No Description
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

TP1_prog2.py.ipynb 406KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960
  1. {
  2. "cells": [
  3. {
  4. "cell_type": "code",
  5. "execution_count": 1,
  6. "id": "530f620c",
  7. "metadata": {},
  8. "outputs": [],
  9. "source": [
  10. "####### Import #######\n",
  11. "from sklearn.datasets import fetch_openml\n",
  12. "from sklearn import model_selection\n",
  13. "from sklearn import neighbors\n",
  14. "import sklearn\n",
  15. "import numpy as np\n",
  16. "from matplotlib import pyplot as plt\n",
  17. "from sklearn.model_selection import KFold\n",
  18. "import time\n",
  19. "import statistics\n",
  20. "from sklearn import metrics"
  21. ]
  22. },
  23. {
  24. "cell_type": "code",
  25. "execution_count": 2,
  26. "id": "68b6a517",
  27. "metadata": {},
  28. "outputs": [],
  29. "source": [
  30. "####### Loading data #######\n",
  31. "mnist = fetch_openml('mnist_784',as_frame=False)"
  32. ]
  33. },
  34. {
  35. "cell_type": "code",
  36. "execution_count": 3,
  37. "id": "eb2c4496",
  38. "metadata": {},
  39. "outputs": [
  40. {
  41. "name": "stdout",
  42. "output_type": "stream",
  43. "text": [
  44. "Classe image 4 : 7\n",
  45. "Classe prédite image 4 : 3\n",
  46. "Score échantillon de test : 0.907\n",
  47. "Score données apprentissage : 0.93375\n"
  48. ]
  49. }
  50. ],
  51. "source": [
  52. "### Create vector of 5000 random indexes\n",
  53. "rand_indexes = np.random.randint(70000, size=5000)\n",
  54. "### Load data with the previous vector\n",
  55. "data = mnist.data[rand_indexes]\n",
  56. "# print(\"Dataset : \", data)\n",
  57. "target = mnist.target[rand_indexes]\n",
  58. "# print(\"Etiquettes : \", target)\n",
  59. "\n",
  60. "### Split the dataset for training and testing\n",
  61. "# xtrain data set d'entraînement et ytrain étiquettes de xtrain\n",
  62. "# xtest dataset de prédiction et ytest étiquettes de xtest\n",
  63. "xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=0.8)\n",
  64. "\n",
  65. "# On entraîne avec k = 10\n",
  66. "n_neighbors = 10\n",
  67. "clf = neighbors.KNeighborsClassifier(n_neighbors)\n",
  68. "# Training with xtrain,ytrain\n",
  69. "clf.fit(xtrain, ytrain)\n",
  70. "# Predicting with xtest\n",
  71. "pred = clf.predict(xtest)\n",
  72. "# print(\"Prédiction : \", pred)\n",
  73. "# Predicting probabilities with xtest\n",
  74. "pred_proba = clf.predict_proba(xtest)\n",
  75. "# print(\"Probabilités : \", pred_proba)\n",
  76. "# Computing the score with xtest,ytest\n",
  77. "score = clf.score(xtest, ytest)\n",
  78. "print(\"Classe image 4 : \", target[3])\n",
  79. "print(\"Classe prédite image 4 : \", pred[3])\n",
  80. "print(\"Score échantillon de test : \", score)\n",
  81. "scoreApp = clf.score(xtrain, ytrain)\n",
  82. "print(\"Score données apprentissage : \", scoreApp)"
  83. ]
  84. },
  85. {
  86. "cell_type": "code",
  87. "execution_count": 18,
  88. "id": "90db6e29",
  89. "metadata": {},
  90. "outputs": [
  91. {
  92. "name": "stdout",
  93. "output_type": "stream",
  94. "text": [
  95. "Computing for k = 2 ...\n",
  96. "Computing for k = 3 ...\n",
  97. "Computing for k = 4 ...\n",
  98. "Computing for k = 5 ...\n",
  99. "Computing for k = 6 ...\n",
  100. "Computing for k = 7 ...\n",
  101. "Computing for k = 8 ...\n",
  102. "Computing for k = 9 ...\n",
  103. "Computing for k = 10 ...\n",
  104. "Computing for k = 11 ...\n",
  105. "Computing for k = 12 ...\n",
  106. "Computing for k = 13 ...\n",
  107. "Computing for k = 14 ...\n",
  108. "Computing for k = 15 ...\n",
  109. "Done\n"
  110. ]
  111. }
  112. ],
  113. "source": [
  114. "####### Variation du nombre k de voisins #######\n",
  115. "\n",
  116. "### Create vector of 5000 random indexes\n",
  117. "rand_indexes = np.random.randint(70000, size=5000)\n",
  118. "### Load data with the previous vector\n",
  119. "data = mnist.data[rand_indexes]\n",
  120. "# print(\"Dataset : \", data)\n",
  121. "target = mnist.target[rand_indexes]\n",
  122. "# print(\"Etiquettes : \", target)\n",
  123. "\n",
  124. "### Split the data in 10 chunks\n",
  125. "kf = KFold(n_splits=10, random_state=None, shuffle=True)\n",
