Skip to content

Instantly share code, notes, and snippets.

@malibayram

malibayram/House Price India with Notater.csv

Last active September 23, 2025 12:32
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save malibayram/3da1a1e08b5f49005ce731eae6f6806c to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save malibayram/3da1a1e08b5f49005ce731eae6f6806c to your computer and use it in GitHub Desktop.
Download ZIP
Raw
House Price India with Notater.csv
Price living_area_renov Price per living_area_renov Notater
4.175 1.39 300.35971223021585 Real Estate Expert
10.8 3.23 334.3653250773994 Other
3.7 1.62 228.39506172839504 Real Estate Expert
2.63 2.15 122.32558139534883 Other
3.35 1.41 237.5886524822695 Real Estate Expert
5.16 2.43 212.34567901234567 Real Estate Expert
4.1 1.55 264.51612903225805 Real Estate Expert
7.2 3.16 227.8481012658228 Real Estate Expert
6.7 2.71 247.2324723247233 Real Estate Expert
4.5 1.86 241.93548387096777 Real Estate Expert
2.25 1.93 116.58031088082902 Other
7.75 2.88 269.09722222222223 Real Estate Expert
2.75 1.72 159.88372093023256 Other
4.6995 1.71 274.82456140350877 Real Estate Expert
12.8 4.03 317.61786600496276 Other
3.07 0.99 310.1010101010101 Other
4.6695 1.11 420.6756756756757 House Owner
8.75 1.91 458.1151832460733 House Owner
5.0 2.84 176.05633802816902 Other
2.66 1.88 141.48936170212767 Other
2.7 1.55 174.19354838709677 Other
5.15 2.08 247.59615384615384 Real Estate Expert
8.199 1.56 525.5769230769231 House Owner
3.22 2.25 143.11111111111111 Other
3.8510000000000004 1.3 296.2307692307692 Real Estate Expert
3.99 1.36 293.38235294117646 Real Estate Expert
8.9 2.3 386.9565217391304 House Owner
4.25 1.54 275.97402597402595 Real Estate Expert
3.1 1.4 221.42857142857144 Real Estate Expert
1.57 1.25 125.6 Other
5.5 1.92 286.4583333333333 Real Estate Expert
4.9795 1.54 323.34415584415586 Other
2.685 1.14 235.52631578947367 Real Estate Expert
3.9 1.47 265.3061224489796 Real Estate Expert
5.5 2.05 268.2926829268293 Real Estate Expert
4.385 1.32 332.1969696969697 Other
3.35 1.29 259.68992248062017 Real Estate Expert
5.35 1.44 371.52777777777777 House Owner
5.55 1.96 283.16326530612247 Real Estate Expert
3.4981 1.25 279.848 Real Estate Expert
3.135 2.12 147.87735849056602 Other
6.53 1.82 358.7912087912088 House Owner
3.98 2.665 149.343339587242 Other
3.1289999999999996 1.83 170.98360655737704 Other
6.2 1.67 371.25748502994014 House Owner
2.57 1.55 165.80645161290323 Other
3.45 1.39 248.20143884892087 Real Estate Expert
2.47 1.73 142.77456647398844 Other
2.67 1.264 211.23417721518987 Real Estate Expert
3.1 2.7 114.8148148148148 Other
3.74 1.67 223.95209580838323 Real Estate Expert
4.45 1.68 264.8809523809524 Real Estate Expert
7.4 2.64 280.3030303030303 Real Estate Expert
2.6 1.99 130.6532663316583 Other
3.5 2.29 152.83842794759826 Other
6.41 2.4 267.0833333333333 Real Estate Expert
7.85 3.32 236.44578313253012 Real Estate Expert
4.89 2.75 177.8181818181818 Other
6.04 2.69 224.53531598513013 Real Estate Expert
4.53 1.7 266.47058823529414 Real Estate Expert
10.5 2.86 367.13286713286715 House Owner
