر <code>df_most_time_consuming</code> ذخیره کنید. \n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:#0099cc\"><b>نکته:</b></span>\n",
+    "    فرمت و ترتیب ستون‌های دیتافریم <code>df_most_time_consuming</code> باید دقیقاً شبیه به دیتافریم <code>df</code> باشد. یعنی این دیتافریم دارای ۱۰ سطر و ۹ ستون خواهد بود.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "ae1f814c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df[\"consuming_time\"]=df.apply(lambda row : row['prep_time'] + row['cook_time'], axis=1 )\n",
+    "df_most_time_consuming = df.sort_values(by='consuming_time' , ascending=False)[0:10]\n",
+    "df.drop(\"consuming_time\" , axis=1,inplace=True)\n",
+    "df_most_time_consuming.drop(\"consuming_time\" , axis=1,inplace=True)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7341ff2d",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "5f63b15c",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['spicy_ratio.npy', 'df_most_time_consuming.csv', 'hindustan.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zipfile\n",
+    "\n",
+    "np.save(\"spicy_ratio\", np.array(spicy_ratio))\n",
+    "df_most_time_consuming.to_csv('df_most_time_consuming.csv', index=False)\n",
+    "\n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"spicy_ratio.npy\", \"df_most_time_consuming.csv\", \"hindustan.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  }
+ ],
+ "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Machine Learning/ch3.pandas/pandas_mad_cyclists/mad_cyclists.ipynb b/Machine Learning/ch3.pandas/pandas_mad_cyclists/mad_cyclists.ipynb
new file mode 100644
index 0000000..6c06e73
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_mad_cyclists/mad_cyclists.ipynb	
@@ -0,0 +1,1141 @@
+{
+ "cells": [
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "دوچرخه‌سواران بی‌طاقت\n",
+    "</font>\n",
+    "</h1>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "دوچرخه‌سوران خیلی روی دما حساس هستند. دمایی که دوچرخه‌سواران احساس می‌کنند معمولاً با سرعت باد و رطوبت هوا رابطه‌ی مستقیم دارد. در این تمرین می‌خواهیم با توجه به مجموعه‌داده‌ی دمای روزهای مختلف، به یک کسب‌و‌کار اجاره‌ی دوچرخه کمک کنیم تا در روزهای مختلف و در دماهای مختلف دوچرخه‌های بیشتری اجاره دهد.   \n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "مجموعه‌داده\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "با اجرای سلول زیر می‌توانید داده‌های این تمرین را به‌شکل یک دیتافریم بخوانید. این مجموعه‌داده شامل ستون‌های زیر است:\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "\n",
+    "<center>\n",
+    "<div dir=rtl style=\"direction: rtl;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "\n",
+    "| <b>شماره گروه</b> | <b>توضیحات</b> |\n",
+    "| :---: | :---: | \n",
+    "| <code>cnt</code> | تعداد دوچرخه‌های اجاره داده‌شده در روز |\n",
+    "| <code>t1</code> | دمای واقعی اندازه‌گیری‌شده در آن روز |\n",
+    "| <code>t2</code> | میانگین دمایی که دوچرخه‌سواران احساس کرده‌اند |\n",
+    "| <code>humidity</code> | رطوبت هوا در آن روز |\n",
+    "| <code>wind_speed</code> | سرعت باد در آن روز |\n",
+    "|‌ <code>is_weekend</code> | آیا آن روز،‌ روز غیرکاری (آخر هفته) هست یا خیر |\n",
+    "| <code>season</code> | چندمین فصل سال |\n",
+    "\n",
+    "</font>\n",
+    "</div>\n",
+    "</center>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>cnt</th>\n",
+       "      <th>t1</th>\n",
+       "      <th>t2</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "      <th>is_weekend</th>\n",
+       "      <th>season</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>182</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>138</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>134</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>72</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>47</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>6.5</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   cnt   t1   t2  humidity  wind_speed  is_weekend  season\n",
+       "0  182  3.0  2.0      93.0         6.0         1.0     3.0\n",
+       "1  138  3.0  2.5      93.0         5.0         1.0     3.0\n",
+       "2  134  2.5  2.5      96.5         0.0         1.0     3.0\n",
+       "3   72  2.0  2.0     100.0         0.0         1.0     3.0\n",
+       "4   47  2.0  0.0      93.0         6.5         1.0     3.0"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "from numpy import nan as NA\n",
+    "\n",
+    "df = pd.read_csv('bikes_borrowed.csv')\n",
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    نحوه‌ی کار با تابع <code>read_csv</code> را در درسنامه‌های آینده یاد می‌گیرید؛ اما به استفاده از <code>head</code> در این کد دقت کنید؛ همان‌طور که گفتیم، این تابع به‌طور پیش‌فرض، ۵ داده‌ی بالای دیتافریم را به ما نشان می‌دهد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت اول\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    همان‌طور که می‌بینید نام ستون‌های <code>t1</code> و <code>t2</code> خیلی کلی است و بیانگر معنای مقدار آن‌ها نیست، به همین دلیل خوب است تا نام آن‌ها را تغییر دهیم.\n",
+    "<br>\n",
+    "    در سلول زیر، نام ستون <code>t1</code> را به <code>t_real</code> و نام ستون <code>t2</code> را به <code>t_feels_like</code> تغییر دهید.\n",
+    "<br>\n",
+    "    برای تغییر نام نمایه‌ها یا ستون‌ها می‌توانیم همچون کد زیر از تابع <code>rename</code> استفاده کنیم:\n",