  126. "\n",
  127. "# Initialisation des métriques\n",
  128. "all_scores = []\n",
  129. "k_scores = []\n",
  130. "all_training_times = []\n",
  131. "k_training_times = []\n",
  132. "all_predicting_times = []\n",
  133. "k_prediction_times = []\n",
  134. "\n",
  135. "# Fais varier le nombre de voisins de 2 à 15\n",
  136. "for k in range(2,16):\n",
  137. " print(\"Computing for k =\", k, \"...\")\n",
  138. " k_scores = []\n",
  139. " k_training_times = []\n",
  140. " k_prediction_times = []\n",
  141. " # Boucle sur chaque paire de jeu de données entraînement/test\n",
  142. " for train_index, test_index in kf.split(data):\n",
  143. " X_train, X_test = data[train_index], data[test_index]\n",
  144. " y_train, y_test = target[train_index], target[test_index]\n",
  145. " \n",
  146. " t1 = round(time.time(),5)\n",
  147. " clf = neighbors.KNeighborsClassifier(k)\n",
  148. " # On entraîne l'algorithme sur xtrain et ytrain\n",
  149. " clf.fit(X_train, y_train)\n",
  150. " t2 = round(time.time(),5)\n",
  151. " # On prédit sur xtest\n",
  152. " pred = clf.predict(X_test)\n",
  153. " t3 = round(time.time(),5)\n",
  154. " # Probabilités des prédictions sur xtest\n",
  155. " pred_proba = clf.predict_proba(X_test)\n",
  156. " # On calcule le score obtenu sur xtest avec les étiquettes ytest\n",
  157. " # et l'ajoute à la liste des scores\n",
  158. " k_scores.append(clf.score(X_test, y_test))\n",
  159. " k_training_times.append(t2-t1)\n",
  160. " k_prediction_times.append(t3-t2)\n",
  161. " all_scores.append(k_scores)\n",
  162. " all_training_times.append(k_training_times)\n",
  163. " all_predicting_times.append(k_prediction_times)\n",
  164. "print(\"Done\")"
  165. ]
  166. },
  167. {
  168. "cell_type": "code",
  169. "execution_count": 23,
  170. "id": "79d055d6",
  171. "metadata": {},
  172. "outputs": [
  173. {
  174. "data": {
  175. "text/plain": [
  176. "Text(108.0, 0.5, 'Predicting times (in ms)')"
  177. ]
  178. },
  179. "execution_count": 23,
  180. "metadata": {},
  181. "output_type": "execute_result"
  182. },
  183. {
  184. "data": {
  185. "image/png": "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\n",
  186. "text/plain": [
  187. "<Figure size 1080x720 with 3 Axes>"
  188. ]
  189. },
  190. "metadata": {
  191. "needs_background": "light"
  192. },
  193. "output_type": "display_data"
  194. }
  195. ],
  196. "source": [
  197. "##### x coordinates\n",
  198. "x = np.arange(2,16,1)\n",
  199. "mean_scores = []\n",
  200. "mean_training_times = []\n",
  201. "mean_predicting_times = []\n",
  202. "for i in range(len(all_scores)):\n",
  203. " mean_scores += [round(statistics.mean(all_scores[i]),5)]\n",
  204. " mean_training_times += [round(statistics.mean(all_training_times[i])*1000,2)]\n",
  205. " mean_predicting_times += [round(statistics.mean(all_predicting_times[i])*1000,2)]\n",
  206. "mean_scores = np.float64(mean_scores)\n",
  207. "mean_training_times = np.float64(mean_training_times)\n",
  208. "mean_predicting_times = np.float64(mean_predicting_times)\n",
  209. "\n",
  210. "### Create plot\n",
  211. "fig, figs = plt.subplots(nrows=3, ncols=1, figsize=(15,10))\n",
  212. "fig.tight_layout(pad=3.0)\n",
  213. "figs[0].plot(x,mean_scores, marker='o', color='r')\n",
  214. "figs[1].plot(x,mean_training_times, marker='o', color='b')\n",
  215. "figs[2].plot(x,mean_predicting_times, marker='o', color='g')\n",
  216. "\n",
  217. "### Add every x coordinates\n",
  218. "figs[0].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  219. "figs[1].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  220. "figs[2].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  221. "\n",
  222. "for i in range(len(x)):\n",
  223. " figs[0].annotate(mean_scores[i], # this is the text\n",
  224. " (x[i],mean_scores[i]), # these are the coordinates to position the label\n",
  225. " textcoords=\"offset points\", # how to position the text\n",
  226. " xytext=(15,5), # distance from text to points (x,y)\n",
  227. " ha='center') # horizontal alignment can be left, right or center\n",
  228. " figs[1].annotate(mean_training_times[i], # this is the text\n",