2.24 1.54 145.45454545454547 Other
4.45 1.4 317.85714285714283 Other
9.6999 4.07 238.32678132678132 Real Estate Expert
7.56 1.85 408.64864864864865 House Owner
4.95 2.44 202.8688524590164 Other
4.92 2.37 207.59493670886076 Real Estate Expert
3.56 1.92 185.41666666666663 Other
4.675 2.07 225.84541062801932 Real Estate Expert
5.95 2.56 232.421875 Real Estate Expert
3.185 1.55 205.48387096774192 Real Estate Expert
3.35 2.27 147.5770925110132 Other
2.94 1.4 210.0 Real Estate Expert
3.645 1.52 239.80263157894737 Real Estate Expert
10.8 2.7 400.0 House Owner
4.65 2.91 159.79381443298968 Other
17.7 3.82 952.8795811518324 House Owner
15.0 2.94 510.2040816326531 House Owner
2.555 1.59 160.69182389937106 Other
2.89 2.74 105.47445255474452 Other
2.34 1.2 195.0 Other
9.9 2.34 423.0769230769231 House Owner
5.85 2.5 234.0 Real Estate Expert
7.35 1.9 386.8421052631579 House Owner
2.7 2.15 125.5813953488372 Other
4.3 2.16 199.07407407407408 Other
16.5 2.6 634.6153846153846 House Owner
2.1 1.04 201.92307692307693 Other
3.1 2.08 149.03846153846155 Other
6.4 2.77 231.04693140794225 Real Estate Expert
1.055 1.62 65.12345679012346 Other
5.8 1.35 429.6296296296296 House Owner
4.2804 2.15 199.08837209302325 Other
2.4 1.53 156.86274509803923 Other
11.0 3.05 360.655737704918 House Owner
5.25 2.1 250.0 Real Estate Expert
3.27 1.5 218.0 Real Estate Expert
2.95 2.59 113.8996138996139 Other
4.35 2.19 198.63013698630135 Other
3.55 2.31 153.67965367965368 Other
Display the source blob
Display the rendered blob
Raw
yapayzeka_102.ipynb
{
"cells": [
{
"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>Price</th>\n",
" <th>living_area_renov</th>\n",
" <th>Price per living_area_renov</th>\n",
" <th>Notater</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>4.175</td>\n",
" <td>1.39</td>\n",
" <td>300.359712</td>\n",
" <td>Real Estate Expert</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>10.800</td>\n",
" <td>3.23</td>\n",
" <td>334.365325</td>\n",
" <td>Other</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>3.700</td>\n",
" <td>1.62</td>\n",
" <td>228.395062</td>\n",
" <td>Real Estate Expert</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2.630</td>\n",
" <td>2.15</td>\n",
" <td>122.325581</td>\n",
" <td>Other</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>3.350</td>\n",
" <td>1.41</td>\n",
" <td>237.588652</td>\n",
" <td>Real Estate Expert</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>95</th>\n",
" <td>5.250</td>\n",
" <td>2.10</td>\n",
" <td>250.000000</td>\n",
" <td>Real Estate Expert</td>\n",
" </tr>\n",
" <tr>\n",
" <th>96</th>\n",
" <td>3.270</td>\n",
" <td>1.50</td>\n",
" <td>218.000000</td>\n",
" <td>Real Estate Expert</td>\n",
" </tr>\n",
" <tr>\n",
" <th>97</th>\n",
" <td>2.950</td>\n",
" <td>2.59</td>\n",
" <td>113.899614</td>\n",
" <td>Other</td>\n",
" </tr>\n",
" <tr>\n",
" <th>98</th>\n",
" <td>4.350</td>\n",
" <td>2.19</td>\n",
" <td>198.630137</td>\n",
" <td>Other</td>\n",
" </tr>\n",
" <tr>\n",
" <th>99</th>\n",
" <td>3.550</td>\n",
" <td>2.31</td>\n",
" <td>153.679654</td>\n",
