-
Notifications
You must be signed in to change notification settings - Fork 15
/
16 실습 - 차원축소
1 lines (1 loc) · 60 KB
/
16 실습 - 차원축소
1
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"16 실습 - 차원축소","provenance":[],"collapsed_sections":[]},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"markdown","metadata":{"id":"4di_nmaFk0Zu"},"source":["# PCA (주성분 분석)"]},{"cell_type":"markdown","metadata":{"id":"-eRv2nFqlMAZ"},"source":["* PCA는 데이터의 분산(variance)을 최대로 하는 축을 찾는 것을 반복하여, 고차원 공간의 표본들을 저차원 공간으로 변환하는 기법\n","\n","\n"]},{"cell_type":"markdown","metadata":{"id":"n45Y8uyUlGMJ"},"source":[""]},{"cell_type":"markdown","metadata":{"id":"ebumLhy25esv"},"source":["## 1 차원 축소의 예\n","\n","변수 갯수를 차원이라고 생각할 수 있음"]},{"cell_type":"code","metadata":{"id":"kSe1NKhQlG9f","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490557315,"user_tz":-540,"elapsed":376,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"e3d21063-eaf7-44af-b067-c5696f255e77"},"source":["국어 = 80\n","영어 = 60\n","수학 = 90\n","과학 = 70\n","\n","평균 = (국어 + 영어 + 수학 + 과학)/4 # 5개 차원을 1개 차원으로 축수\n","\n","평균"],"execution_count":42,"outputs":[{"output_type":"execute_result","data":{"text/plain":["75.0"]},"metadata":{},"execution_count":42}]},{"cell_type":"code","metadata":{"id":"wWnVJ9vK5YY9","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490588404,"user_tz":-540,"elapsed":256,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"dd8c43e0-9048-4417-bfa5-b715da4db9d7"},"source":["몸무게 = 70\n","키 = 180\n","\n","BMI = 몸무게 / (키*키) # 2개 차원을 1개 차원으로 축소\n","\n","BMI "],"execution_count":43,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.0021604938271604936"]},"metadata":{},"execution_count":43}]},{"cell_type":"markdown","metadata":{"id":"zc4GQj2oNn3G"},"source":["\n","\n","---\n","\n"]},{"cell_type":"markdown","metadata":{"id":"c_viMcGaNpEN"},"source":["## 2 PCA 예제"]},{"cell_type":"markdown","metadata":{"id":"EIDZeP-lOB52"},"source":["### 1) 예제 데이터"]},{"cell_type":"code","metadata":{"id":"zwp8In2uNWgi","colab":{"base_uri":"https://localhost:8080/","height":205},"executionInfo":{"status":"ok","timestamp":1637490614823,"user_tz":-540,"elapsed":337,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"c05923eb-2aba-468d-de22-86d29d648242"},"source":["import pandas as pd\n","# Eating, exercise habbit and their body shape\n","df = pd.DataFrame(columns=['calory', 'breakfast', 'lunch', 'dinner', 'exercise', 'body_shape'])\n","\n","df.loc[0] = [1200, 1, 0, 0, 2, 'Skinny']\n","df.loc[1] = [2800, 1, 1, 1, 1, 'Normal']\n","df.loc[2] = [3500, 2, 2, 1, 0, 'Fat']\n","df.loc[3] = [1400, 0, 1, 0, 3, 'Skinny']\n","df.loc[4] = [5000, 2, 2, 2, 0, 'Fat']\n","df.loc[5] = [1300, 0, 0, 1, 2, 'Skinny']\n","df.loc[6] = [3000, 1, 0, 1, 1, 'Normal']\n","df.loc[7] = [4000, 2, 2, 2, 0, 'Fat']\n","df.loc[8] = [2600, 0, 2, 0, 0, 'Normal']\n","df.loc[9] = [3000, 1, 2, 1, 1, 'Fat']\n","\n","df.head()"],"execution_count":44,"outputs":[{"output_type":"execute_result","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>calory</th>\n"," <th>breakfast</th>\n"," <th>lunch</th>\n"," <th>dinner</th>\n"," <th>exercise</th>\n"," <th>body_shape</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1200</td>\n"," <td>1</td>\n"," <td>0</td>\n"," <td>0</td>\n"," <td>2</td>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2800</td>\n"," <td>1</td>\n"," <td>1</td>\n"," <td>1</td>\n"," <td>1</td>\n"," <td>Normal</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3500</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>1</td>\n"," <td>0</td>\n"," <td>Fat</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>1400</td>\n"," <td>0</td>\n"," <td>1</td>\n"," <td>0</td>\n"," <td>3</td>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5000</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>0</td>\n"," <td>Fat</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" calory