  229. " (x[i],mean_training_times[i]), # these are the coordinates to position the label\n",
  230. " textcoords=\"offset points\", # how to position the text\n",
  231. " xytext=(15,5), # distance from text to points (x,y)\n",
  232. " ha='center') # horizontal alignment can be left, right or center\n",
  233. " figs[2].annotate(mean_predicting_times[i], # this is the text\n",
  234. " (x[i],mean_predicting_times[i]), # these are the coordinates to position the label\n",
  235. " textcoords=\"offset points\", # how to position the text\n",
  236. " xytext=(15,5), # distance from text to points (x,y)\n",
  237. " ha='center') # horizontal alignment can be left, right or center\n",
  238. "\n",
  239. "figs[0].set_xticks(x)\n",
  240. "figs[1].set_xticks(x)\n",
  241. "figs[2].set_xticks(x)\n",
  242. " \n",
  243. "### Add title and axis names\n",
  244. "figs[0].title.set_text('Mean scores for k neighbors')\n",
  245. "figs[1].title.set_text('Mean training times for k neighbors')\n",
  246. "figs[2].title.set_text('Mean predicting times for k neighbors')\n",
  247. "figs[0].set_xlabel('N_neighbors')\n",
  248. "figs[1].set_xlabel('N_neighbors')\n",
  249. "figs[2].set_xlabel('N_neighbors')\n",
  250. "figs[0].set_ylabel('Score')\n",
  251. "figs[1].set_ylabel('Training times (in ms)')\n",
  252. "figs[2].set_ylabel('Predicting times (in ms)')"
  253. ]
  254. },
  255. {
  256. "cell_type": "code",
  257. "execution_count": 28,
  258. "id": "cc24e898",
  259. "metadata": {},
  260. "outputs": [
  261. {
  262. "name": "stdout",
  263. "output_type": "stream",
  264. "text": [
  265. "Computing for j = 10 %...\n",
  266. "Computing for j = 20 %...\n",
  267. "Computing for j = 30 %...\n",
  268. "Computing for j = 40 %...\n",
  269. "Computing for j = 50 %...\n",
  270. "Computing for j = 60 %...\n",
  271. "Computing for j = 70 %...\n",
  272. "Computing for j = 80 %...\n",
  273. "Computing for j = 90 %...\n",
  274. "Done\n"
  275. ]
  276. }
  277. ],
  278. "source": [
  279. "####### Variation du pourcentage des échantillons #######\n",
  280. "\n",
  281. "### Create vector of 5000 random indexes\n",
  282. "rand_indexes = np.random.randint(70000, size=5000)\n",
  283. "### Load data with the previous vector\n",
  284. "data = mnist.data[rand_indexes]\n",
  285. "# print(\"Dataset : \", data)\n",
  286. "target = mnist.target[rand_indexes]\n",
  287. "# print(\"Etiquettes : \", target)\n",
  288. "\n",
  289. "scores = []\n",
  290. "training_times = []\n",
  291. "prediction_times = []\n",
  292. "\n",
  293. "### Train the algorithm with various percentage of the dataset used for training\n",
  294. "### from 10% to 90% by 10% increment\n",
  295. "for j in range (1, 10):\n",
  296. " print(\"Computing for j =\", j*10, \"%...\")\n",
  297. " # Split the dataset\n",
  298. " xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=(j/10))\n",
  299. " \n",
  300. " # Training on xtrain,ytrain\n",
  301. " t1 = round(time.time(),5)\n",
  302. " clf = neighbors.KNeighborsClassifier(n_neighbors=3)\n",
  303. " clf.fit(xtrain, ytrain)\n",
  304. " t2 = round(time.time(),5)\n",
  305. " \n",
  306. " # Predicting on xtest\n",
  307. " pred = clf.predict(xtest)\n",
  308. " t3 = round(time.time(),5)\n",
  309. " training_times.append(round(t2-t1,5))\n",
  310. " prediction_times.append(round(t3-t2,5))\n",
  311. " # Probabilités des prédictions sur xtest\n",
  312. " pred_proba = clf.predict_proba(xtest)\n",
  313. " # On calcule le score obtenu sur xtest avec les étiquettes ytest\n",
  314. " scores.append(round(clf.score(xtest, ytest),3))\n",
  315. "print(\"Done\")"
  316. ]
  317. },
  318. {
  319. "cell_type": "code",
  320. "execution_count": 30,
  321. "id": "cbb5eda6",
  322. "metadata": {},
  323. "outputs": [
  324. {
  325. "name": "stdout",
  326. "output_type": "stream",
  327. "text": [
  328. "training_times : \n",
  329. " [0.00296, 0.002, 0.003, 0.00399, 0.00698, 0.00499, 0.00698, 0.00698, 0.00898]\n",
  330. "prediction_times : \n",
  331. " [0.19548, 0.25033, 0.24833, 0.2244, 0.24136, 0.22241, 0.17752, 0.12467, 0.08975]\n"
  332. ]
  333. },
  334. {
  335. "data": {
  336. "text/plain": [
  337. "Text(108.0, 0.5, 'Times (in ms)')"
  338. ]
  339. },
  340. "execution_count": 30,