" <td>Other</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>100 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" Price living_area_renov Price per living_area_renov Notater\n",
"0 4.175 1.39 300.359712 Real Estate Expert\n",
"1 10.800 3.23 334.365325 Other\n",
"2 3.700 1.62 228.395062 Real Estate Expert\n",
"3 2.630 2.15 122.325581 Other\n",
"4 3.350 1.41 237.588652 Real Estate Expert\n",
".. ... ... ... ...\n",
"95 5.250 2.10 250.000000 Real Estate Expert\n",
"96 3.270 1.50 218.000000 Real Estate Expert\n",
"97 2.950 2.59 113.899614 Other\n",
"98 4.350 2.19 198.630137 Other\n",
"99 3.550 2.31 153.679654 Other\n",
"\n",
"[100 rows x 4 columns]"
]
},
"execution_count": 1,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import pandas as pd\n",
"\n",
"data = pd.read_csv('House Price India with Notater.csv')\n",
"\n",
"data"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# plot the data color coded by notater\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"\n",
"sns.scatterplot(data=data, x='living_area_renov', y='Price', hue='Notater')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Kavramlar\n",
"\n",
"- **Linear Reagression**: Doğrusal regresyon, bir bağımlı değişkenin bir veya daha fazla bağımsız değişkenle olan ilişkisini modellemek için kullanılan bir regresyon analizidir.\n",
"- **Forward Pass**: İleri geçiş, bir sinir ağındaki girdilerin ağırlıklarla çarpılması ve ardından aktivasyon fonksiyonuna uygulanmasıdır."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 165,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(0.417022004702574, 0.7203244934421581)"
]
},
"execution_count": 165,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import numpy as np\n",
"\n",
"# seed\n",
"np.random.seed(1)\n",
"# random weight and bias\n",
"weight = np.random.rand() # 2.5\n",
"bias = np.random.rand() # -0.01\n",
"\n",
"weight, bias"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
"def fiyat_tahmin_et(m2):\n",
" global weight, bias\n",
" return m2 * weight + bias\n",
"\n",
"def hata_hesapla(m2, fiyat):\n",
" tahmin = fiyat_tahmin_et(m2)\n",
" return fiyat - tahmin"
]
},
{
"cell_type": "code",
"execution_count": 168,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# draw the line with colored dots\n",
"plt.plot(data['living_area_renov'], fiyat_tahmin_et(data['living_area_renov']), color='red')\n",
"sns.scatterplot(data=data, x='living_area_renov', y='Price', hue='Notater')\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(1.39, 4.175)"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data.loc[0]['living_area_renov'], data.loc[0]['Price']"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(1.299985079978736, 2.875014920021264)"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"tahmin, hata"
]
},
{
"cell_type": "code",
"execution_count": 161,
"metadata": {},
"outputs": [],
"source": [
"epoch = 100\n",
"learning_rate = 1e-3"
]
},
{
"cell_type": "code",
"execution_count": 154,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(2.2256066779578036, 2.0214645461437666)"
]
},