breakfast lunch dinner exercise body_shape\n","0 1200 1 0 0 2 Skinny\n","1 2800 1 1 1 1 Normal\n","2 3500 2 2 1 0 Fat\n","3 1400 0 1 0 3 Skinny\n","4 5000 2 2 2 0 Fat"]},"metadata":{},"execution_count":44}]},{"cell_type":"markdown","metadata":{"id":"RU435e2UNWgn"},"source":["### 2) vector 추출"]},{"cell_type":"code","metadata":{"id":"aTsVKtciNWgn","colab":{"base_uri":"https://localhost:8080/","height":205},"executionInfo":{"status":"ok","timestamp":1637490290093,"user_tz":-540,"elapsed":26,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"e1b218ec-45cb-4614-df58-4311a3b1be7c"},"source":["# X is feature vectors\n","X = df[['calory', 'breakfast', 'lunch', 'dinner', 'exercise']]\n","X.head()"],"execution_count":24,"outputs":[{"output_type":"execute_result","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>calory</th>\n"," <th>breakfast</th>\n"," <th>lunch</th>\n"," <th>dinner</th>\n"," <th>exercise</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1200</td>\n"," <td>1</td>\n"," <td>0</td>\n"," <td>0</td>\n"," <td>2</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2800</td>\n"," <td>1</td>\n"," <td>1</td>\n"," <td>1</td>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3500</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>1</td>\n"," <td>0</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>1400</td>\n"," <td>0</td>\n"," <td>1</td>\n"," <td>0</td>\n"," <td>3</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5000</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>2</td>\n"," <td>0</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" calory breakfast lunch dinner exercise\n","0 1200 1 0 0 2\n","1 2800 1 1 1 1\n","2 3500 2 2 1 0\n","3 1400 0 1 0 3\n","4 5000 2 2 2 0"]},"metadata":{},"execution_count":24}]},{"cell_type":"code","metadata":{"id":"W7pLlFliNWgq","colab":{"base_uri":"https://localhost:8080/","height":205},"executionInfo":{"status":"ok","timestamp":1637490290093,"user_tz":-540,"elapsed":25,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"4199f40e-1e6f-4887-f67a-c6a51b2ea4db"},"source":["# Y is labels\n","Y = df[['body_shape']]\n","Y.head()"],"execution_count":25,"outputs":[{"output_type":"execute_result","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>body_shape</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>Normal</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>Fat</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>Fat</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" body_shape\n","0 Skinny\n","1 Normal\n","2 Fat\n","3 Skinny\n","4 Fat"]},"metadata":{},"execution_count":25}]},{"cell_type":"markdown","metadata":{"id":"_-jESafVNWgv"},"source":["### 3) feature vector 표준화"]},{"cell_type":"code","metadata":{"id":"jTYWHiC6NWgv","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290094,"user_tz":-540,"elapsed":25,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"4149b022-f3d2-406b-f246-8bdc7e74c494"},"source":["from