├── PCA.ipynb
└── README.md


/PCA.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "# Principal Component Analysis(PCA)"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 1,
 13 |    "metadata": {},
 14 |    "outputs": [],
 15 |    "source": [
 16 |     "import matplotlib.pyplot as plt\n",
 17 |     "import numpy as np\n",
 18 |     "import pandas as pd\n",
 19 |     "%matplotlib inline"
 20 |    ]
 21 |   },
 22 |   {
 23 |    "cell_type": "code",
 24 |    "execution_count": 2,
 25 |    "metadata": {},
 26 |    "outputs": [],
 27 |    "source": [
 28 |     "from sklearn.datasets import load_breast_cancer"
 29 |    ]
 30 |   },
 31 |   {
 32 |    "cell_type": "code",
 33 |    "execution_count": 3,
 34 |    "metadata": {},
 35 |    "outputs": [],
 36 |    "source": [
 37 |     "cancer=load_breast_cancer()"
 38 |    ]
 39 |   },
 40 |   {
 41 |    "cell_type": "code",
 42 |    "execution_count": 4,
 43 |    "metadata": {},
 44 |    "outputs": [
 45 |     {
 46 |      "data": {
 47 |       "text/plain": [
 48 |        "dict_keys(['data', 'target', 'target_names', 'DESCR', 'feature_names'])"
 49 |       ]
 50 |      },
 51 |      "execution_count": 4,
 52 |      "metadata": {},
 53 |      "output_type": "execute_result"
 54 |     }
 55 |    ],
 56 |    "source": [
 57 |     "cancer.keys()"
 58 |    ]
 59 |   },
 60 |   {
 61 |    "cell_type": "code",
 62 |    "execution_count": 5,
 63 |    "metadata": {},
 64 |    "outputs": [
 65 |     {
 66 |      "name": "stdout",
 67 |      "output_type": "stream",
 68 |      "text": [
 69 |       "Breast Cancer Wisconsin (Diagnostic) Database\n",
 70 |       "=============================================\n",
 71 |       "\n",
 72 |       "Notes\n",
 73 |       "-----\n",
 74 |       "Data Set Characteristics:\n",
 75 |       "    :Number of Instances: 569\n",
 76 |       "\n",
 77 |       "    :Number of Attributes: 30 numeric, predictive attributes and the class\n",
 78 |       "\n",
 79 |       "    :Attribute Information:\n",
 80 |       "        - radius (mean of distances from center to points on the perimeter)\n",
 81 |       "        - texture (standard deviation of gray-scale values)\n",
 82 |       "        - perimeter\n",
 83 |       "        - area\n",
 84 |       "        - smoothness (local variation in radius lengths)\n",
 85 |       "        - compactness (perimeter^2 / area - 1.0)\n",
 86 |       "        - concavity (severity of concave portions of the contour)\n",
 87 |       "        - concave points (number of concave portions of the contour)\n",
 88 |       "        - symmetry \n",
 89 |       "        - fractal dimension (\"coastline approximation\" - 1)\n",
 90 |       "\n",
 91 |       "        The mean, standard error, and \"worst\" or largest (mean of the three\n",
 92 |       "        largest values) of these features were computed for each image,\n",
 93 |       "        resulting in 30 features.  For instance, field 3 is Mean Radius, field\n",
 94 |       "        13 is Radius SE, field 23 is Worst Radius.\n",
 95 |       "\n",
 96 |       "        - class:\n",
 97 |       "                - WDBC-Malignant\n",
 98 |       "                - WDBC-Benign\n",
 99 |       "\n",
100 |       "    :Summary Statistics:\n",
101 |       "\n",
102 |       "    ===================================== ====== ======\n",
103 |       "                                           Min    Max\n",
104 |       "    ===================================== ====== ======\n",
105 |       "    radius (mean):                        6.981  28.11\n",
106 |       "    texture (mean):                       9.71   39.28\n",
107 |       "    perimeter (mean):                     43.79  188.5\n",
108 |       "    area (mean):                          143.5  2501.0\n",
109 |       "    smoothness (mean):                    0.053  0.163\n",
110 |       "    compactness (mean):                   0.019  0.345\n",