"execution_count": 154,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tahmin = fiyat_tahmin_et(data.loc[0]['living_area_renov'])\n",
"hata = hata_hesapla(data.loc[0]['living_area_renov'], data.loc[0]['Price'])\n",
"\n",
"weight = weight + hata * learning_rate * data.loc[0]['living_area_renov']\n",
"bias = bias + hata * learning_rate\n",
"\n",
"weight, bias"
]
},
{
"cell_type": "code",
"execution_count": 163,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch: 0, Weight: 1.0712010550832147, Bias: 0.9944412860478115\n",
"Epoch: 1, Weight: 1.45100666991369, Bias: 1.1441126661629624\n",
"Epoch: 2, Weight: 1.6729910456510892, Bias: 1.2222767405017227\n",
"Epoch: 3, Weight: 1.8041841117270405, Bias: 1.2593779826624356\n",
"Epoch: 4, Weight: 1.8831353082482558, Bias: 1.2729246762299729\n",
"Epoch: 5, Weight: 1.9320148108930402, Bias: 1.2729854450276896\n",
"Epoch: 6, Weight: 1.9635730020260191, Bias: 1.265350333374193\n",
"Epoch: 7, Weight: 1.9851428951366512, Bias: 1.2533487581123504\n",
"Epoch: 8, Weight: 2.000941947887785, Bias: 1.2388950213055652\n",
"Epoch: 9, Weight: 2.013395845590675, Bias: 1.2230895894570606\n",
"Epoch: 10, Weight: 2.0238998152925265, Bias: 1.2065648919766574\n",
"Epoch: 11, Weight: 2.0332564610854518, Bias: 1.1896841914706962\n",
"Epoch: 12, Weight: 2.041927565255095, Bias: 1.1726559549496889\n",
"Epoch: 13, Weight: 2.050178900312414, Bias: 1.15559962921829\n",
"Epoch: 14, Weight: 2.0581635110833396, Bias: 1.138583468583192\n",
"Epoch: 15, Weight: 2.0659696106404857, Bias: 1.1216462897183885\n",
"Epoch: 16, Weight: 2.0736481254896373, Bias: 1.1048099829636011\n",
"Epoch: 17, Weight: 2.081228536947278, Bias: 1.0880867076043237\n",
"Epoch: 18, Weight: 2.088727991602714, Bias: 1.0714830298850264\n",
"Epoch: 19, Weight: 2.0961565407990865, Bias: 1.0550023027740651\n",
"Epoch: 20, Weight: 2.103520153894995, Bias: 1.0386460345519755\n",
"Epoch: 21, Weight: 2.1108224512167877, Bias: 1.0224146758675705\n",
"Epoch: 22, Weight: 2.118065700698891, Bias: 1.0063080723523132\n",
"Epoch: 23, Weight: 2.1252513910675987, Bias: 0.9903257248959255\n",
"Epoch: 24, Weight: 2.132380561492924, Bias: 0.9744669393072282\n",
"Epoch: 25, Weight: 2.1394539911840065, Bias: 0.9587309123601331\n",
"Epoch: 26, Weight: 2.1464723084373323, Bias: 0.9431167812546074\n",
"Epoch: 27, Weight: 2.153436053361787, Bias: 0.9276236520376229\n",
"Epoch: 28, Weight: 2.160345713962952, Bias: 0.9122506159240722\n",
"Epoch: 29, Weight: 2.167201746906117, Bias: 0.8969967586590891\n",
"Epoch: 30, Weight: 2.1740045894678386, Bias: 0.8818611658786386\n",
"Epoch: 31, Weight: 2.180754666419927, Bias: 0.8668429261688649\n",
"Epoch: 32, Weight: 2.1874523939989508, Bias: 0.8519411328021818\n",
"Epoch: 33, Weight: 2.1940981821995496, Bias: 0.8371548847125201\n",
"Epoch: 34, Weight: 2.2006924361036533, Bias: 0.8224832870331981\n",
"Epoch: 35, Weight: 2.2072355566551867, Bias: 0.8079254513834285\n",
"Epoch: 36, Weight: 2.2137279411157906, Bias: 0.7934804960104459\n",
"Epoch: 37, Weight: 2.220169983337, Bias: 0.7791475458487711\n",
"Epoch: 38, Weight: 2.2265620739267895, Bias: 0.7649257325320015\n",