sklearn.preprocessing import StandardScaler\n","x_std = StandardScaler().fit_transform(X)\n","\n","x_std"],"execution_count":26,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[-1.35205803, 0. , -1.3764944 , -1.28571429, 1. ],\n"," [ 0.01711466, 0. , -0.22941573, 0.14285714, 0. ],\n"," [ 0.61612771, 1.29099445, 0.91766294, 0.14285714, -1. ],\n"," [-1.18091145, -1.29099445, -0.22941573, -1.28571429, 2. ],\n"," [ 1.89972711, 1.29099445, 0.91766294, 1.57142857, -1. ],\n"," [-1.26648474, -1.29099445, -1.3764944 , 0.14285714, 1. ],\n"," [ 0.18826125, 0. , -1.3764944 , 0.14285714, 0. ],\n"," [ 1.04399418, 1.29099445, 0.91766294, 1.57142857, -1. ],\n"," [-0.15403193, -1.29099445, 0.91766294, -1.28571429, -1. ],\n"," [ 0.18826125, 0. , 0.91766294, 0.14285714, 0. ]])"]},"metadata":{},"execution_count":26}]},{"cell_type":"code","metadata":{"id":"g_KzSYIMNWgx","executionInfo":{"status":"ok","timestamp":1637490290094,"user_tz":-540,"elapsed":23,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}}},"source":["StandardScaler?"],"execution_count":27,"outputs":[]},{"cell_type":"code","metadata":{"id":"2Y5sAh8VOiK4","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290248,"user_tz":-540,"elapsed":177,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"74e86ea4-8030-4acd-af31-4c52c0751b0d"},"source":["print(x_std.mean())\n","print(x_std.std())"],"execution_count":28,"outputs":[{"output_type":"stream","name":"stdout","text":["3.552713678800501e-17\n","1.0\n"]}]},{"cell_type":"code","metadata":{"id":"tVAlWTMC7zXF","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490791671,"user_tz":-540,"elapsed":276,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"52273b66-12ff-48b9-8dfc-7efc352ebda6"},"source":["x_std"],"execution_count":45,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[-1.35205803, 0. , -1.3764944 , -1.28571429, 1. ],\n"," [ 0.01711466, 0. , -0.22941573, 0.14285714, 0. ],\n"," [ 0.61612771, 1.29099445, 0.91766294, 0.14285714, -1. ],\n"," [-1.18091145, -1.29099445, -0.22941573, -1.28571429, 2. ],\n"," [ 1.89972711, 1.29099445, 0.91766294, 1.57142857, -1. ],\n"," [-1.26648474, -1.29099445, -1.3764944 , 0.14285714, 1. ],\n"," [ 0.18826125, 0. , -1.3764944 , 0.14285714, 0. ],\n"," [ 1.04399418, 1.29099445, 0.91766294, 1.57142857, -1. ],\n"," [-0.15403193, -1.29099445, 0.91766294, -1.28571429, -1. ],\n"," [ 0.18826125, 0. , 0.91766294, 0.14285714, 0. ]])"]},"metadata":{},"execution_count":45}]},{"cell_type":"code","metadata":{"id":"q6UAaEze75Sq","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490795451,"user_tz":-540,"elapsed":267,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"8a3b04bf-f0e1-44cf-938b-b14289cf6394"},"source":["x_std.T"],"execution_count":46,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[-1.35205803, 0.01711466, 0.61612771, -1.18091145, 1.89972711,\n"," -1.26648474, 0.18826125, 1.04399418, -0.15403193, 0.18826125],\n"," [ 0. , 0. , 1.29099445, -1.29099445, 1.29099445,\n"," -1.29099445, 0. , 1.29099445, -1.29099445, 0. ],\n"," [-1.3764944 , -0.22941573, 0.91766294, -0.22941573, 0.91766294,\n"," -1.3764944 , -1.3764944 , 0.91766294, 0.91766294, 0.91766294],\n"," [-1.28571429, 0.14285714, 0.14285714, -1.28571429, 1.57142857,\n"," 0.14285714, 0.14285714, 1.57142857, -1.28571429, 0.14285714],\n"," [ 1. , 0. , -1. , 2. , -1. ,\n"," 1. , 0. , -1. , -1. , 0. ]])"]},"metadata":{},"execution_count":46}]},{"cell_type":"markdown","metadata":{"id":"S0OJ6rcXNWgz"},"source":["### 