+    "<br>\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "`df.rename(columns={\"col1\": \"new_col1\", \"col2\": \"new_col2\"} , inplace = True)`\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>cnt</th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "      <th>is_weekend</th>\n",
+       "      <th>season</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>182</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>138</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>134</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>72</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>47</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>6.5</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   cnt  t_real  t_feels_like  humidity  wind_speed  is_weekend  season\n",
+       "0  182     3.0           2.0      93.0         6.0         1.0     3.0\n",
+       "1  138     3.0           2.5      93.0         5.0         1.0     3.0\n",
+       "2  134     2.5           2.5      96.5         0.0         1.0     3.0\n",
+       "3   72     2.0           2.0     100.0         0.0         1.0     3.0\n",
+       "4   47     2.0           0.0      93.0         6.5         1.0     3.0"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.rename(columns={\"t1\":\"t_real\",\"t2\":\"t_feels_like\"},inplace=True)\n",
+    "\n",
+    "df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    به پارامتر <code>inplace = True</code> دقت کنید؛ وجود این پارامتر باعث می‌شود که تابع <code>rename</code>، به‌جای آن‌که یک کپی از دیتافریم تهیه کرده و بعد از اعمال تغییرات آن کپی را خروجی دهد، تغییرات را مستقیماً روی دیتافریم اصلی اعمال کند.\n",
+    "<br>\n",
+    "برای درک بهتر این موضوع می‌توانید یک‌بار این پارامتر را حذف کرده و سپس مجدد <code dir=ltr>df.head()</code> را خروجی بگیرید و بررسی کنید که آیا نام ستون‌ها تغییر کرده است یا خیر؟\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت دوم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "از آن‌جایی که اغلب روز‌هایی که دوچرخه‌سواران از هوا گلایه دارند روز‌های بادی است، تصمیم گرفتیم که محاسبات را فقط برای روز‌هایی که سرعت باد بیشتر از ۱۰ است انجام دهیم.\n",
+    "<br>\n",
+    "    بنابراین در دیتافریم <code>windy_days_df</code>، تنها داده‌هایی را ذخیره کنید که سرعت باد آن‌ها بیشتر از ۱۰ باشد (خود ۱۰ را در نظر نگیرید).\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>cnt</th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "      <th>is_weekend</th>\n",
+       "      <th>season</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>528</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>727</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>862</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>13.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>916</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>15.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>869</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>11.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>3.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    cnt  t_real  t_feels_like  humidity  wind_speed  is_weekend  season\n",
+       "10  528     3.0          -0.5      93.0        12.0         1.0     3.0\n",
+       "11  727     2.0          -1.5     100.0        12.0         1.0     3.0\n",
+       "12  862     2.0          -1.5      96.5        13.0         1.0     3.0\n",
+       "13  916     3.0          -0.5      87.0        15.0         1.0     3.0\n",
+       "15  869     2.0          -1.5      93.0        11.0         1.0     3.0"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "windy_days_df = df[df[\"wind_speed\"]>10]\n",
+    "\n",
+    "windy_days_df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت سوم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "برای انجام باقی محاسبات فقط ستون‌های مخصوص دما را نیاز داریم؛ پس برای دیتافریم <code>windy_days_df</code>، ستون‌های <code>humidity</code> , <code>t_feels_like</code> , <code>t_real</code> و <code>wind_speed</code> را ذخیره کرده و باقی ستون‌ها را نادیده بگیرید.\n",
+    "\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>12.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>12.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>13.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>15.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>11.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    t_real  t_feels_like  humidity  wind_speed\n",
+       "10     3.0          -0.5      93.0        12.0\n",
+       "11     2.0          -1.5     100.0        12.0\n",
+       "12     2.0          -1.5      96.5        13.0\n",
+       "13     3.0          -0.5      87.0        15.0\n",
+       "15     2.0          -1.5      93.0        11.0"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "windy_days_df = windy_days_df[[\"t_real\",\"t_feels_like\",\"humidity\",\"wind_speed\"]]\n",
+    "\n",
+    "windy_days_df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در مراحل بعدی می‌خواهیم برخی مقادیر دیتافریم را تغییر دهیم پس برای آن‌که دیتافریم اصلی و <code>windy_days_df</code> تغییر نکنند لازم است تا از تابع <code>copy</code> استفاده کنیم.\n",
+    "<br>\n",
+    "به‌طور کلی استفاده از عملگر مساوی برای دیتافریم‌ها در پانداز مانند نامپای عمل می‌کنند و تنها یک اشاره‌گر (pointer) جدید به همان دیتافریم اضافه می‌شود. بنابراین، تغییر دادن هر کدام از دیتافریم‌ها باعث تغییر دیگری نیز می‌شود.\n",
+    "<br>\n",
+    "    استفاده از تابع <code>copy</code> این قابلیت را برای ما ایجاد می‌کند که یک کپی از دیتافریم اولیه دریافت کنیم و تغییر یکی تاثیری روی دیگری نداشته باشد.\n",