"Epoch: 39, Weight: 2.232904600355323, Bias: 0.750814194377467\n",
"Epoch: 40, Weight: 2.2391979470256054, Bias: 0.7368120763554514\n",
"Epoch: 41, Weight: 2.2454424953239362, Bias: 0.7229185300497145\n",
"Epoch: 42, Weight: 2.25163862365861, Bias: 0.7091327136131707\n",
"Epoch: 43, Weight: 2.2577867074918307, Bias: 0.6954537917209562\n",
"Epoch: 44, Weight: 2.2638871193676287, Bias: 0.6818809355221599\n",
"Epoch: 45, Weight: 2.2699402289373998, Bias: 0.6684133225909488\n",
"Epoch: 46, Weight: 2.2759464029840104, Bias: 0.6550501368775088\n",
"Epoch: 47, Weight: 2.281906005445006, Bias: 0.6417905686590435\n",
"Epoch: 48, Weight: 2.2878193974352232, Bias: 0.6286338144909661\n",
"Epoch: 49, Weight: 2.2936869372690056, Bias: 0.615579077158366\n",
"Epoch: 50, Weight: 2.2995089804820905, Bias: 0.6026255656277847\n",
"Epoch: 51, Weight: 2.3052858798532694, Bias: 0.5897724949993342\n",
"Epoch: 52, Weight: 2.311017985425824, Bias: 0.5770190864591654\n",
"Epoch: 53, Weight: 2.3167056445287684, Bias: 0.5643645672322931\n",
"Epoch: 54, Weight: 2.322349201797912, Bias: 0.5518081705357769\n",
"Epoch: 55, Weight: 2.3279489991967637, Bias: 0.5393491355322699\n",
"Epoch: 56, Weight: 2.3335053760372593, Bias: 0.5269867072839172\n",
"Epoch: 57, Weight: 2.339018669000311, Bias: 0.5147201367066115\n",
"Epoch: 58, Weight: 2.3444892121562284, Bias: 0.5025486805246081\n",
"Epoch: 59, Weight: 2.349917336984954, Bias: 0.49047160122548367\n",
"Epoch: 60, Weight: 2.355303372396144, Bias: 0.478488167015451\n",
"Epoch: 61, Weight: 2.3606476447491076, Bias: 0.46659765177501694\n",
"Epoch: 62, Weight: 2.3659504778725857, Bias: 0.4547993350149826\n",
"Epoch: 63, Weight: 2.3712121930843546, Bias: 0.4430925018327875\n",
"Epoch: 64, Weight: 2.3764331092107183, Bias: 0.43147644286919173\n",
"Epoch: 65, Weight: 2.38161354260581, Bias: 0.4199504542652922\n",
"Epoch: 66, Weight: 2.3867538071707712, Bias: 0.40851383761987387\n",
"Epoch: 67, Weight: 2.391854214372774, Bias: 0.39716589994709084\n",
"Epoch: 68, Weight: 2.396915073263878, Bias: 0.38590595363447533\n",
"Epoch: 69, Weight: 2.4019366904997765, Bias: 0.3747333164012749\n",
"Epoch: 70, Weight: 2.406919370358368, Bias: 0.3636473112571081\n",
"Epoch: 71, Weight: 2.411863414758192, Bias: 0.35264726646094385\n",
"Epoch: 72, Weight: 2.4167691232767314, Bias: 0.3417325154803991\n",
"Epoch: 73, Weight: 2.421636793168553, Bias: 0.3309023969513507\n",
"Epoch: 74, Weight: 2.4264667193833307, Bias: 0.32015625463786074\n",
"Epoch: 75, Weight: 2.4312591945837125, Bias: 0.3094934373924129\n",
"Epoch: 76, Weight: 2.4360145091630536, Bias: 0.2989132991164577\n",
"Epoch: 77, Weight: 2.4407329512630134, Bias: 0.2884151987212622\n",
"Epoch: 78, Weight: 2.445414806791015, Bias: 0.2779985000890644\n",
"Epoch: 79, Weight: 2.4500603594375683, Bias: 0.2676625720345289\n",
"Epoch: 80, Weight: 2.454669890693462, Bias: 0.2574067882665007\n",
"Epoch: 81, Weight: 2.4592436798668147, Bias: 0.2472305273500566\n",
"Epoch: 82, Weight: 2.4637820041000116, Bias: 0.23713317266884962\n",
"Epoch: 83, Weight: 2.468285138386481, Bias: 0.22711411238774556\n",