4) feature 간 공분산 행렬계산"]},{"cell_type":"code","metadata":{"id":"EsRD5rBxNWg1","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290249,"user_tz":-540,"elapsed":173,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"f449bd24-4307-4390-f38e-49f92a09e84d"},"source":["import numpy as np\n","\n","# features are columns from x_std\n","features = x_std.T \n","covariance_matrix = np.cov(features)\n","\n","\n","\n","print(covariance_matrix)"],"execution_count":31,"outputs":[{"output_type":"stream","name":"stdout","text":["[[ 1.11111111 0.88379717 0.76782385 0.89376551 -0.93179808]\n"," [ 0.88379717 1.11111111 0.49362406 0.81967902 -0.71721914]\n"," [ 0.76782385 0.49362406 1.11111111 0.40056715 -0.76471911]\n"," [ 0.89376551 0.81967902 0.40056715 1.11111111 -0.63492063]\n"," [-0.93179808 -0.71721914 -0.76471911 -0.63492063 1.11111111]]\n"]}]},{"cell_type":"code","metadata":{"id":"dcCx3i7e8ihj","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490842287,"user_tz":-540,"elapsed":287,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"1db28d4a-7b4a-40b3-ee92-b47d3845e7ba"},"source":["x_std.T.dot(x_std)/9"],"execution_count":47,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[ 1.11111111, 0.88379717, 0.76782385, 0.89376551, -0.93179808],\n"," [ 0.88379717, 1.11111111, 0.49362406, 0.81967902, -0.71721914],\n"," [ 0.76782385, 0.49362406, 1.11111111, 0.40056715, -0.76471911],\n"," [ 0.89376551, 0.81967902, 0.40056715, 1.11111111, -0.63492063],\n"," [-0.93179808, -0.71721914, -0.76471911, -0.63492063, 1.11111111]])"]},"metadata":{},"execution_count":47}]},{"cell_type":"markdown","metadata":{"id":"fPQTE54yNWg2"},"source":["### 5) 고유 벡터(Eigen Vectors)와 고유값(Eigen Values) 산출"]},{"cell_type":"code","metadata":{"id":"SZM7B525NWg3","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290249,"user_tz":-540,"elapsed":170,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"f9e71730-56fd-4bf8-f58d-dd84c62435a9"},"source":["eig_vals, eig_vecs = np.linalg.eig(covariance_matrix)\n","print('Eigenvalues \\n%s' %eig_vals)\n","print('Eigenvectors \\n%s' %eig_vecs)"],"execution_count":33,"outputs":[{"output_type":"stream","name":"stdout","text":["Eigenvalues \n","[4.0657343 0.8387565 0.07629538 0.27758568 0.2971837 ]\n","Eigenvectors \n","[[-0.508005 -0.0169937 -0.84711404 0.11637853 0.10244985]\n"," [-0.44660335 -0.36890361 0.12808055 -0.63112016 -0.49973822]\n"," [-0.38377913 0.70804084 0.20681005 -0.40305226 0.38232213]\n"," [-0.42845209 -0.53194699 0.3694462 0.22228235 0.58954327]\n"," [ 0.46002038 -0.2816592 -0.29450345 -0.61341895 0.49601841]]\n"]}]},{"cell_type":"code","metadata":{"id":"gswIA3qUNWg7","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290250,"user_tz":-540,"elapsed":169,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"b91241ed-742b-463c-add5-3749c47a51f1"},"source":["# 1개 차원으로 축소했을때 73%의 정보를 보유\n","eig_vals[0] / sum(eig_vals)"],"execution_count":34,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.7318321731427545"]},"metadata":{},"execution_count":34}]},{"cell_type":"markdown","metadata":{"id":"Iz3WqbWdNWg9"},"source":["### 6) 고유벡터(Eigen Vector)를 기준으로 데이터를 투영"]},{"cell_type":"code","metadata":{"id":"LVITokrDEOos","executionInfo":{"status":"ok","timestamp":1637490950034,"user_tz":-540,"elapsed":261,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"284037fb-35fc-43f2-817e-0d8e20df49a2","colab":{"base_uri":"https://localhost:8080/"}},"source":["eig_vecs.T[0]"],"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([-0.508005 , -0.44660335, -0.38377913, -0.42845209, 0.46002038])"]},"metadata":{},"execution_count":48}]},{"cell_type":"code","metadata":{"id":"g1Z4VErxNWg9","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290250,"user_tz":-540,"elapsed":167,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"de4ac759-4b6c-497a-d8f9-6920d201c3fe"},"source":["projected_X = x_std.dot(eig_vecs.T[0])\n","projected_X"],"execution_count":35,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ 2.22600943, 0.0181432 , -1.76296611, 2.73542407, -3.02711544,\n"," 2.14702579, 0.37142473, -2.59239883, 0.39347815, -0.50902498])"]},"metadata":{},"execution_count":35}]},{"cell_type":"code","metadata":{"id":"-ZBMM62kNWhJ","colab":{"base_uri":"https://localhost:8080/","height":205},"executionInfo":{"status":"ok","timestamp":1637490290250,"user_tz":-540,"elapsed":166,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"9367686c-830d-4637-cf2c-43073c4fa8c9"},"source":["result = pd.DataFrame(projected_X, columns=['PC1'])\n","result['y-axis'] = 0.0\n","result['label'] = Y\n","result.head()"],"execution_count":36,"outputs":[{"output_type":"execute_result","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>PC1</th>\n"," <th>y-axis</th>\n"," <th>label</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>2.226009</td>\n"," <td>0.0</td>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>0.018143</td>\n"," <td>0.0</td>\n"," <td>Normal</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>-1.762966</td>\n"," <td>0.0</td>\n"," <td>Fat</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>2.735424</td>\n"," <td>0.0</td>\n"," <td>Skinny</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>-3.027115</td>\n"," <td>0.0</td>\n"," <td>Fat</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" PC1 y-axis label\n","0 2.226009 0.0 Skinny\n","1 0.018143 0.0 Normal\n","2 -1.762966 0.0 Fat\n","3 2.735424 0.0 Skinny\n","4 -3.027115 0.0 Fat"]},"metadata":{},"execution_count":36}]},{"cell_type":"code","metadata":{"id":"qpgqiL4QNWhM","colab":{"base_uri":"https://localhost:8080/","height":458},"executionInfo":{"status":"ok","timestamp":1637490290958,"user_tz":-540,"elapsed":873,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"56f69778-3260-4e31-fab7-475a4085a82e"},"source":["import matplotlib.pyplot as plt\n","import seaborn as sns\n","%matplotlib inline\n","\n","sns.lmplot('PC1', 'y-axis', data=result, fit_reg=False, # x-axis, y-axis, data, no line\n"," scatter_kws={\"s\": 50}, # marker size\n"," hue=\"label\") # color\n","\n","# title\n","plt.title('PCA result')"],"execution_count":37,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n"," FutureWarning\n"]},{"output_type":"execute_result","data":{"text/plain":["Text(0.5, 1.0, 'PCA result')"]},"metadata":{},"execution_count":37},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432.125x360 with 1 Axes>"]},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"ZJ6gr23YPhfr"},"source":["\n","\n","---\n","\n"]},{"cell_type":"markdown","metadata":{"id":"VhprUX7gNWhP"},"source":["## 3 scikit-learn 활용 PCA"]},{"cell_type":"code","metadata":{"id":"8hNAo6pDNWhQ","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490290959,"user_tz":-540,"elapsed":7,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"720f2c91-ccad-416f-f101-1fccd0697c94"},"source":["from