+    "<br>\n",
+    "در ادامه بیشتر  این ویژگی را بررسی می‌کنیم اما فعلاً کافیست سلول زیر را اجرا کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "temperature_df = windy_days_df.copy()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت چهارم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "برای انجام تحلیل‌های بهتر نیاز داریم بدانیم که یک روز نسبت به بازده‌ی کلی دما تا چه‌اندازه گرم است. برای این کار به‌ترتیب مراحل زیر را انجام دهید:\n",
+    "<br>\n",
+    "    ۱- در متغیر <code>t_max</code> مقدار ماکزیمم ستون <code>t_real</code> را ذخیره کنید.\n",
+    "<br>\n",
+    "    ۲- در متغیر <code>t_min</code> مقدار مینیمم ستون <code>t_real</code> را ذخیره کنید.\n",
+    "<br>\n",
+    "۳- ستون جدیدی به نام <code>t_percent</code> به دیتافریم اضافه کنید. در این ستون از فرمول زیر برای نرمال‌سازی و پیدا کردن دمای نسبی استفاده کنید.\n",
+    "<br>\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "`((temp - min) / (max - min)) * 100`\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:green\"><b>راهنمایی:</b></span>\n",
+    " تمام عملیات این مرحله را می‌توانید با توابع نامپای انجام دهید. همان‌طور که گفتیم پانداز بر اساس نامپای نوشته شده، پس تمامی عملیات‌های نامپای در این‌جا هم قابل استفاده است.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "      <th>t_percent</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>11.428571</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>13.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>15.0</td>\n",
+       "      <td>11.428571</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>11.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    t_real  t_feels_like  humidity  wind_speed  t_percent\n",
+       "10     3.0          -0.5      93.0        12.0  11.428571\n",
+       "11     2.0          -1.5     100.0        12.0   8.571429\n",
+       "12     2.0          -1.5      96.5        13.0   8.571429\n",
+       "13     3.0          -0.5      87.0        15.0  11.428571\n",
+       "15     2.0          -1.5      93.0        11.0   8.571429"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t_max = np.max(temperature_df[\"t_real\"])\n",
+    "t_min = np.min(temperature_df[\"t_real\"])\n",
+    "temperature_df['t_percent'] = ((temperature_df[\"t_real\"] - t_min) / (t_max - t_min)) * 100\n",
+    "\n",
+    "temperature_df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت پنجم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "حال می‌خواهیم دمایی که احساس می‌شود را طبق اطلاعات موجود پیش‌بینی کرده و مقادیر این ستون (<code>t_feels_like</code>) را جایگزین کنیم.\n",
+    "برای مقداردهی این ستون از فرمول زیر استفاده کنید:\n",
+    "<br>\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "`t_feels_like = t_real + (humidity * t_real)/1000 - (wind_speed)/10 -2`\n",
+    "<br>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: right;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    <span style=\"color:orange\"><b>یادآوری:</b></span>\n",
+    " هر ستون دیتافریم را می‌توانید مانند یک سری جدا کرده و مانند یک آرایه‌ی نامپای با آن رفتار کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "      <th>t_percent</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0.079</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>11.428571</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.000</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.107</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>13.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.239</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>15.0</td>\n",
+       "      <td>11.428571</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-0.914</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>11.0</td>\n",
+       "      <td>8.571429</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    t_real  t_feels_like  humidity  wind_speed  t_percent\n",
+       "10     3.0         0.079      93.0        12.0  11.428571\n",
+       "11     2.0        -1.000     100.0        12.0   8.571429\n",
+       "12     2.0        -1.107      96.5        13.0   8.571429\n",
+       "13     3.0        -0.239      87.0        15.0  11.428571\n",
+       "15     2.0        -0.914      93.0        11.0   8.571429"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "temperature_df['t_feels_like'] = df.t_real + (df.humidity * df.t_real)/1000 - (df.wind_speed)/10 -2\n",
+    "\n",
+    "temperature_df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    همان‌طور که می‌بینید مقدار ستون <code>t_feels_like</code> تغییر کرده و دیگر مانند آن‌چه در دیتافریم اولیه داشتیم نیست. حال بیایید کارکرد تابع ‍<code>copy</code> را بررسی کنیم. اگر تمام مراحل را درست پیش رفته باشید، مقدار ستون <code>t_feels_like</code> در دیتافریم ‍‍<code>windy_days_df</code> نباید تغییر کرده باشد. برای اطمینان از این موضوع با استفاده از تابع <code>head</code> می‌توانیم مقادیر این دو دیتافریم را مقایسه کنیم.\n",
+    "\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>t_real</th>\n",
+       "      <th>t_feels_like</th>\n",
+       "      <th>humidity</th>\n",
+       "      <th>wind_speed</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>12.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>100.0</td>\n",
+       "      <td>12.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>96.5</td>\n",
+       "      <td>13.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>3.0</td>\n",
+       "      <td>-0.5</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>15.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-1.5</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>11.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    t_real  t_feels_like  humidity  wind_speed\n",
+       "10     3.0          -0.5      93.0        12.0\n",
+       "11     2.0          -1.5     100.0        12.0\n",
+       "12     2.0          -1.5      96.5        13.0\n",
+       "13     3.0          -0.5      87.0        15.0\n",
+       "15     2.0          -1.5      93.0        11.0"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "windy_days_df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت ششم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "در نهایت می‌خواهیم دقت این پیش‌بینی را محاسبه کنیم. برای این کار از فرمول زیر استفاده می‌کنیم که با نام <i>Mean Absolute Error</i> یا <i>MAE</i> شناخته می‌شود.\n",
+    "<br>\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "`np.mean(np.abs(temperature_df['t_feels_like'] - windy_days_df['t_feels_like']))`\n",
+    "<br>\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: right;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    این مقدار را محاسبه و در متغیر <code>diffrence</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2.028340505491597\n"
+     ]
+    }
+   ],
+   "source": [
+    "diffrence = np.mean(np.abs(temperature_df['t_feels_like'] - windy_days_df['t_feels_like']))\n",
+    "print(diffrence)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "قسمت هفتم\n",
+    "    </font>\n",
+    "</h2>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    در این تمرین با تابع <code>head(n)</code> کار کردیم اما بیایید به بررسی عمیق‌تر این تابع بپردازیم.\n",
+    "<br>\n",
+    "به‌طور کلی این تابع تنها برای نمایش داده‌ها استفاده نمی‌شود، بلکه اعمال آن روی دیتافریم یک کپی از <code>n</code> داده‌ی اول آن را برمی‌گرداند و می‌توان آن را در دیتافریم جداگانه‌ای ذخیره کرد.\n",
+    "<br>\n",
+    "    در دیتافریم زیر، ۱۰۰ داده‌ی اول دیتافریم <code>temperature_df</code> را ذخیره کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "final_df = temperature_df.head(100)"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl; text-align: justify; line-height:200%; font-family:vazir; font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برای ساخته‌شدن فایل <code>result.zip</code> سلول زیر را اجرا کنید. توجه داشته باشید که پیش از اجرای سلول زیر تغییرات اعمال شده در نت‌بوک را ذخیره کرده باشید (<code>ctrl+s</code>) تا در صورت نیاز به پشتیبانی امکان بررسی کد شما وجود داشته باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['bikes.csv', 'windy_days.csv', 'temperature.csv', 'final.csv', 'answers.npz', 'mad_cyclists.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zlib\n",
+    "import zipfile\n",
+    "\n",
+    "np.savez(\"answers.npz\", diffrence=diffrence, t_min=t_min , t_max=t_max)\n",
+    "df.to_csv('bikes.csv', index=True)\n",
+    "windy_days_df.to_csv('windy_days.csv', index=True)\n",
+    "temperature_df.to_csv(\"temperature.csv\", index=True)\n",
+    "final_df.to_csv(\"final.csv\", index=True)\n",
+    "\n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"bikes.csv\", \"windy_days.csv\", \"temperature.csv\", \n",
+    "              \"final.csv\", \"answers.npz\" , \"mad_cyclists.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  }
+ ],
+ "metadata": {
+  "colab": {
+   "collapsed_sections": [],
+   "name": "exercise1.ipynb",
+   "provenance": []
+  },
+  "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Machine Learning/ch3.pandas/pandas_shakgram/shakhgram.ipynb b/Machine Learning/ch3.pandas/pandas_shakgram/shakhgram.ipynb
new file mode 100644
index 0000000..cac0131
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_shakgram/shakhgram.ipynb	
@@ -0,0 +1,444 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "d64d01af",
+   "metadata": {},
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    شاخگرام\n",
+    "</font>\n",
+    "</h1>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "4e177dbd-7e65-45b6-8221-1dec02f863e2",
+   "metadata": {},
+   "source": [
+    "<div dir=rtl id=\"start\" style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "\t<font face=\"vazir\" size=3>\n",
+    "در درسنامه‌های پیشین با ساختمان‌داده‌های پرکاربرد در پانداز آشنا شدیم. در این تمرین، قصد داریم انتخاب داده از یک دیتافریم را بیشتر تمرین کنیم.\n",
+    "مجموعه‌داده‌ای که برای این تمرین در دست داریم، مربوط به یک شبکه‌ی اجتماعی فرضی به نام شاخگرام است که در آن افراد سعی می‌کنند هرچه بیشتر شاخ شوند!\n",
+    " هدف اصلی این تمرین بررسی داده‌های شاخگرام و نوشتن شرط‌های گوناگون جهت فیلتر کردن شاخ‌های سطوح مختلف است.\n",
+    "    </font>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "3934f0cc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "1792e7ff",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "مجموعه‌داده\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "داده‌های مربوط به این تمرین در فایل <code>shakhgram.csv</code> قرار گرفته و به کمک سلول زیر می‌توانید آن را به شکل یک دیتافریم بخوانید و در متغیر <code>df</code> ذخیره کنید. این دیتافریم شامل ستون‌های زیر است:\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<center>\n",
+    "<div dir=rtl style=\"direction: rtl;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "\n",
+    "| نام ستون | توضیحات |\n",
+    "| :---: | :---: |\n",