"Epoch: 84, Weight: 2.4727533555873693, Bias: 0.2171727394157507\n",
"Epoch: 85, Weight: 2.47718692644807, Bias: 0.20730845136922585\n",
"Epoch: 86, Weight: 2.4815861196146285, Bias: 0.19752065053538523\n",
"Epoch: 87, Weight: 2.4859512016500283, Bias: 0.18780874383607893\n",
"Epoch: 88, Weight: 2.4902824370503267, Bias: 0.17817214279185667\n",
"Epoch: 89, Weight: 2.494580088260701, Bias: 0.16861026348630886\n",
"Epoch: 90, Weight: 2.498844415691336, Bias: 0.159122526530685\n",
"Epoch: 91, Weight: 2.5030756777332015, Bias: 0.1497083570287885\n",
"Epoch: 92, Weight: 2.507274130773733, Bias: 0.14036718454213915\n",
"Epoch: 93, Weight: 2.5114400292123333, Bias: 0.1310984430554094\n",
"Epoch: 94, Weight: 2.5155736254758176, Bias: 0.12190157094212721\n",
"Epoch: 95, Weight: 2.519675170033688, Bias: 0.1127760109306445\n",
"Epoch: 96, Weight: 2.5237449114133206, Bias: 0.10372121007037152\n",
"Epoch: 97, Weight: 2.5277830962150216, Bias: 0.0947366196982703\n",
"Epoch: 98, Weight: 2.531789969126971, Bias: 0.08582169540560919\n",
"Epoch: 99, Weight: 2.5357657729400507, Bias: 0.07697589700497587\n"
]
}
],
"source": [
"\n",
"\n",
"for e in range(epoch):\n",
" for i in range(len(data)):\n",
" tahmin = fiyat_tahmin_et(data.loc[i]['living_area_renov'])\n",
" hata = hata_hesapla(data.loc[i]['living_area_renov'], data.loc[i]['Price'])\n",
" weight = weight + hata * learning_rate * data.loc[i]['living_area_renov']\n",
" bias = bias + hata * learning_rate\n",
" print(f'Epoch: {e}, Weight: {weight}, Bias: {bias}')"
]
},
{
"cell_type": "code",
"execution_count": 167,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch: 0, Weight: 1.5375382732028664, Bias: 1.2044004000246418\n",
"Epoch: 1, Weight: 1.8119064765248365, Bias: 1.2924281543917684\n",
"Epoch: 2, Weight: 1.8875162200133127, Bias: 1.2880939090692922\n",
"Epoch: 3, Weight: 1.9162531766862105, Bias: 1.262631743928874\n",
"Epoch: 4, Weight: 1.9337556195798278, Bias: 1.2327417743456355\n",
"Epoch: 5, Weight: 1.9483895842999528, Bias: 1.2023304738391405\n",
"Epoch: 6, Weight: 1.96212281471835, Bias: 1.1723031219118623\n",
"Epoch: 7, Weight: 1.9754218798699001, Bias: 1.1428611476879496\n",
"Epoch: 8, Weight: 1.9884010603044935, Bias: 1.114041091046188\n",
"Epoch: 9, Weight: 2.00109188391094, Bias: 1.0858410400163774\n",
"Epoch: 10, Weight: 2.013506362399409, Bias: 1.0582502885021143\n",
"Epoch: 11, Weight: 2.025651829051381, Bias: 1.0312562891981158\n",
"Epoch: 12, Weight: 2.0375344218778904, Bias: 1.0048462799719227\n",
"Epoch: 13, Weight: 2.049159902896563, Bias: 0.9790076607220877\n",
"Epoch: 14, Weight: 2.0605338524053214, Bias: 0.9537280771945735\n",
"Epoch: 15, Weight: 2.0716617165829674, Bias: 0.9289954361497844\n",
"Epoch: 16, Weight: 2.0825488206358695, Bias: 0.9047979045214405\n",
"Epoch: 17, Weight: 2.0932003738187874, Bias: 0.8811239049150731\n",
"Epoch: 18, Weight: 2.1036214725074394, Bias: 0.8579621103408925\n",
"Epoch: 19, Weight: 2.113817102773028, Bias: 0.8353014388581537\n",
"Epoch: 20, Weight: 2.12379214279979, Bias: 0.813131048287889\n",
"Epoch: 21, Weight: 2.1335513652264915, Bias: 0.7914403310288943\n",