sklearn import decomposition\n","pca = decomposition.PCA(n_components=1)\n","sklearn_pca_x = pca.fit_transform(x_std)\n","sklearn_pca_x"],"execution_count":38,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[-2.22600943],\n"," [-0.0181432 ],\n"," [ 1.76296611],\n"," [-2.73542407],\n"," [ 3.02711544],\n"," [-2.14702579],\n"," [-0.37142473],\n"," [ 2.59239883],\n"," [-0.39347815],\n"," [ 0.50902498]])"]},"metadata":{},"execution_count":38}]},{"cell_type":"code","metadata":{"id":"uMdCh2hFNWhR","colab":{"base_uri":"https://localhost:8080/","height":442},"executionInfo":{"status":"ok","timestamp":1637490291260,"user_tz":-540,"elapsed":305,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"b13f0765-7e63-4793-cb0d-bfc970cb3caf"},"source":["sklearn_result = pd.DataFrame(sklearn_pca_x, columns=['PC1'])\n","sklearn_result['y-axis'] = 0.0\n","sklearn_result['label'] = Y\n","\n","sns.lmplot('PC1', 'y-axis', data=sklearn_result, fit_reg=False, # x-axis, y-axis, data, no line\n"," scatter_kws={\"s\": 50}, # marker size\n"," hue=\"label\") # color"],"execution_count":39,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n"," FutureWarning\n"]},{"output_type":"execute_result","data":{"text/plain":["<seaborn.axisgrid.FacetGrid at 0x7f502f1aac90>"]},"metadata":{},"execution_count":39},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432.125x360 with 1 Axes>"]},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"ZRbZndmmcqMe","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1637490291261,"user_tz":-540,"elapsed":8,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"d6040b30-0651-4f27-f115-a908e5a925df"},"source":["from sklearn import decomposition\n","pca = decomposition.PCA(n_components=2)\n","sklearn_pca_x = pca.fit_transform(x_std)\n","sklearn_pca_x"],"execution_count":40,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[-2.22600943, -0.54936514],\n"," [-0.0181432 , -0.23871898],\n"," [ 1.76296611, 0.36868681],\n"," [-2.73542407, 0.4544983 ],\n"," [ 3.02711544, -0.41305057],\n"," [-2.14702579, -0.83449111],\n"," [-0.37142473, -1.05380594],\n"," [ 2.59239883, -0.3985085 ],\n"," [-0.39347815, 2.09420397],\n"," [ 0.50902498, 0.57055116]])"]},"metadata":{},"execution_count":40}]},{"cell_type":"code","metadata":{"id":"MVe_ILkacs95","colab":{"base_uri":"https://localhost:8080/","height":442},"executionInfo":{"status":"ok","timestamp":1637490291950,"user_tz":-540,"elapsed":435,"user":{"displayName":"이민호","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GiFPPatrtQJJCEfMd6D3DoTVRog9gVm7Ovj5Lex=s64","userId":"15829449822908558555"}},"outputId":"ee7b55a1-b7a9-4e9c-d63f-8a6a8e110b41"},"source":["sklearn_result = pd.DataFrame(sklearn_pca_x, columns=['PC1', 'PC2'])\n","sklearn_result['label'] = Y\n","\n","sns.lmplot('PC1', 'PC2', data=sklearn_result, fit_reg=False, # x-axis, y-axis, data, no line\n"," scatter_kws={\"s\": 50}, # marker size\n"," hue=\"label\") # color"],"execution_count":41,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n"," FutureWarning\n"]},{"output_type":"execute_result","data":{"text/plain":["<seaborn.axisgrid.FacetGrid at 0x7f502a92a210>"]},"metadata":{},"execution_count":41},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432.125x360 with 1 Axes>"]},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"qxxHu8ooPgJ0"},"source":["\n","\n","---\n","\n"]}]}