  341. "metadata": {},
  342. "output_type": "execute_result"
  343. },
  344. {
  345. "data": {
  346. "image/png": "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\n",
  347. "text/plain": [
  348. "<Figure size 1080x720 with 3 Axes>"
  349. ]
  350. },
  351. "metadata": {
  352. "needs_background": "light"
  353. },
  354. "output_type": "display_data"
  355. }
  356. ],
  357. "source": [
  358. "##### x coordinates\n",
  359. "x = np.arange(10,100,10)\n",
  360. "print(\"training_times : \\n\", training_times)\n",
  361. "print(\"prediction_times : \\n\", prediction_times)\n",
  362. "\n",
  363. "training_times_ms = [round(i*1000,2) for i in training_times]\n",
  364. "prediction_times_ms = [round(i*1000,2) for i in prediction_times]\n",
  365. "\n",
  366. "### Create plot\n",
  367. "fig, figs = plt.subplots(nrows=3, ncols=1, figsize=(15,10))\n",
  368. "fig.tight_layout(pad=3.0)\n",
  369. "figs[0].plot(x,scores, marker='o', color='r')\n",
  370. "figs[1].plot(x,training_times_ms, marker='o', color='b')\n",
  371. "figs[2].plot(x,prediction_times_ms, marker='o', color='g')\n",
  372. "### Add every x coordinates\n",
  373. "figs[0].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  374. "figs[1].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  375. "figs[2].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  376. "\n",
  377. "for i in range(len(x)):\n",
  378. " figs[0].annotate(scores[i], # this is the text\n",
  379. " (x[i],scores[i]), # these are the coordinates to position the label\n",
  380. " textcoords=\"offset points\", # how to position the text\n",
  381. " xytext=(15,5), # distance from text to points (x,y)\n",
  382. " ha='center') # horizontal alignment can be left, right or center\n",
  383. " figs[1].annotate(training_times_ms[i], # this is the text\n",
  384. " (x[i],training_times_ms[i]), # these are the coordinates to position the label\n",
  385. " textcoords=\"offset points\", # how to position the text\n",
  386. " xytext=(15,5), # distance from text to points (x,y)\n",
  387. " ha='center') # horizontal alignment can be left, right or center\n",
  388. " figs[2].annotate(prediction_times_ms[i], # this is the text\n",
  389. " (x[i],prediction_times_ms[i]), # these are the coordinates to position the label\n",
  390. " textcoords=\"offset points\", # how to position the text\n",
  391. " xytext=(15,5), # distance from text to points (x,y)\n",
  392. " ha='center') # horizontal alignment can be left, right or center\n",
  393. "\n",
  394. "figs[0].set_xticks(x)\n",
  395. "figs[1].set_xticks(x)\n",
  396. "figs[2].set_xticks(x)\n",
  397. " \n",
  398. "### Add title and axis names\n",
  399. "figs[0].title.set_text('Scores for various percentage of training dataset (k=3)')\n",
  400. "figs[1].title.set_text('Training times for various percentage of training dataset (k=3)')\n",
  401. "figs[2].title.set_text('Prediction times for various percentage of training dataset (k=3)')\n",
  402. "figs[0].set_xlabel('Percentage of training dataset')\n",
  403. "figs[1].set_xlabel('Percentage of training dataset')\n",
  404. "figs[2].set_xlabel('Percentage of training dataset')\n",
  405. "figs[0].set_ylabel('Score')\n",
  406. "figs[1].set_ylabel('Times (in ms)')\n",
  407. "figs[2].set_ylabel('Times (in ms)')"
  408. ]
  409. },
  410. {
  411. "cell_type": "code",
  412. "execution_count": 33,
  413. "id": "572d0a52",
  414. "metadata": {},
  415. "outputs": [
  416. {
  417. "name": "stdout",
  418. "output_type": "stream",
  419. "text": [
  420. "Computing for taille échantillon = 5000 données...\n",
  421. "Computing for taille échantillon = 10000 données...\n",
  422. "Computing for taille échantillon = 15000 données...\n",
  423. "Computing for taille échantillon = 20000 données...\n",
  424. "Computing for taille échantillon = 25000 données...\n",
  425. "Computing for taille échantillon = 30000 données...\n",
  426. "Computing for taille échantillon = 35000 données...\n",
  427. "Computing for taille échantillon = 40000 données...\n",
  428. "Computing for taille échantillon = 45000 données...\n",
  429. "Computing for taille échantillon = 50000 données...\n",