"Epoch: 22, Weight: 2.14309943943169, Bias: 0.7702189089832662\n",
"Epoch: 23, Weight: 2.152440933768261, Bias: 0.7494566285911066\n",
"Epoch: 24, Weight: 2.1615803177492436, Bias: 0.7291435559725168\n",
"Epoch: 25, Weight: 2.1705219641863556, Bias: 0.7092699721746657\n",
"Epoch: 26, Weight: 2.17927015128219, Bias: 0.6898263685216784\n",
"Epoch: 27, Weight: 2.187829064677166, Bias: 0.6708034420651353\n",
"Epoch: 28, Weight: 2.1962027994521764, Bias: 0.6521920911330035\n",
"Epoch: 29, Weight: 2.2043953620879164, Bias: 0.6339834109748658\n",
"Epoch: 30, Weight: 2.2124106723818167, Bias: 0.6161686895013723\n",
"Epoch: 31, Weight: 2.220252565323498, Bias: 0.5987394031158727\n",
"Epoch: 32, Weight: 2.2279247929296604, Bias: 0.5816872126362319\n",
"Epoch: 33, Weight: 2.235431026039253, Bias: 0.5650039593048809\n",
"Epoch: 34, Weight: 2.242774856069832, Bias: 0.548681660885195\n",
"Epoch: 35, Weight: 2.249959796735887, Bias: 0.5327125078423264\n",
"Epoch: 36, Weight: 2.2569892857300236, Bias: 0.5170888596066717\n",
"Epoch: 37, Weight: 2.263866686367739, Bias: 0.5018032409181794\n",
"Epoch: 38, Weight: 2.270595289196635, Bias: 0.48684833824974955\n",
"Epoch: 39, Weight: 2.277178313570807, Bias: 0.47221699630802116\n",
"Epoch: 40, Weight: 2.2836189091911714, Bias: 0.45790221460986313\n",
"Epoch: 41, Weight: 2.2899201576124573, Bias: 0.44389714413293574\n",
"Epoch: 42, Weight: 2.296085073717612, Bias: 0.43019508403872325\n",
"Epoch: 43, Weight: 2.302116607160288, Bias: 0.4167894784664617\n",
"Epoch: 44, Weight: 2.3080176437761284, Bias: 0.403673913396435\n",
"Epoch: 45, Weight: 2.313791006963528, Bias: 0.3908421135811379\n",
"Epoch: 46, Weight: 2.3194394590344953, Bias: 0.37828793954283013\n",
"Epoch: 47, Weight: 2.324965702536317, Bias: 0.3660053846360565\n",
"Epoch: 48, Weight: 2.3303723815446054, Bias: 0.35398857217371676\n",
"Epoch: 49, Weight: 2.3356620829283825, Bias: 0.34223175261532085\n",
"Epoch: 50, Weight: 2.340837337587784, Bias: 0.3307293008160708\n",
"Epoch: 51, Weight: 2.34590062166499, Bias: 0.3194757133354673\n",
"Epoch: 52, Weight: 2.3508543577289545, Bias: 0.30846560580414606\n",
"Epoch: 53, Weight: 2.3557009159345084, Bias: 0.29769371034767944\n",
"Epoch: 54, Weight: 2.360442615156374, Bias: 0.2871548730661219\n",
"Epoch: 55, Weight: 2.3650817240986473, Bias: 0.2768440515680847\n",
"Epoch: 56, Weight: 2.3696204623802815, Bias: 0.2667563125581632\n",
"Epoch: 57, Weight: 2.374061001597085, Bias: 0.25688682947656066\n",
"Epoch: 58, Weight: 2.378405466360725, Bias: 0.24723088018978473\n",
"Epoch: 59, Weight: 2.3826559353152836, Bias: 0.23778384473130248\n",
"Epoch: 60, Weight: 2.386814442131792, Bias: 0.2285412030910814\n",
"Epoch: 61, Weight: 2.39088297648125, Bias: 0.21949853305295314\n",
"Epoch: 62, Weight: 2.394863484986615, Bias: 0.21065150807877053\n",
"Epoch: 63, Weight: 2.39875787215417, Bias: 0.20199589523833816\n",
"Epoch: 64, Weight: 2.402568001284751, Bias: 0.19352755318413484\n",
"Epoch: 65, Weight: 2.4062956953652592, Bias: 0.18524243016985129\n",
"Epoch: 66, Weight: 2.4099427379408818, Bias: 0.1771365621118013\n",