111 |       "    concavity (mean):                     0.0    0.427\n",
112 |       "    concave points (mean):                0.0    0.201\n",
113 |       "    symmetry (mean):                      0.106  0.304\n",
114 |       "    fractal dimension (mean):             0.05   0.097\n",
115 |       "    radius (standard error):              0.112  2.873\n",
116 |       "    texture (standard error):             0.36   4.885\n",
117 |       "    perimeter (standard error):           0.757  21.98\n",
118 |       "    area (standard error):                6.802  542.2\n",
119 |       "    smoothness (standard error):          0.002  0.031\n",
120 |       "    compactness (standard error):         0.002  0.135\n",
121 |       "    concavity (standard error):           0.0    0.396\n",
122 |       "    concave points (standard error):      0.0    0.053\n",
123 |       "    symmetry (standard error):            0.008  0.079\n",
124 |       "    fractal dimension (standard error):   0.001  0.03\n",
125 |       "    radius (worst):                       7.93   36.04\n",
126 |       "    texture (worst):                      12.02  49.54\n",
127 |       "    perimeter (worst):                    50.41  251.2\n",
128 |       "    area (worst):                         185.2  4254.0\n",
129 |       "    smoothness (worst):                   0.071  0.223\n",
130 |       "    compactness (worst):                  0.027  1.058\n",
131 |       "    concavity (worst):                    0.0    1.252\n",
132 |       "    concave points (worst):               0.0    0.291\n",
133 |       "    symmetry (worst):                     0.156  0.664\n",
134 |       "    fractal dimension (worst):            0.055  0.208\n",
135 |       "    ===================================== ====== ======\n",
136 |       "\n",
137 |       "    :Missing Attribute Values: None\n",
138 |       "\n",
139 |       "    :Class Distribution: 212 - Malignant, 357 - Benign\n",
140 |       "\n",
141 |       "    :Creator:  Dr. William H. Wolberg, W. Nick Street, Olvi L. Mangasarian\n",
142 |       "\n",
143 |       "    :Donor: Nick Street\n",
144 |       "\n",
145 |       "    :Date: November, 1995\n",
146 |       "\n",
147 |       "This is a copy of UCI ML Breast Cancer Wisconsin (Diagnostic) datasets.\n",
148 |       "https://goo.gl/U2Uwz2\n",
149 |       "\n",
150 |       "Features are computed from a digitized image of a fine needle\n",
151 |       "aspirate (FNA) of a breast mass.  They describe\n",
152 |       "characteristics of the cell nuclei present in the image.\n",
153 |       "\n",
154 |       "Separating plane described above was obtained using\n",
155 |       "Multisurface Method-Tree (MSM-T) [K. P. Bennett, \"Decision Tree\n",
156 |       "Construction Via Linear Programming.\" Proceedings of the 4th\n",
157 |       "Midwest Artificial Intelligence and Cognitive Science Society,\n",
158 |       "pp. 97-101, 1992], a classification method which uses linear\n",
159 |       "programming to construct a decision tree.  Relevant features\n",
160 |       "were selected using an exhaustive search in the space of 1-4\n",
161 |       "features and 1-3 separating planes.\n",
162 |       "\n",
163 |       "The actual linear program used to obtain the separating plane\n",
164 |       "in the 3-dimensional space is that described in:\n",
165 |       "[K. P. Bennett and O. L. Mangasarian: \"Robust Linear\n",
166 |       "Programming Discrimination of Two Linearly Inseparable Sets\",\n",
167 |       "Optimization Methods and Software 1, 1992, 23-34].\n",
168 |       "\n",
169 |       "This database is also available through the UW CS ftp server:\n",
170 |       "\n",
171 |       "ftp ftp.cs.wisc.edu\n",
172 |       "cd math-prog/cpo-dataset/machine-learn/WDBC/\n",
173 |       "\n",
174 |       "References\n",
175 |       "----------\n",
176 |       "   - W.N. Street, W.H. Wolberg and O.L. Mangasarian. Nuclear feature extraction \n",
177 |       "     for breast tumor diagnosis. IS&T/SPIE 1993 International Symposium on \n",
178 |       "     Electronic Imaging: Science and Technology, volume 1905, pages 861-870,\n",
179 |       "     San Jose, CA, 1993.\n",
180 |       "   - O.L. Mangasarian, W.N. Street and W.H. Wolberg. Breast cancer diagnosis and \n",
181 |       "     prognosis via linear programming. Operations Research, 43(4), pages 570-577, \n",
182 |       "     July-August 1995.\n",
183 |       "   - W.H. Wolberg, W.N. Street, and O.L. Mangasarian. Machine learning techniques\n",
184 |       "     to diagnose breast cancer from fine-needle aspirates. Cancer Letters 77 (1994) \n",
185 |       "     163-171.\n",
186 |       "\n"
187 |      ]
188 |     }
189 |    ],
190 |    "source": [
191 |     "print(cancer['DESCR'])"
192 |    ]
193 |   },
194 |   {
195 |    "cell_type": "code",
196 |    "execution_count": 6,
197 |    "metadata": {},
198 |    "outputs": [],
199 |    "source": [
200 |     "df=pd.DataFrame(cancer['data'],columns=cancer['feature_names'])"
201 |    ]
202 |   },
203 |   {
204 |    "cell_type": "code",
205 |    "execution_count": 7,
206 |    "metadata": {},
207 |    "outputs": [
208 |     {
209 |      "data": {
210 |       "text/html": [
211 |        "<div>\n",
212 |        "<style scoped>\n",
213 |        "    .dataframe tbody tr th:only-of-type {\n",
214 |        "        vertical-align: middle;\n",
215 |        "    }\n",
216 |        "\n",
217 |        "    .dataframe tbody tr th {\n",
218 |        "        vertical-align: top;\n",
219 |        "    }\n",
220 |        "\n",
221 |        "    .dataframe thead th {\n",
222 |        "        text-align: right;\n",
223 |        "    }\n",
224 |        "</style>\n",
225 |        "<table border=\"1\" class=\"dataframe\">\n",
226 |        "  <thead>\n",
227 |        "    <tr style=\"text-align: right;\">\n",
228 |        "      <th></th>\n",
229 |        "      <th>mean radius</th>\n",
230 |        "      <th>mean texture</th>\n",
231 |        "      <th>mean perimeter</th>\n",
232 |        "      <th>mean area</th>\n",
233 |        "      <th>mean smoothness</th>\n",
234 |        "      <th>mean compactness</th>\n",
235 |        "      <th>mean concavity</th>\n",
236 |        "      <th>mean concave points</th>\n",
237 |        "      <th>mean symmetry</th>\n",
238 |        "      <th>mean fractal dimension</th>\n",
239 |        "      <th>...</th>\n",
240 |        "      <th>worst radius</th>\n",
241 |        "      <th>worst texture</th>\n",
242 |        "      <th>worst perimeter</th>\n",
243 |        "      <th>worst area</th>\n",
244 |        "      <th>worst smoothness</th>\n",
245 |        "      <th>worst compactness</th>\n",
246 |        "      <th>worst concavity</th>\n",
247 |        "      <th>worst concave points</th>\n",
248 |        "      <th>worst symmetry</th>\n",
249 |        "      <th>worst fractal dimension</th>\n",
250 |        "    </tr>\n",
251 |        "  </thead>\n",
252 |        "  <tbody>\n",
253 |        "    <tr>\n",
254 |        "      <th>0</th>\n",
255 |        "      <td>17.99</td>\n",
256 |        "      <td>10.38</td>\n",
257 |        "      <td>122.80</td>\n",
258 |        "      <td>1001.0</td>\n",
259 |        "      <td>0.11840</td>\n",
260 |        "      <td>0.27760</td>\n",
261 |        "      <td>0.3001</td>\n",
262 |        "      <td>0.14710</td>\n",
263 |        "      <td>0.2419</td>\n",
264 |        "      <td>0.07871</td>\n",
265 |        "      <td>...</td>\n",
266 |        "      <td>25.38</td>\n",
267 |        "      <td>17.33</td>\n",
268 |        "      <td>184.60</td>\n",
269 |        "      <td>2019.0</td>\n",
270 |        "      <td>0.1622</td>\n",
271 |        "      <td>0.6656</td>\n",
272 |        "      <td>0.7119</td>\n",
273 |        "      <td>0.2654</td>\n",
274 |        "      <td>0.4601</td>\n",
275 |        "      <td>0.11890</td>\n",
276 |        "    </tr>\n",
277 |        "    <tr>\n",
278 |        "      <th>1</th>\n",
279 |        "      <td>20.57</td>\n",
280 |        "      <td>17.77</td>\n",
281 |        "      <td>132.90</td>\n",
282 |        "      <td>1326.0</td>\n",
283 |        "      <td>0.08474</td>\n",
284 |        "      <td>0.07864</td>\n",
285 |        "      <td>0.0869</td>\n",
286 |        "      <td>0.07017</td>\n",
287 |        "      <td>0.1812</td>\n",
288 |        "      <td>0.05667</td>\n",
289 |        "      <td>...</td>\n",
290 |        "      <td>24.99</td>\n",
291 |        "      <td>23.41</td>\n",
292 |        "      <td>158.80</td>\n",
293 |        "      <td>1956.0</td>\n",
294 |        "      <td>0.1238</td>\n",
295 |        "      <td>0.1866</td>\n",
296 |        "      <td>0.2416</td>\n",
297 |        "      <td>0.1860</td>\n",
298 |        "      <td>0.2750</td>\n",
299 |        "      <td>0.08902</td>\n",
300 |        "    </tr>\n",
301 |        "    <tr>\n",
302 |        "      <th>2</th>\n",
303 |        "      <td>19.69</td>\n",
304 |        "      <td>21.25</td>\n",
305 |        "      <td>130.00</td>\n",
306 |        "      <td>1203.0</td>\n",
307 |        "      <td>0.10960</td>\n",
308 |        "      <td>0.15990</td>\n",
309 |        "      <td>0.1974</td>\n",
310 |        "      <td>0.12790</td>\n",
311 |        "      <td>0.2069</td>\n",
312 |        "      <td>0.05999</td>\n",
313 |        "      <td>...</td>\n",
314 |        "      <td>23.57</td>\n",
315 |        "      <td>25.53</td>\n",
316 |        "      <td>152.50</td>\n",
317 |        "      <td>1709.0</td>\n",
318 |        "      <td>0.1444</td>\n",
319 |        "      <td>0.4245</td>\n",
320 |        "      <td>0.4504</td>\n",
321 |        "      <td>0.2430</td>\n",
322 |        "      <td>0.3613</td>\n",
323 |        "      <td>0.08758</td>\n",
324 |        "    </tr>\n",
325 |        "    <tr>\n",
326 |        "      <th>3</th>\n",
327 |        "      <td>11.42</td>\n",
328 |        "      <td>20.38</td>\n",
329 |        "      <td>77.58</td>\n",
330 |        "      <td>386.1</td>\n",
331 |        "      <td>0.14250</td>\n",
332 |        "      <td>0.28390</td>\n",
333 |        "      <td>0.2414</td>\n",
334 |        "      <td>0.10520</td>\n",
335 |        "      <td>0.2597</td>\n",
336 |        "      <td>0.09744</td>\n",
337 |        "      <td>...</td>\n",
338 |        "      <td>14.91</td>\n",
339 |        "      <td>26.50</td>\n",
340 |        "      <td>98.87</td>\n",
341 |        "      <td>567.7</td>\n",
342 |        "      <td>0.2098</td>\n",
343 |        "      <td>0.8663</td>\n",
344 |        "      <td>0.6869</td>\n",
345 |        "      <td>0.2575</td>\n",
346 |        "      <td>0.6638</td>\n",
347 |        "      <td>0.17300</td>\n",
348 |        "    </tr>\n",
349 |        "    <tr>\n",
350 |        "      <th>4</th>\n",
351 |        "      <td>20.29</td>\n",
352 |        "      <td>14.34</td>\n",
353 |        "      <td>135.10</td>\n",
354 |        "      <td>1297.0</td>\n",
355 |        "      <td>0.10030</td>\n",
356 |        "      <td>0.13280</td>\n",
357 |        "      <td>0.1980</td>\n",
358 |        "      <td>0.10430</td>\n",
359 |        "      <td>0.1809</td>\n",
360 |        "      <td>0.05883</td>\n",
361 |        "      <td>...</td>\n",
362 |        "      <td>22.54</td>\n",
363 |        "      <td>16.67</td>\n",
364 |        "      <td>152.20</td>\n",
365 |        "      <td>1575.0</td>\n",
366 |        "      <td>0.1374</td>\n",
367 |        "      <td>0.2050</td>\n",
368 |        "      <td>0.4000</td>\n",
369 |        "      <td>0.1625</td>\n",
370 |        "      <td>0.2364</td>\n",
371 |        "      <td>0.07678</td>\n",
372 |        "    </tr>\n",
373 |        "  </tbody>\n",
374 |        "</table>\n",
375 |        "<p>5 rows × 30 columns</p>\n",
376 |        "</div>"
377 |       ],
378 |       "text/plain": [
379 |        "   mean radius  mean texture  mean perimeter  mean area  mean smoothness  \\\n",
380 |        "0        17.99         10.38          122.80     1001.0          0.11840   \n",
381 |        "1        20.57         17.77          132.90     1326.0          0.08474   \n",
382 |        "2        19.69         21.25          130.00     1203.0          0.10960   \n",
383 |        "3        11.42         20.38           77.58      386.1          0.14250   \n",
384 |        "4        20.29         14.34          135.10     1297.0          0.10030   \n",
385 |        "\n",
386 |        "   mean compactness  mean concavity  mean concave points  mean symmetry  \\\n",
387 |        "0           0.27760          0.3001              0.14710         0.2419   \n",
388 |        "1           0.07864          0.0869              0.07017         0.1812   \n",
389 |        "2           0.15990          0.1974              0.12790         0.2069   \n",
390 |        "3           0.28390          0.2414              0.10520         0.2597   \n",
391 |        "4           0.13280          0.1980              0.10430         0.1809   \n",
392 |        "\n",
393 |        "   mean fractal dimension           ...             worst radius  \\\n",
394 |        "0                 0.07871           ...                    25.38   \n",
395 |        "1                 0.05667           ...                    24.99   \n",
396 |        "2                 0.05999           ...                    23.57   \n",
397 |        "3                 0.09744           ...                    14.91   \n",
398 |        "4                 0.05883           ...                    22.54   \n",
399 |        "\n",
400 |        "   worst texture  worst perimeter  worst area  worst smoothness  \\\n",
401 |        "0          17.33           184.60      2019.0            0.1622   \n",
402 |        "1          23.41           158.80      1956.0            0.1238   \n",
403 |        "2          25.53           152.50      1709.0            0.1444   \n",
404 |        "3          26.50            98.87       567.7            0.2098   \n",
405 |        "4          16.67           152.20      1575.0            0.1374   \n",
406 |        "\n",
407 |        "   worst compactness  worst concavity  worst concave points  worst symmetry  \\\n",
408 |        "0             0.6656           0.7119                0.2654          0.4601   \n",
409 |        "1             0.1866           0.2416                0.1860          0.2750   \n",
410 |        "2             0.4245           0.4504                0.2430          0.3613   \n",
411 |        "3             0.8663           0.6869                0.2575          0.6638   \n",
412 |        "4             0.2050           0.4000                0.1625          0.2364   \n",
413 |        "\n",
414 |        "   worst fractal dimension  \n",
415 |        "0                  0.11890  \n",
416 |        "1                  0.08902  \n",
417 |        "2                  0.08758  \n",
418 |        "3                  0.17300  \n",
419 |        "4                  0.07678  \n",
420 |        "\n",
421 |        "[5 rows x 30 columns]"
422 |       ]
423 |      },
424 |      "execution_count": 7,
425 |      "metadata": {},
426 |      "output_type": "execute_result"
427 |     }
428 |    ],
429 |    "source": [
430 |     "df.head(5)"
431 |    ]
432 |   },
433 |   {
434 |    "cell_type": "code",
435 |    "execution_count": 8,
436 |    "metadata": {},
437 |    "outputs": [],
438 |    "source": [
439 |     "from sklearn.preprocessing import MinMaxScaler"
440 |    ]
441 |   },
442 |   {
443 |    "cell_type": "code",
444 |    "execution_count": 9,
445 |    "metadata": {},
446 |    "outputs": [],
447 |    "source": [
448 |     "from sklearn.preprocessing import StandardScaler"
449 |    ]
450 |   },
451 |   {
452 |    "cell_type": "code",
453 |    "execution_count": 10,
454 |    "metadata": {},
455 |    "outputs": [
456 |     {
457 |      "data": {
458 |       "text/plain": [
459 |        "StandardScaler(copy=True, with_mean=True, with_std=True)"
460 |       ]
461 |      },
462 |      "execution_count": 10,
463 |      "metadata": {},
464 |      "output_type": "execute_result"
465 |     }
466 |    ],
467 |    "source": [
468 |     "scaler=StandardScaler()\n",
469 |     "scaler.fit(df)"
470 |    ]
471 |   },
472 |   {
473 |    "cell_type": "code",
474 |    "execution_count": 11,
475 |    "metadata": {},
476 |    "outputs": [],
477 |    "source": [
478 |     "scaled_data=scaler.transform(df)"
479 |    ]
480 |   },
481 |   {
482 |    "cell_type": "code",
483 |    "execution_count": 12,
484 |    "metadata": {},
485 |    "outputs": [
486 |     {
487 |      "data": {
488 |       "text/plain": [
489 |        "array([[ 1.09706398, -2.07333501,  1.26993369, ...,  2.29607613,\n",
490 |        "         2.75062224,  1.93701461],\n",
491 |        "       [ 1.82982061, -0.35363241,  1.68595471, ...,  1.0870843 ,\n",
492 |        "        -0.24388967,  0.28118999],\n",
493 |        "       [ 1.57988811,  0.45618695,  1.56650313, ...,  1.95500035,\n",
494 |        "         1.152255  ,  0.20139121],\n",
495 |        "       ...,\n",
496 |        "       [ 0.70228425,  2.0455738 ,  0.67267578, ...,  0.41406869,\n",
497 |        "        -1.10454895, -0.31840916],\n",
498 |        "       [ 1.83834103,  2.33645719,  1.98252415, ...,  2.28998549,\n",
499 |        "         1.91908301,  2.21963528],\n",
500 |        "       [-1.80840125,  1.22179204, -1.81438851, ..., -1.74506282,\n",
501 |        "        -0.04813821, -0.75120669]])"
502 |       ]
503 |      },
504 |      "execution_count": 12,
505 |      "metadata": {},
506 |      "output_type": "execute_result"
507 |     }
508 |    ],
509 |    "source": [
510 |     "scaled_data"
511 |    ]
512 |   },
513 |   {
514 |    "cell_type": "code",
515 |    "execution_count": 13,
516 |    "metadata": {},
517 |    "outputs": [],
518 |    "source": [
519 |     "from sklearn.decomposition import PCA"
520 |    ]
521 |   },
522 |   {
523 |    "cell_type": "code",
524 |    "execution_count": 14,
525 |    "metadata": {},
526 |    "outputs": [],
527 |    "source": [
528 |     "pca=PCA(n_components=2)"
529 |    ]
530 |   },
531 |   {
532 |    "cell_type": "code",
533 |    "execution_count": 15,
534 |    "metadata": {},
535 |    "outputs": [
536 |     {
537 |      "data": {
538 |       "text/plain": [
539 |        "PCA(copy=True, iterated_power='auto', n_components=2, random_state=None,\n",
540 |        "  svd_solver='auto', tol=0.0, whiten=False)"
541 |       ]
542 |      },
543 |      "execution_count": 15,
544 |      "metadata": {},
545 |      "output_type": "execute_result"
546 |     }
547 |    ],
548 |    "source": [
549 |     "pca.fit(scaled_data)"
550 |    ]
551 |   },
552 |   {
553 |    "cell_type": "code",
554 |    "execution_count": 16,
555 |    "metadata": {},
556 |    "outputs": [],
557 |    "source": [
558 |     "x_pca=pca.transform(scaled_data)"
559 |    ]
560 |   },
561 |   {
562 |    "cell_type": "code",
563 |    "execution_count": 17,
564 |    "metadata": {},
565 |    "outputs": [
566 |     {
567 |      "data": {
568 |       "text/plain": [
569 |        "(569, 30)"
570 |       ]
571 |      },
572 |      "execution_count": 17,
573 |      "metadata": {},
574 |      "output_type": "execute_result"
575 |     }
576 |    ],
577 |    "source": [
578 |     "scaled_data.shape"
579 |    ]
580 |   },
581 |   {
582 |    "cell_type": "code",
583 |    "execution_count": 18,
584 |    "metadata": {},
585 |    "outputs": [
586 |     {
587 |      "data": {
588 |       "text/plain": [
589 |        "(569, 2)"
590 |       ]
591 |      },
592 |      "execution_count": 18,
593 |      "metadata": {},
594 |      "output_type": "execute_result"
595 |     }
596 |    ],
597 |    "source": [
598 |     "x_pca.shape"