  430. "Computing for taille échantillon = 55000 données...\n",
  431. "Computing for taille échantillon = 60000 données...\n",
  432. "Computing for taille échantillon = 65000 données...\n",
  433. "Computing for taille échantillon = 70000 données...\n",
  434. "Done\n"
  435. ]
  436. }
  437. ],
  438. "source": [
  439. "####### Variation de la taille de l'échantillon #######\n",
  440. "scores = []\n",
  441. "training_times = []\n",
  442. "prediction_times = []\n",
  443. "\n",
  444. "for i in range (1,15):\n",
  445. " print(\"Computing for taille échantillon = \", i*5000, \" données...\")\n",
  446. " ### Create vector of 5000 random indexes\n",
  447. " rand_indexes = np.random.randint(70000, size=i*5000)\n",
  448. " ### Load data with the previous vector\n",
  449. " data = mnist.data[rand_indexes]\n",
  450. " target = mnist.target[rand_indexes]\n",
  451. "\n",
  452. " # Split the dataset\n",
  453. " xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=(0.9))\n",
  454. "\n",
  455. " # Training on xtrain,ytrain\n",
  456. " t1 = round(time.time(),5)\n",
  457. " clf = neighbors.KNeighborsClassifier(n_neighbors=3)\n",
  458. " clf.fit(xtrain, ytrain)\n",
  459. " t2 = round(time.time(),5)\n",
  460. "\n",
  461. " # Predicting on xtest\n",
  462. " pred = clf.predict(xtest)\n",
  463. " t3 = round(time.time(),5)\n",
  464. " training_times.append(round(t2-t1,5))\n",
  465. " prediction_times.append(round(t3-t2,5))\n",
  466. " # Probabilités des prédictions sur xtest\n",
  467. " pred_proba = clf.predict_proba(xtest)\n",
  468. " # On calcule le score obtenu sur xtest avec les étiquettes ytest\n",
  469. " scores.append(round(clf.score(xtest, ytest),3))\n",
  470. "print(\"Done\")"
  471. ]
  472. },
  473. {
  474. "cell_type": "code",
  475. "execution_count": 34,
  476. "id": "61279a41",
  477. "metadata": {},
  478. "outputs": [
  479. {
  480. "data": {
  481. "text/plain": [
  482. "Text(108.0, 0.5, 'Times (in ms)')"
  483. ]
  484. },
  485. "execution_count": 34,
  486. "metadata": {},
  487. "output_type": "execute_result"
  488. },
  489. {
  490. "data": {
  491. "image/png": "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\n",
  492. "text/plain": [
  493. "<Figure size 1080x720 with 3 Axes>"
  494. ]
  495. },
  496. "metadata": {
  497. "needs_background": "light"
  498. },
  499. "output_type": "display_data"
  500. }
  501. ],
  502. "source": [
  503. "##### x coordinates\n",
  504. "x = np.arange(5000,75000,5000)\n",
  505. "\n",
  506. "training_times_ms = [round(i*1000,2) for i in training_times]\n",
  507. "prediction_times_ms = [round(i*1000,2) for i in prediction_times]\n",
  508. "\n",
  509. "### Create plot\n",
  510. "fig, figs = plt.subplots(nrows=3, ncols=1, figsize=(15,10))\n",
  511. "fig.tight_layout(pad=3.0)\n",
  512. "figs[0].plot(x,scores, marker='o', color='r')\n",
  513. "figs[1].plot(x,training_times_ms, marker='o', color='b')\n",
  514. "figs[2].plot(x,prediction_times_ms, marker='o', color='g')\n",
  515. "### Add every x coordinates\n",
  516. "figs[0].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  517. "figs[1].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  518. "figs[2].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  519. "\n",
  520. "for i in range(len(x)):\n",
  521. " figs[0].annotate(scores[i], # this is the text\n",
  522. " (x[i],scores[i]), # these are the coordinates to position the label\n",
  523. " textcoords=\"offset points\", # how to position the text\n",
  524. " xytext=(15,5), # distance from text to points (x,y)\n",
  525. " ha='center') # horizontal alignment can be left, right or center\n",
  526. " figs[1].annotate(training_times_ms[i], # this is the text\n",
  527. " (x[i],training_times_ms[i]), # these are the coordinates to position the label\n",
  528. " textcoords=\"offset points\", # how to position the text\n",
  529. " xytext=(15,5), # distance from text to points (x,y)\n",
  530. " ha='center') # horizontal alignment can be left, right or center\n",
  531. " figs[2].annotate(prediction_times_ms[i], # this is the text\n",
  532. " (x[i],prediction_times_ms[i]), # these are the coordinates to position the label\n",
  533. " textcoords=\"offset points\", # how to position the text\n",