"Epoch: 67, Weight: 2.4135108739684457, Bias: 0.16920607069227345\n",
"Epoch: 68, Weight: 2.4170018106513123, Bias: 0.16144716150391786\n",
"Epoch: 69, Weight: 2.4204172182561887, Bias: 0.15385612223428083\n",
"Epoch: 70, Weight: 2.42375873091229, Bias: 0.14642932088962124\n",
"Epoch: 71, Weight: 2.4270279473932006, Bias: 0.1391632040571527\n",
"Epoch: 72, Weight: 2.4302264318818154, Bias: 0.13205429520488576\n",
"Epoch: 73, Weight: 2.4333557147187443, Bias: 0.1250991930182516\n",
"Epoch: 74, Weight: 2.4364172931345047, Bias: 0.11829456977271688\n",
"Epoch: 75, Weight: 2.4394126319658955, Bias: 0.11163716974160745\n",
"Epoch: 76, Weight: 2.442343164356852, Bias: 0.10512380763837732\n",
"Epoch: 77, Weight: 2.445210292444143, Bias: 0.09875136709258651\n",
"Epoch: 78, Weight: 2.4480153880282467, Bias: 0.09251679915884697\n",
"Epoch: 79, Weight: 2.4507597932296843, Bias: 0.08641712085802802\n",
"Epoch: 80, Weight: 2.4534448211311806, Bias: 0.08044941375002966\n",
"Epoch: 81, Weight: 2.4560717564059136, Bias: 0.07461082253742772\n",
"Epoch: 82, Weight: 2.458641855932173, Bias: 0.0688985536993371\n",
"Epoch: 83, Weight: 2.461156349394732, Bias: 0.06330987415482785\n",
"Epoch: 84, Weight: 2.463616439873196, Bias: 0.05784210995526014\n",
"Epoch: 85, Weight: 2.4660233044176216, Bias: 0.05249264500491319\n",
"Epoch: 86, Weight: 2.4683780946116927, Bias: 0.04725891980929251\n",
"Epoch: 87, Weight: 2.470681937123703, Bias: 0.04213843025051945\n",
"Epoch: 88, Weight: 2.4729359342456143, Bias: 0.0371287263892152\n",
"Epoch: 89, Weight: 2.4751411644204615, Bias: 0.03222741129230891\n",
"Epoch: 90, Weight: 2.47729868275835, Bias: 0.027432139886204847\n",
"Epoch: 91, Weight: 2.4794095215412577, Bias: 0.02274061783476307\n",
"Epoch: 92, Weight: 2.48147469071697, Bias: 0.01815060044155677\n",
"Epoch: 93, Weight: 2.483495178382273, Bias: 0.013659891575878064\n",
"Epoch: 94, Weight: 2.485471951255725, Bias: 0.009266342621981876\n",
"Epoch: 95, Weight: 2.4874059551401957, Bias: 0.0049678514510644945\n",
"Epoch: 96, Weight: 2.4892981153753846, Bias: 0.0007623614154813733\n",
"Epoch: 97, Weight: 2.491149337280562, Bias: -0.003352139635270969\n",
"Epoch: 98, Weight: 2.492960506587724, Bias: -0.007377620317284189\n",
"Epoch: 99, Weight: 2.4947324898653815, Bias: -0.011316006654140311\n"
]
}
],
"source": [
"\n",
"for e in range(epoch):\n",
" for i in range(len(data)):\n",
" tahmin = fiyat_tahmin_et(data.loc[i]['living_area_renov'])\n",
" hata = hata_hesapla(data.loc[i]['living_area_renov'], data.loc[i]['Price'])\n",
" notater = data.loc[i]['Notater']\n",
" notater_katsayisi = 1\n",
" if notater == 'Real Estate Expert':\n",
" notater_katsayisi = 5\n",
" elif notater == 'other':\n",
" notater_katsayisi = 2\n",
" weight = weight + hata * learning_rate * data.loc[i]['living_area_renov'] * notater_katsayisi\n",
" bias = bias + hata * learning_rate * notater_katsayisi\n",
" print(f'Epoch: {e}, Weight: {weight}, Bias: {bias}')"
]
}
],
"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": "3.9.6"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment