aboutsummaryrefslogtreecommitdiff
path: root/CS105MiniProject.ipynb
diff options
context:
space:
mode:
authorGravatar Anshul Gupta <ansg191@anshulg.com> 2024-02-15 17:13:38 -0800
committerGravatar Anshul Gupta <ansg191@anshulg.com> 2024-02-15 17:13:38 -0800
commit0441d9bdead1e968b8d05fdac0001b7609059d0e (patch)
tree6dddd83e1418b058a9308bd508fab96734d60b9d /CS105MiniProject.ipynb
parentaa9c60b320a3a2e5679826b471f44934ef64fdae (diff)
downloadCS105MiniProject-0441d9bdead1e968b8d05fdac0001b7609059d0e.tar.gz
CS105MiniProject-0441d9bdead1e968b8d05fdac0001b7609059d0e.tar.zst
CS105MiniProject-0441d9bdead1e968b8d05fdac0001b7609059d0e.zip
Adds working and non-working dataframe
Diffstat (limited to 'CS105MiniProject.ipynb')
-rw-r--r--CS105MiniProject.ipynb61
1 files changed, 51 insertions, 10 deletions
diff --git a/CS105MiniProject.ipynb b/CS105MiniProject.ipynb
index dda76e0..603dbfc 100644
--- a/CS105MiniProject.ipynb
+++ b/CS105MiniProject.ipynb
@@ -12,8 +12,8 @@
"height": 614
},
"ExecuteTime": {
- "end_time": "2024-02-16T00:58:32.054806Z",
- "start_time": "2024-02-16T00:58:31.371211Z"
+ "end_time": "2024-02-16T01:13:23.980165Z",
+ "start_time": "2024-02-16T01:13:23.265957Z"
}
},
"outputs": [
@@ -44,11 +44,21 @@
"metadata": {
"id": "29889175",
"ExecuteTime": {
- "end_time": "2024-02-16T00:58:32.059881Z",
- "start_time": "2024-02-16T00:58:32.056070Z"
+ "end_time": "2024-02-16T01:13:23.990972Z",
+ "start_time": "2024-02-16T01:13:23.981386Z"
}
},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/plain": " Timestamp What is your current class standing? \\\n0 2/9/2024 20:12:14 Senior \n1 2/9/2024 20:16:34 Junior \n2 2/9/2024 20:18:55 Junior \n3 2/9/2024 20:24:00 Senior \n4 2/9/2024 20:26:16 Graduate \n.. ... ... \n255 2/14/2024 19:46:28 Junior \n256 2/15/2024 0:28:38 NaN \n257 2/15/2024 8:33:45 Senior \n258 2/15/2024 16:10:40 Sophomore \n259 2/15/2024 16:14:11 Sophomore \n\n What gender do you identify as? Who do you live with? \\\n0 Male Neither \n1 Female Both \n2 Female Friends \n3 Male Neither \n4 Male Neither \n.. ... ... \n255 Male Friends \n256 Male Family \n257 Male Family \n258 Female Family \n259 Female Friends \n\n Do you currently live in a house, apartnment, or dorm? \\\n0 House \n1 Apartment \n2 House \n3 Apartment \n4 Apartment \n.. ... \n255 House \n256 Apartment \n257 House \n258 Apartment \n259 Dorm \n\n How many people live in your household? Do you currently work? \\\n0 6 Yes \n1 4 No \n2 4 No \n3 1 No \n4 1 Yes \n.. ... ... \n255 5 Yes \n256 4 No \n257 9 No \n258 4 Yes \n259 3 No \n\n How many hours do you work per week on average? \\\n0 5 - 10 \n1 NaN \n2 NaN \n3 NaN \n4 10 - 20 \n.. ... \n255 10 - 20 \n256 NaN \n257 1 - 5 \n258 5 - 10 \n259 NaN \n\n Do you work on or off campus? \\\n0 Off-campus \n1 NaN \n2 NaN \n3 NaN \n4 Off-campus \n.. ... \n255 On-campus \n256 NaN \n257 Off-campus \n258 On-campus \n259 NaN \n\n Do you work in a department related to your major? \\\n0 No \n1 NaN \n2 NaN \n3 No \n4 Yes \n.. ... \n255 No \n256 NaN \n257 No \n258 No \n259 NaN \n\n Do you have roommates that are part of your major? \n0 No \n1 Yes \n2 No \n3 No \n4 No \n.. ... \n255 No \n256 No \n257 No \n258 No \n259 Yes \n\n[260 rows x 11 columns]",
+ "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>Timestamp</th>\n <th>What is your current class standing?</th>\n <th>What gender do you identify as?</th>\n <th>Who do you live with?</th>\n <th>Do you currently live in a house, apartnment, or dorm?</th>\n <th>How many people live in your household?</th>\n <th>Do you currently work?</th>\n <th>How many hours do you work per week on average?</th>\n <th>Do you work on or off campus?</th>\n <th>Do you work in a department related to your major?</th>\n <th>Do you have roommates that are part of your major?</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>0</th>\n <td>2/9/2024 20:12:14</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Neither</td>\n <td>House</td>\n <td>6</td>\n <td>Yes</td>\n <td>5 - 10</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>1</th>\n <td>2/9/2024 20:16:34</td>\n <td>Junior</td>\n <td>Female</td>\n <td>Both</td>\n <td>Apartment</td>\n <td>4</td>\n <td>No</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>Yes</td>\n </tr>\n <tr>\n <th>2</th>\n <td>2/9/2024 20:18:55</td>\n <td>Junior</td>\n <td>Female</td>\n <td>Friends</td>\n <td>House</td>\n <td>4</td>\n <td>No</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>No</td>\n </tr>\n <tr>\n <th>3</th>\n <td>2/9/2024 20:24:00</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Neither</td>\n <td>Apartment</td>\n <td>1</td>\n <td>No</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>4</th>\n <td>2/9/2024 20:26:16</td>\n <td>Graduate</td>\n <td>Male</td>\n <td>Neither</td>\n <td>Apartment</td>\n <td>1</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>Off-campus</td>\n <td>Yes</td>\n <td>No</td>\n </tr>\n <tr>\n <th>...</th>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n </tr>\n <tr>\n <th>255</th>\n <td>2/14/2024 19:46:28</td>\n <td>Junior</td>\n <td>Male</td>\n <td>Friends</td>\n <td>House</td>\n <td>5</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>256</th>\n <td>2/15/2024 0:28:38</td>\n <td>NaN</td>\n <td>Male</td>\n <td>Family</td>\n <td>Apartment</td>\n <td>4</td>\n <td>No</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>No</td>\n </tr>\n <tr>\n <th>257</th>\n <td>2/15/2024 8:33:45</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Family</td>\n <td>House</td>\n <td>9</td>\n <td>No</td>\n <td>1 - 5</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>258</th>\n <td>2/15/2024 16:10:40</td>\n <td>Sophomore</td>\n <td>Female</td>\n <td>Family</td>\n <td>Apartment</td>\n <td>4</td>\n <td>Yes</td>\n <td>5 - 10</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>259</th>\n <td>2/15/2024 16:14:11</td>\n <td>Sophomore</td>\n <td>Female</td>\n <td>Friends</td>\n <td>Dorm</td>\n <td>3</td>\n <td>No</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>NaN</td>\n <td>Yes</td>\n </tr>\n </tbody>\n</table>\n<p>260 rows × 11 columns</p>\n</div>"
+ },
+ "execution_count": 2,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"df['How many people live in your household?'] = (df['How many people live in your household?']\n",
" .fillna(0)\n",
@@ -59,22 +69,53 @@
" .replace('5 including me', '5')\n",
" .replace('North District 4 bed 2 bath', '4')\n",
" .replace('3 (room), 8 (hall), ~70 (building)', '3')\n",
- " .astype(int))"
+ " .astype(int))\n",
+ "df"
]
},
{
"cell_type": "code",
+ "outputs": [
+ {
+ "data": {
+ "text/plain": " Timestamp What is your current class standing? \\\n0 2/9/2024 20:12:14 Senior \n4 2/9/2024 20:26:16 Graduate \n8 2/9/2024 22:02:49 Junior \n9 2/9/2024 22:08:43 Senior \n13 2/9/2024 22:15:13 Junior \n.. ... ... \n246 2/13/2024 19:37:02 Graduate \n247 2/13/2024 21:39:14 Senior \n252 2/14/2024 9:48:12 Junior \n255 2/14/2024 19:46:28 Junior \n258 2/15/2024 16:10:40 Sophomore \n\n What gender do you identify as? Who do you live with? \\\n0 Male Neither \n4 Male Neither \n8 Male Friends \n9 Male Family \n13 Male Family \n.. ... ... \n246 Male Family \n247 Male Friends \n252 Male Family \n255 Male Friends \n258 Female Family \n\n Do you currently live in a house, apartnment, or dorm? \\\n0 House \n4 Apartment \n8 House \n9 House \n13 Apartment \n.. ... \n246 House \n247 Apartment \n252 House \n255 House \n258 Apartment \n\n How many people live in your household? Do you currently work? \\\n0 6 Yes \n4 1 Yes \n8 6 Yes \n9 5 Yes \n13 4 Yes \n.. ... ... \n246 2 Yes \n247 3 Yes \n252 5 Yes \n255 5 Yes \n258 4 Yes \n\n How many hours do you work per week on average? \\\n0 5 - 10 \n4 10 - 20 \n8 10 - 20 \n9 1 - 5 \n13 10 - 20 \n.. ... \n246 10 - 20 \n247 20 - 40 \n252 20 - 40 \n255 10 - 20 \n258 5 - 10 \n\n Do you work on or off campus? \\\n0 Off-campus \n4 Off-campus \n8 On-campus \n9 On-campus \n13 Off-campus \n.. ... \n246 On-campus \n247 Off-campus \n252 Off-campus \n255 On-campus \n258 On-campus \n\n Do you work in a department related to your major? \\\n0 No \n4 Yes \n8 No \n9 No \n13 No \n.. ... \n246 Yes \n247 No \n252 No \n255 No \n258 No \n\n Do you have roommates that are part of your major? \n0 No \n4 No \n8 No \n9 No \n13 No \n.. ... \n246 No \n247 Yes \n252 No \n255 No \n258 No \n\n[77 rows x 11 columns]",
+ "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>Timestamp</th>\n <th>What is your current class standing?</th>\n <th>What gender do you identify as?</th>\n <th>Who do you live with?</th>\n <th>Do you currently live in a house, apartnment, or dorm?</th>\n <th>How many people live in your household?</th>\n <th>Do you currently work?</th>\n <th>How many hours do you work per week on average?</th>\n <th>Do you work on or off campus?</th>\n <th>Do you work in a department related to your major?</th>\n <th>Do you have roommates that are part of your major?</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>0</th>\n <td>2/9/2024 20:12:14</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Neither</td>\n <td>House</td>\n <td>6</td>\n <td>Yes</td>\n <td>5 - 10</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>4</th>\n <td>2/9/2024 20:26:16</td>\n <td>Graduate</td>\n <td>Male</td>\n <td>Neither</td>\n <td>Apartment</td>\n <td>1</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>Off-campus</td>\n <td>Yes</td>\n <td>No</td>\n </tr>\n <tr>\n <th>8</th>\n <td>2/9/2024 22:02:49</td>\n <td>Junior</td>\n <td>Male</td>\n <td>Friends</td>\n <td>House</td>\n <td>6</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>9</th>\n <td>2/9/2024 22:08:43</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Family</td>\n <td>House</td>\n <td>5</td>\n <td>Yes</td>\n <td>1 - 5</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>13</th>\n <td>2/9/2024 22:15:13</td>\n <td>Junior</td>\n <td>Male</td>\n <td>Family</td>\n <td>Apartment</td>\n <td>4</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>...</th>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n </tr>\n <tr>\n <th>246</th>\n <td>2/13/2024 19:37:02</td>\n <td>Graduate</td>\n <td>Male</td>\n <td>Family</td>\n <td>House</td>\n <td>2</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>On-campus</td>\n <td>Yes</td>\n <td>No</td>\n </tr>\n <tr>\n <th>247</th>\n <td>2/13/2024 21:39:14</td>\n <td>Senior</td>\n <td>Male</td>\n <td>Friends</td>\n <td>Apartment</td>\n <td>3</td>\n <td>Yes</td>\n <td>20 - 40</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>Yes</td>\n </tr>\n <tr>\n <th>252</th>\n <td>2/14/2024 9:48:12</td>\n <td>Junior</td>\n <td>Male</td>\n <td>Family</td>\n <td>House</td>\n <td>5</td>\n <td>Yes</td>\n <td>20 - 40</td>\n <td>Off-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>255</th>\n <td>2/14/2024 19:46:28</td>\n <td>Junior</td>\n <td>Male</td>\n <td>Friends</td>\n <td>House</td>\n <td>5</td>\n <td>Yes</td>\n <td>10 - 20</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n <tr>\n <th>258</th>\n <td>2/15/2024 16:10:40</td>\n <td>Sophomore</td>\n <td>Female</td>\n <td>Family</td>\n <td>Apartment</td>\n <td>4</td>\n <td>Yes</td>\n <td>5 - 10</td>\n <td>On-campus</td>\n <td>No</td>\n <td>No</td>\n </tr>\n </tbody>\n</table>\n<p>77 rows × 11 columns</p>\n</div>"
+ },
+ "execution_count": 3,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Working DataFrame\n",
+ "w_df = df[df['Do you currently work?'] == 'Yes']\n",
+ "# Not working DataFrame\n",
+ "nw_df = df[df['Do you currently work?'] == 'No']\n",
+ "w_df"
+ ],
+ "metadata": {
+ "collapsed": false,
+ "ExecuteTime": {
+ "end_time": "2024-02-16T01:13:24.001347Z",
+ "start_time": "2024-02-16T01:13:23.991767Z"
+ }
+ },
+ "id": "de4448fd64205d85",
+ "execution_count": 3
+ },
+ {
+ "cell_type": "code",
"outputs": [],
"source": [],
"metadata": {
"collapsed": false,
"ExecuteTime": {
- "end_time": "2024-02-16T00:58:32.062613Z",
- "start_time": "2024-02-16T00:58:32.061050Z"
+ "end_time": "2024-02-16T01:13:24.004250Z",
+ "start_time": "2024-02-16T01:13:24.002783Z"
}
},
- "id": "de4448fd64205d85",
- "execution_count": 2
+ "id": "a9df972e44277de7",
+ "execution_count": 3
}
],
"metadata": {