599 |    ]
600 |   },
601 |   {
602 |    "cell_type": "code",
603 |    "execution_count": 19,
604 |    "metadata": {},
605 |    "outputs": [
606 |     {
607 |      "data": {
608 |       "text/plain": [
609 |        "array([[ 1.09706398, -2.07333501,  1.26993369, ...,  2.29607613,\n",
610 |        "         2.75062224,  1.93701461],\n",
611 |        "       [ 1.82982061, -0.35363241,  1.68595471, ...,  1.0870843 ,\n",
612 |        "        -0.24388967,  0.28118999],\n",
613 |        "       [ 1.57988811,  0.45618695,  1.56650313, ...,  1.95500035,\n",
614 |        "         1.152255  ,  0.20139121],\n",
615 |        "       ...,\n",
616 |        "       [ 0.70228425,  2.0455738 ,  0.67267578, ...,  0.41406869,\n",
617 |        "        -1.10454895, -0.31840916],\n",
618 |        "       [ 1.83834103,  2.33645719,  1.98252415, ...,  2.28998549,\n",
619 |        "         1.91908301,  2.21963528],\n",
620 |        "       [-1.80840125,  1.22179204, -1.81438851, ..., -1.74506282,\n",
621 |        "        -0.04813821, -0.75120669]])"
622 |       ]
623 |      },
624 |      "execution_count": 19,
625 |      "metadata": {},
626 |      "output_type": "execute_result"
627 |     }
628 |    ],
629 |    "source": [
630 |     "scaled_data"
631 |    ]
632 |   },
633 |   {
634 |    "cell_type": "code",
635 |    "execution_count": 20,
636 |    "metadata": {},
637 |    "outputs": [
638 |     {
639 |      "data": {
640 |       "text/plain": [
641 |        "array([[ 9.19283683,  1.94858307],\n",
642 |        "       [ 2.3878018 , -3.76817174],\n",
643 |        "       [ 5.73389628, -1.0751738 ],\n",
644 |        "       ...,\n",
645 |        "       [ 1.25617928, -1.90229671],\n",
646 |        "       [10.37479406,  1.67201011],\n",
647 |        "       [-5.4752433 , -0.67063679]])"
648 |       ]
649 |      },
650 |      "execution_count": 20,
651 |      "metadata": {},
652 |      "output_type": "execute_result"
653 |     }
654 |    ],
655 |    "source": [
656 |     "x_pca\n"
657 |    ]
658 |   },
659 |   {
660 |    "cell_type": "code",
661 |    "execution_count": 21,
662 |    "metadata": {},
663 |    "outputs": [
664 |     {
665 |      "data": {
666 |       "text/plain": [
667 |        "Text(0,0.5,'Second principle component')"
668 |       ]
669 |      },
670 |      "execution_count": 21,
671 |      "metadata": {},
672 |      "output_type": "execute_result"
673 |     },
674 |     {
675 |      "data": {
676 |       "image/png": "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\n",
677 |       "text/plain": [
678 |        "<matplotlib.figure.Figure at 0x1e8d4161588>"
679 |       ]
680 |      },
681 |      "metadata": {},
682 |      "output_type": "display_data"
683 |     }
684 |    ],
685 |    "source": [
686 |     "plt.figure(figsize=(8,6))\n",
687 |     "plt.scatter(x_pca[:,0],x_pca[:,1],c=cancer['target'])\n",
688 |     "plt.xlabel('First principle component')\n",
689 |     "plt.ylabel('Second principle component')\n",
690 |     "\n",
691 |     "\n",
692 |     "\n"
693 |    ]
694 |   },
695 |   {
696 |    "cell_type": "code",
697 |    "execution_count": null,
698 |    "metadata": {},
699 |    "outputs": [],
700 |    "source": []
701 |   }
702 |  ],
703 |  "metadata": {
704 |   "kernelspec": {
705 |    "display_name": "Python 3",
706 |    "language": "python",
707 |    "name": "python3"
708 |   },
709 |   "language_info": {
710 |    "codemirror_mode": {
711 |     "name": "ipython",
712 |     "version": 3
713 |    },
714 |    "file_extension": ".py",
715 |    "mimetype": "text/x-python",
716 |    "name": "python",
717 |    "nbconvert_exporter": "python",
718 |    "pygments_lexer": "ipython3",
719 |    "version": "3.6.4"
720 |   }
721 |  },
722 |  "nbformat": 4,
723 |  "nbformat_minor": 2
724 | }
725 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # Principle-Component-Analysis


--------------------------------------------------------------------------------