+    "| `owner_name` | اسم صاحب پیج |\n",
+    "| `follower` | تعداد دنبال‌کنندگان پیج |\n",
+    "| `following` | تعداد دنبال‌شوندگان پیج |\n",
+    "| `post_number` | تعداد پست‌های پیج |\n",
+    "\n",
+    "</font>\n",
+    "</div>\n",
+    "</center>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "c5312517",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(\"shakhgram.csv\")"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "b562fdc1-2071-47e5-8aae-eb3581aadd99",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    مشاهده‌ی داده‌ها\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "با اجرای سلول زیر، ۵ سطر اول دیتافریم نمایش داده می‌شود تا یک نگاه کلی به داده‌ها داشته باشیم.\n",
+    "پیشنهاد می‌کنیم سعی کنید ستون‌ها و سطرهای مختلفی از دیتافریم را انتخاب و مشاهده کنید تا علاوه بر این‌که با مجموعه‌داده بیشتر آشنا می‌شوید، برای پیاده‌سازی پاسخ تمرین نیز آماده‌تر شوید!\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "856f8d8c",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>owner_name</th>\n",
+       "      <th>follower</th>\n",
+       "      <th>following</th>\n",
+       "      <th>post_number</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Muhammad Schaefer</td>\n",
+       "      <td>108751787.0</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>61.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Ezra Kemp</td>\n",
+       "      <td>172731383.0</td>\n",
+       "      <td>71.0</td>\n",
+       "      <td>43.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Eli Good</td>\n",
+       "      <td>105194939.0</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>44.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Lillianna Pruitt</td>\n",
+       "      <td>14807976.0</td>\n",
+       "      <td>11.0</td>\n",
+       "      <td>88.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Carlie Woodward</td>\n",
+       "      <td>228517005.0</td>\n",
+       "      <td>68.0</td>\n",
+       "      <td>52.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          owner_name     follower  following  post_number\n",
+       "0  Muhammad Schaefer  108751787.0       45.0         61.0\n",
+       "1          Ezra Kemp  172731383.0       71.0         43.0\n",
+       "2           Eli Good  105194939.0       20.0         44.0\n",
+       "3   Lillianna Pruitt   14807976.0       11.0         88.0\n",
+       "4    Carlie Woodward  228517005.0       68.0         52.0"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "aa82e222",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت اول\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    در این شبکه‌ی اجتماعی، ۵ دسته از شاخ‌بودن وجود دارد. برای این‌که عضو هر دسته باشیم، می‌بایست پیج و اسم ما شرایط خاصی داشته باشد.\n",
+    "    <br>\n",
+    "    اولین دسته‌ از شاخ‌های شاخگرام، شاخ‌های برنجی هستند؛ شاخ‌های برنجی،  افرادی هستند که تعداد افرادی که توسط پیج‌شان دنبال می‌شوند (<code>following</code>) <b>از ۱۱ کمتر</b> باشد.\n",
+    "    <br>\n",
+    "    در سلول زیر، بخشی از دیتافریم اصلی را که شرایط شاخ‌های برنجی دارند جدا کرده و در متغیر <code>berenji_df</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "f16be0eb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "berenji_df = df[df.following<11]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "77416747",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت دوم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    در متغیر <code>mesi_df</code>، شاخ‌های مسی را ذخیره کنید؛ شاخ‌های مسی، افرادی هستند که تعداد پست‌های پیج‌شان از تعداد افرادی که آن پیج دنبال‌شان می‌کند (<code>following</code>) بیشتر باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "51f5a12b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "mesi_df = df[df.following<df.post_number]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "5bf37fd8-813b-4882-b8ba-b361fdc62f0e",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت سوم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    در متغیر <code>boronzi_df</code>، شاخ‌های برنزی را ذخیره کنید؛\n",
+    "    شاخ‌های برنزی، افرادی هستند که تعداد دنبال‌کنندگان (<code>follower</code>) پیج‌شان بیشتر از یک میلیون نفر باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "76184e27",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "boronzi_df = df[df.follower>1000000]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "e8ef45c6-0367-4e24-82c0-9ab6b97a34b2",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت چهارم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    در متغیر <code>noghreyee_df</code>، شاخ‌های نقره‌ای را ذخیره کنید؛ شاخ‌های نقره‌ای،‌ افرادی هستند که دو شرط زیر همزمان برای آن‌ها برقرار باشد:\n",
+    "    <ol dir=rtl>\n",
+    "        <li>تعداد دنبال‌کنندگان (<code>follower</code>) پیج‌شان بزرگ‌تر یا مساوی ۱۰ میلیون باشد.</li>\n",
+    "        <li>تعداد دنبال‌شوندگان توسط پیج (<code>following</code>) کم‌تر از ۵۰ باشد.</li>\n",
+    "    </ol>\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "6b8b4f92",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "noghreyee_df = df[(df.follower>=10000000) & (df.following<50)]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "79b40e28-7007-48b5-95b1-bd8c3e5f05ae",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت پنجم\n",
+    "    </font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "        در متغیر <code>talayee_df</code>، شاخ‌های طلایی را ذخیره کنید؛\n",
+    "شاخ‌های طلایی افرادی هستند که حداقل یکی از دو شرط زیر را داشته باشند (یعنی شرط اول یا شرط دوم)\n",
+    "<ol dir=rtl>\n",
+    "        <li>تعداد دنبال‌کنندگان (<code>follower</code>) پیج‌شان بزرگ‌تر از ۱۰ میلیون و تعداد دنبال‌شوندگان (<code>following</code>) آن‌ها کمتر از ۲۰ باشد.</li>\n",
+    "        <li>تعداد پست‌های‌شان کمتر از ۱۵ باشد و حرف <code>R</code> در اسم صاحب پیج وجود داشته باشد (مثل <code dir=ltr>!BooAzaR</code>).</li>\n",
+    "    </ol>\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "099d1ff4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "talayee_df = df[((df.follower>10000000) & (df.following<20) | (df.post_number<15) & (df.owner_name.str.contains('R')))]"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "ad56a791-6354-4597-babf-c9df1c83ca73",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl; text-align: justify; line-height:200%; font-family:vazir; font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برای ساخته‌شدن فایل <code>result.zip</code> سلول زیر را اجرا کنید. توجه داشته باشید که پیش از اجرای سلول زیر تغییرات اعمال شده در نت‌بوک را ذخیره کرده باشید (<code>ctrl+s</code>) تا در صورت نیاز به پشتیبانی امکان بررسی کد شما وجود داشته باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "9b01fa06",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['berenji.csv', 'mesi.csv', 'boronzi.csv', 'noghreyee.csv', 'talayee.csv', 'shakhgram.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zlib\n",
+    "import zipfile\n",
+    "\n",
+    "berenji_df.to_csv('berenji.csv', index=True)\n",
+    "mesi_df.to_csv('mesi.csv', index=True)\n",
+    "boronzi_df.to_csv(\"boronzi.csv\", index=True)\n",
+    "noghreyee_df.to_csv(\"noghreyee.csv\", index=True)\n",
+    "talayee_df.to_csv(\"talayee.csv\", index=True)\n",
+    "\n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"berenji.csv\", \"mesi.csv\", \"boronzi.csv\",\n",
+    "              \"noghreyee.csv\", \"talayee.csv\", \"shakhgram.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  }
+ ],
+ "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Machine Learning/ch3.pandas/pandas_yektanet/yektanet.ipynb b/Machine Learning/ch3.pandas/pandas_yektanet/yektanet.ipynb
new file mode 100644
index 0000000..2bdc72a
--- /dev/null
+++ b/Machine Learning/ch3.pandas/pandas_yektanet/yektanet.ipynb	
@@ -0,0 +1,591 @@
+{
+ "cells": [
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "ac1d21fa-520d-4a4c-9b26-7942fb833449",
+   "metadata": {},
+   "source": [
+    "<h1 align=center style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    یکتانت\n",
+    "</font>\n",
+    "</h1>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "داده‌هایی که در این تمرین استفاده می‌کنیم، داده‌هایی است که شرکت یکتانت از وبسایت‌های مختلف جمع‌آوری کرده است. قصد داریم با بررسی آمار مشاهده‌ی کاربران از وب‌سایت‌ها، درصد شهرت وب‌سایت‌های فارسی با کلید‌واژه‌های پرکاربردتر را محاسبه و بررسی کنیم.\n",
+    "</font>\n",
+    "</p>\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "e76aae32",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "مجموعه‌داده\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "داده‌های این تمرین از طریق فایل <code>yektanet.csv</code> در دسترس شما قرار گرفته است. ابتدا این فایل را خوانده و نگاهی به آن بیندازید. توضیح هرکدام از ستون‌های این مجموعه‌داده در جدول زیر آمده است:\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<center>\n",
+    "<div dir=rtl style=\"direction: rtl;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "\n",
+    "| ستون | توضیح |\n",
+    "| :---: | :---: |\n",
+    "| `keywords` | کلیدواژه‌هایی که هر وب‌سایت با آن‌ها تعریف می‌شود و معمولاً در جست‌و‌جوها با آن موضوع نمایش داده می‌شود. |\n",
+    "| `segments` | دسته‌بندی‌هایی که برای هر وب‌سایت در نظر می‌گیریم و برای تبلیغات استفاده می‌شوند. |\n",
+    "| `visit count` | تعداد بازدیدهای متوسط صفحه در یک روز |\n",
+    "\n",
+    "</font>\n",
+    "</div>\n",
+    "</center>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "65ff8052",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd \n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "995dac76",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>keywords</th>\n",
+       "      <th>segments</th>\n",
+       "      <th>visit count</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>219614</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment', 'varzeshi']</td>\n",
+       "      <td>641</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219615</th>\n",
+       "      <td>['زیرنویس', 'فارسی فیلم', 'دانلود زیرنویس']</td>\n",
+       "      <td>['femaleSegment']</td>\n",
+       "      <td>355</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219616</th>\n",
+       "      <td>['سامان', 'ساعت']</td>\n",
+       "      <td>['femaleSegment']</td>\n",
+       "      <td>40</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219617</th>\n",
+       "      <td>['چشم', 'تنظیم', 'رنگ', 'نرم']</td>\n",
+       "      <td>['technology']</td>\n",
+       "      <td>523</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219618</th>\n",
+       "      <td>['خارجی', 'قوی', 'ایرانی خارجی', 'ایرانی']</td>\n",
+       "      <td>['femaleSegment']</td>\n",
+       "      <td>906</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                           keywords  \\\n",
+       "219614              ['نتایج زنده', 'نتایج', 'زنده']   \n",
+       "219615  ['زیرنویس', 'فارسی فیلم', 'دانلود زیرنویس']   \n",
+       "219616                            ['سامان', 'ساعت']   \n",
+       "219617               ['چشم', 'تنظیم', 'رنگ', 'نرم']   \n",
+       "219618   ['خارجی', 'قوی', 'ایرانی خارجی', 'ایرانی']   \n",
+       "\n",
+       "                           segments  visit count  \n",
+       "219614  ['maleSegment', 'varzeshi']          641  \n",
+       "219615            ['femaleSegment']          355  \n",
+       "219616            ['femaleSegment']           40  \n",
+       "219617               ['technology']          523  \n",
+       "219618            ['femaleSegment']          906  "
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df = pd.read_csv(\"yektanet.csv\")\n",
+    "df.tail()"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "4d3fbe6b",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت اول\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    با استفاده از گروه‌بندی و تابع <code>filter</code>، وب‌سایت‌هایی را انتخاب کنید که کلیدواژه‌ی مربوط به آن‌ها، بیشتر از ۱۴۰۰ وب‌سایت را شامل می‌شود. به بیان دیگر، آن لیست کلیدواژه بیش از ۱۴۰۰ بار در دیتافریم تکرار شده باشد. حاصل را در متغیر <code>filtered_df</code> ذخیره کنید.\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:green\"><b>راهنمایی:</b></span>\n",
+    " ابتدا تابع <code>groupby</code> را برای کلید‌واژه‌ها اعمال کرده و سپس از از تابع <code>len</code> برای بررسی تعداد اعضا استفاده کنید. توجه کنید <b> لازم نیست </b>گروه‌بندی براساس هر عضو داخل آرایه‌ی کلیدواژه‌ها انجام شود.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "1f051715-851e-4865-b14f-dfe975ec0b2b",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>keywords</th>\n",
+       "      <th>segments</th>\n",
+       "      <th>visit count</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>822</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>307</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>778</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment', 'varzeshi']</td>\n",
+       "      <td>418</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>324</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219597</th>\n",
+       "      <td>['صفحه اخبار', 'صفحه', 'اخبار']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>669</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219601</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>742</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219608</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['salamat', 'laghari']</td>\n",
+       "      <td>472</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219612</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>460</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219614</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment', 'varzeshi']</td>\n",
+       "      <td>641</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>24247 rows × 3 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                               keywords                     segments  \\\n",
+       "1       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "3       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "4       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "10      ['نتایج زنده', 'نتایج', 'زنده']  ['maleSegment', 'varzeshi']   \n",
+       "16      ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "...                                 ...                          ...   \n",
+       "219597  ['صفحه اخبار', 'صفحه', 'اخبار']              ['maleSegment']   \n",
+       "219601  ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "219608  ['نتایج زنده', 'نتایج', 'زنده']       ['salamat', 'laghari']   \n",
+       "219612  ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "219614  ['نتایج زنده', 'نتایج', 'زنده']  ['maleSegment', 'varzeshi']   \n",
+       "\n",
+       "        visit count  \n",
+       "1               822  \n",
+       "3               307  \n",
+       "4               778  \n",
+       "10              418  \n",
+       "16              324  \n",
+       "...             ...  \n",
+       "219597          669  \n",
+       "219601          742  \n",
+       "219608          472  \n",
+       "219612          460  \n",
+       "219614          641  \n",
+       "\n",
+       "[24247 rows x 3 columns]"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "filtered_df = df.groupby(\"keywords\").filter(lambda x : len(x) > 1400)\n",
+    "filtered_df"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "81e5af66",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "    قسمت دوم\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    اکنون ستونی به نام <code>popularity</code> را به دیتافریم <code>filtered_df</code> اضافه کنید که نسبت تعداد بازدید هر وب‌سایت به مجموع تعداد بازدید تمام وب‌سایت‌های آن کلیدواژه را شامل شود.\n",
+    "</font>\n",
+    "</p>\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl;text-align: justify;line-height:200%;font-family:vazir;font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "<span style=\"color:green\"><b>راهنمایی:</b></span>\n",
+    "بعد از انجام گروه‌بندی بر اساس <code>keywords</code>، تابع <code>transfrom</code> را روی ستون <code>visit count</code> اعمال کرده و نسبت بازید هر وب‌سایت به مجموع بازدیدهای آن گروه کلیدواژه را محاسبه و در ۱۰۰ ضرب کنید.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "a7ccd820-70b9-4d19-ad6e-36fe65c8c16b",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>keywords</th>\n",
+       "      <th>segments</th>\n",
+       "      <th>visit count</th>\n",
+       "      <th>popularity</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>822</td>\n",
+       "      <td>0.016966</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>307</td>\n",
+       "      <td>0.006337</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>778</td>\n",
+       "      <td>0.016058</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment', 'varzeshi']</td>\n",
+       "      <td>418</td>\n",
+       "      <td>0.008628</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>324</td>\n",
+       "      <td>0.006687</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219597</th>\n",
+       "      <td>['صفحه اخبار', 'صفحه', 'اخبار']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>669</td>\n",
+       "      <td>0.014774</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219601</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>742</td>\n",
+       "      <td>0.015315</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219608</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['salamat', 'laghari']</td>\n",
+       "      <td>472</td>\n",
+       "      <td>0.009742</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219612</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment']</td>\n",
+       "      <td>460</td>\n",
+       "      <td>0.009494</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>219614</th>\n",
+       "      <td>['نتایج زنده', 'نتایج', 'زنده']</td>\n",
+       "      <td>['maleSegment', 'varzeshi']</td>\n",
+       "      <td>641</td>\n",
+       "      <td>0.013230</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>24247 rows × 4 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                               keywords                     segments  \\\n",
+       "1       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "3       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "4       ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "10      ['نتایج زنده', 'نتایج', 'زنده']  ['maleSegment', 'varzeshi']   \n",
+       "16      ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "...                                 ...                          ...   \n",
+       "219597  ['صفحه اخبار', 'صفحه', 'اخبار']              ['maleSegment']   \n",
+       "219601  ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "219608  ['نتایج زنده', 'نتایج', 'زنده']       ['salamat', 'laghari']   \n",
+       "219612  ['نتایج زنده', 'نتایج', 'زنده']              ['maleSegment']   \n",
+       "219614  ['نتایج زنده', 'نتایج', 'زنده']  ['maleSegment', 'varzeshi']   \n",
+       "\n",
+       "        visit count  popularity  \n",
+       "1               822    0.016966  \n",
+       "3               307    0.006337  \n",
+       "4               778    0.016058  \n",
+       "10              418    0.008628  \n",
+       "16              324    0.006687  \n",
+       "...             ...         ...  \n",
+       "219597          669    0.014774  \n",
+       "219601          742    0.015315  \n",
+       "219608          472    0.009742  \n",
+       "219612          460    0.009494  \n",
+       "219614          641    0.013230  \n",
+       "\n",
+       "[24247 rows x 4 columns]"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "filtered_df['popularity']  = filtered_df.groupby('keywords')['visit count'].transform(lambda x: (x/x.sum()*100))\n",
+    "filtered_df"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "f42669a5-fb7f-4f64-acb8-e026fb310789",
+   "metadata": {},
+   "source": [
+    "<h2 align=right style=\"line-height:200%;font-family:vazir;color:#0099cc\">\n",
+    "<font face=\"vazir\" color=\"#0099cc\">\n",
+    "<b>سلول جواب‌ساز</b>\n",
+    "</font>\n",
+    "</h2>\n",
+    "\n",
+    "\n",
+    "<p dir=rtl style=\"direction: rtl; text-align: justify; line-height:200%; font-family:vazir; font-size:medium\">\n",
+    "<font face=\"vazir\" size=3>\n",
+    "    برای ساخته‌شدن فایل <code>result.zip</code> سلول زیر را اجرا کنید. توجه داشته باشید که پیش از اجرای سلول زیر تغییرات اعمال شده در نت‌بوک را ذخیره کرده باشید (<code>ctrl+s</code>) تا در صورت نیاز به پشتیبانی امکان بررسی کد شما وجود داشته باشد.\n",
+    "</font>\n",
+    "</p>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "6bc08279-93f4-4c56-bd9d-4e3d457ff6ad",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File Paths:\n",
+      "['filtered_df.csv', 'yektanet.ipynb']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zlib\n",
+    "import zipfile\n",
+    "\n",
+    "filtered_df.to_csv('filtered_df.csv', index=True)\n",
+    "\n",
+    "def compress(file_names):\n",
+    "    print(\"File Paths:\")\n",
+    "    print(file_names)\n",
+    "    compression = zipfile.ZIP_DEFLATED\n",
+    "    with zipfile.ZipFile(\"result.zip\", mode=\"w\") as zf:\n",
+    "        for file_name in file_names:\n",
+    "            zf.write('./' + file_name, file_name, compress_type=compression)\n",
+    "\n",
+    "file_names = [\"filtered_df.csv\", \"yektanet.ipynb\"]\n",
+    "compress(file_names)"
+   ]
+  }
+ ],
+ "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": "3.8.1"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "44e7e1b8fa2096bd5707ed7fd18b1724a2db25f4c565a7673f8b6e7bfc49d25d"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}