  534. " xytext=(15,5), # distance from text to points (x,y)\n",
  535. " ha='center') # horizontal alignment can be left, right or center\n",
  536. "\n",
  537. "figs[0].set_xticks(x)\n",
  538. "figs[1].set_xticks(x)\n",
  539. "figs[2].set_xticks(x)\n",
  540. " \n",
  541. "### Add title and axis names\n",
  542. "figs[0].title.set_text('Scores for various size of the dataset (k=3,train_size=0.9)')\n",
  543. "figs[1].title.set_text('Training times for various size of the dataset (k=3,train_size=0.9)')\n",
  544. "figs[2].title.set_text('Prediction times for various size of the dataset (k=3,train_size=0.9)')\n",
  545. "figs[0].set_xlabel('Size of the dataset')\n",
  546. "figs[1].set_xlabel('Size of the dataset')\n",
  547. "figs[2].set_xlabel('Size of the dataset')\n",
  548. "figs[0].set_ylabel('Score')\n",
  549. "figs[1].set_ylabel('Times (in ms)')\n",
  550. "figs[2].set_ylabel('Times (in ms)')"
  551. ]
  552. },
  553. {
  554. "cell_type": "code",
  555. "execution_count": 35,
  556. "id": "b766a47e",
  557. "metadata": {
  558. "scrolled": false
  559. },
  560. "outputs": [
  561. {
  562. "name": "stdout",
  563. "output_type": "stream",
  564. "text": [
  565. "Computing for d = manhattan ...\n",
  566. "Fitting...\n",
  567. "Predicting...\n",
  568. "Computing for d = euclidean ...\n",
  569. "Fitting...\n",
  570. "Predicting...\n",
  571. "Computing for d = minkowski ...\n",
  572. "Fitting...\n",
  573. "Predicting...\n",
  574. "Done\n"
  575. ]
  576. }
  577. ],
  578. "source": [
  579. "####### Variation des types de distances utilisées #######\n",
  580. "\n",
  581. "### Create vector of 5000 random indexes\n",
  582. "rand_indexes = np.random.randint(70000, size=25000)\n",
  583. "### Load data with the previous vector\n",
  584. "data = mnist.data[rand_indexes]\n",
  585. "# print(\"Dataset : \", data)\n",
  586. "target = mnist.target[rand_indexes]\n",
  587. "# print(\"Etiquettes : \", target)\n",
  588. "\n",
  589. "scores = []\n",
  590. "training_times = []\n",
  591. "prediction_times = []\n",
  592. "\n",
  593. "### Train the algorithm with various distance types\n",
  594. "distances = [1,2,3]\n",
  595. "names_distances = [\"manhattan\",\"euclidean\",\"minkowski\"]\n",
  596. "\n",
  597. "for i in distances:\n",
  598. " print(\"Computing for d =\", names_distances[i-1],\"...\")\n",
  599. " # Split the dataset\n",
  600. " xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=0.9)\n",
  601. " \n",
  602. " # Training on xtrain,ytrain\n",
  603. " t1 = round(time.time(),5)\n",
  604. " clf = neighbors.KNeighborsClassifier(n_neighbors=3,p=i)\n",
  605. " print(\"Fitting...\")\n",
  606. " clf.fit(xtrain, ytrain)\n",
  607. " t2 = round(time.time(),5)\n",
  608. " \n",
  609. " # Predicting on xtest\n",
  610. " print(\"Predicting...\")\n",
  611. " pred = clf.predict(xtest)\n",
  612. " t3 = round(time.time(),5)\n",
  613. " training_times.append(round(t2-t1,5))\n",
  614. " prediction_times.append(round(t3-t2,5))\n",
  615. " # Probabilités des prédictions sur xtest\n",
  616. " pred_proba = clf.predict_proba(xtest)\n",
  617. " # On calcule le score obtenu sur xtest avec les étiquettes ytest\n",
  618. " scores.append(round(clf.score(xtest, ytest),3))\n",
  619. "print(\"Done\")"
  620. ]
  621. },
  622. {
  623. "cell_type": "code",
  624. "execution_count": 40,
  625. "id": "8f4fd158",
  626. "metadata": {},
  627. "outputs": [
  628. {
  629. "name": "stdout",
  630. "output_type": "stream",
  631. "text": [
  632. "names_distances : ['manhattan', 'euclidean', 'minkowski']\n",
  633. "training_times_ms : [62.87, 46.87, 49.9]\n",
  634. "prediction_times_ms : [55381.76, 1318.48, 667718.1]\n"
  635. ]
  636. },
  637. {
  638. "data": {
  639. "text/plain": [
  640. "Text(63.0, 0.5, 'Times (in ms)')"
  641. ]
  642. },
  643. "execution_count": 40,
  644. "metadata": {},
  645. "output_type": "execute_result"
  646. },
  647. {
  648. "data": {
  649. "image/png": "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\n",
  650. "text/plain": [
  651. "<Figure size 720x720 with 3 Axes>"
  652. ]
  653. },
  654. "metadata": {
  655. "needs_background": "light"
  656. },
  657. "output_type": "display_data"
  658. }
  659. ],
  660. "source": [
  661. "##### x coordinates\n",
  662. "x = names_distances\n",
  663. "print(\"names_distances : \", names_distances)\n",
  664. "training_times_ms = [round(i*1000,2) for i in training_times]\n",
  665. "print(\"training_times_ms : \", training_times_ms)\n",
  666. "prediction_times_ms = [round(i*1000,2) for i in prediction_times]\n",
  667. "print(\"prediction_times_ms : \", prediction_times_ms)\n",
  668. "\n",
  669. "### Create plot\n",
  670. "fig, figs = plt.subplots(nrows=3, ncols=1, figsize=(10,10))\n",
  671. "fig.tight_layout(pad=3.0)\n",
  672. "figs[0].plot(x,scores, marker='o', color='r')\n",
  673. "figs[1].plot(x,training_times_ms, marker='o', color='b')\n",
  674. "figs[2].plot(x,prediction_times_ms, marker='o', color='g')\n",
  675. "### Add every x coordinates\n",
  676. "figs[0].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  677. "figs[1].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  678. "figs[2].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  679. "\n",
  680. "for i in range(len(x)):\n",
  681. " figs[0].annotate(scores[i], # this is the text\n",
  682. " (x[i],scores[i]), # these are the coordinates to position the label\n",
  683. " textcoords=\"offset points\", # how to position the text\n",
  684. " xytext=(15,5), # distance from text to points (x,y)\n",
  685. " ha='center') # horizontal alignment can be left, right or center\n",
  686. " figs[1].annotate(training_times_ms[i], # this is the text\n",
  687. " (x[i],training_times_ms[i]), # these are the coordinates to position the label\n",
  688. " textcoords=\"offset points\", # how to position the text\n",
  689. " xytext=(15,5), # distance from text to points (x,y)\n",
  690. " ha='center') # horizontal alignment can be left, right or center\n",
  691. " figs[2].annotate(prediction_times_ms[i], # this is the text\n",
  692. " (x[i],prediction_times_ms[i]), # these are the coordinates to position the label\n",
  693. " textcoords=\"offset points\", # how to position the text\n",
  694. " xytext=(15,5), # distance from text to points (x,y)\n",
  695. " ha='center') # horizontal alignment can be left, right or center\n",
  696. "\n",
  697. "figs[0].set_xticks(x)\n",
  698. "figs[1].set_xticks(x)\n",
  699. "figs[2].set_xticks(x)\n",
  700. " \n",
  701. "### Add title and axis names\n",
  702. "figs[0].title.set_text('Scores for various distance type (k=3,train_size=0.9,dataset_size=25000)')\n",
  703. "figs[1].title.set_text('Training times for various distance type (k=3,train_size=0.9,dataset_size=25000)')\n",
  704. "figs[2].title.set_text('Prediction times for various distance type (k=3,train_size=0.9,dataset_size=25000)')\n",
  705. "figs[0].set_xlabel('Distance type')\n",
  706. "figs[1].set_xlabel('Distance type')\n",
  707. "figs[2].set_xlabel('Distance type')\n",
  708. "figs[0].set_ylabel('Score')\n",
  709. "figs[1].set_ylabel('Times (in ms)')\n",
  710. "figs[2].set_ylabel('Times (in ms)')"
  711. ]
  712. },
  713. {
  714. "cell_type": "code",
  715. "execution_count": 5,
  716. "id": "24b641ef",
  717. "metadata": {},
  718. "outputs": [
  719. {
  720. "name": "stdout",
  721. "output_type": "stream",
  722. "text": [
  723. "Computing for n_jobs = -1 ...\n",
  724. "Fitting...\n",
  725. "Predicting...\n",
  726. "Computing for n_jobs = 1 ...\n",
  727. "Fitting...\n",
  728. "Predicting...\n",
  729. "Done\n"
  730. ]
  731. }
  732. ],
  733. "source": [
  734. "####### Variation de n_jobs #######\n",
  735. "\n",
  736. "### Create vector of 5000 random indexes\n",
  737. "rand_indexes = np.random.randint(70000, size=25000)\n",
  738. "### Load data with the previous vector\n",
  739. "data = mnist.data[rand_indexes]\n",
  740. "# print(\"Dataset : \", data)\n",
  741. "target = mnist.target[rand_indexes]\n",
  742. "# print(\"Etiquettes : \", target)\n",
  743. "\n",
  744. "scores = []\n",
  745. "training_times = []\n",
  746. "prediction_times = []\n",
  747. "\n",
  748. "### Train the algorithm with various distance types\n",
  749. "nb_jobs = [-1,1]\n",
  750. "\n",
  751. "for i in nb_jobs:\n",
  752. " print(\"Computing for n_jobs =\", i,\"...\")\n",
  753. " # Split the dataset\n",
  754. " xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=0.9)\n",
  755. " \n",
  756. " # Training on xtrain,ytrain\n",
  757. " t1 = round(time.time(),5)\n",
  758. " clf = neighbors.KNeighborsClassifier(n_neighbors=3,p=2,n_jobs=i)\n",
  759. " print(\"Fitting...\")\n",
  760. " clf.fit(xtrain, ytrain)\n",
  761. " t2 = round(time.time(),5)\n",
  762. " \n",
  763. " # Predicting on xtest\n",
  764. " print(\"Predicting...\")\n",
  765. " pred = clf.predict(xtest)\n",
  766. " t3 = round(time.time(),5)\n",
  767. " training_times.append(round(t2-t1,5))\n",
  768. " prediction_times.append(round(t3-t2,5))\n",
  769. " # Probabilités des prédictions sur xtest\n",
  770. " pred_proba = clf.predict_proba(xtest)\n",
  771. " # On calcule le score obtenu sur xtest avec les étiquettes ytest\n",
  772. " scores.append(round(clf.score(xtest, ytest),3))\n",
  773. "print(\"Done\")"
  774. ]
  775. },
  776. {
  777. "cell_type": "code",
  778. "execution_count": 7,
  779. "id": "343e5d14",
  780. "metadata": {},
  781. "outputs": [
  782. {
  783. "name": "stdout",
  784. "output_type": "stream",
  785. "text": [
  786. "training_times_ms : [39.89, 48.87]\n",
  787. "prediction_times_ms : [1778.06, 1312.49]\n"
  788. ]
  789. },
  790. {
  791. "data": {
  792. "text/plain": [
  793. "Text(36.0, 0.5, 'Times (in ms)')"
  794. ]
  795. },
  796. "execution_count": 7,
  797. "metadata": {},
  798. "output_type": "execute_result"
  799. },
  800. {
  801. "data": {
  802. "image/png": "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\n",
  803. "text/plain": [
  804. "<Figure size 504x720 with 3 Axes>"
  805. ]
  806. },
  807. "metadata": {
  808. "needs_background": "light"
  809. },
  810. "output_type": "display_data"
  811. }
  812. ],
  813. "source": [
  814. "##### x coordinates\n",
  815. "x = nb_jobs\n",
  816. "training_times_ms = [round(i*1000,2) for i in training_times]\n",
  817. "print(\"training_times_ms : \", training_times_ms)\n",
  818. "prediction_times_ms = [round(i*1000,2) for i in prediction_times]\n",
  819. "print(\"prediction_times_ms : \", prediction_times_ms)\n",
  820. "\n",
  821. "### Create plot\n",
  822. "fig, figs = plt.subplots(nrows=3, ncols=1, figsize=(7,10))\n",
  823. "fig.tight_layout(pad=3.0)\n",
  824. "figs[0].bar(x,scores, color='r')\n",
  825. "figs[1].bar(x,training_times_ms, color='b')\n",
  826. "figs[2].bar(x,prediction_times_ms, color='g')\n",
  827. "### Add every x coordinates\n",
  828. "figs[0].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  829. "figs[1].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  830. "figs[2].tick_params(axis='both', which='both', labelsize=7, labelbottom=True)\n",
  831. "\n",
  832. "for i in range(len(x)):\n",
  833. " figs[0].annotate(scores[i], # this is the text\n",
  834. " (x[i],scores[i]), # these are the coordinates to position the label\n",
  835. " textcoords=\"offset points\", # how to position the text\n",
  836. " xytext=(15,5), # distance from text to points (x,y)\n",
  837. " ha='center') # horizontal alignment can be left, right or center\n",
  838. " figs[1].annotate(training_times_ms[i], # this is the text\n",
  839. " (x[i],training_times_ms[i]), # these are the coordinates to position the label\n",
  840. " textcoords=\"offset points\", # how to position the text\n",
  841. " xytext=(15,5), # distance from text to points (x,y)\n",
  842. " ha='center') # horizontal alignment can be left, right or center\n",
  843. " figs[2].annotate(prediction_times_ms[i], # this is the text\n",
  844. " (x[i],prediction_times_ms[i]), # these are the coordinates to position the label\n",
  845. " textcoords=\"offset points\", # how to position the text\n",
  846. " xytext=(15,5), # distance from text to points (x,y)\n",
  847. " ha='center') # horizontal alignment can be left, right or center\n",
  848. "\n",
  849. "figs[0].set_xticks(x)\n",
  850. "figs[1].set_xticks(x)\n",
  851. "figs[2].set_xticks(x)\n",
  852. " \n",
  853. "### Add title and axis names\n",
  854. "figs[0].title.set_text('Scores for various n_jobs (k=3,train_size=0.9,dataset_size=25000,distance=euclidean)')\n",
  855. "figs[1].title.set_text('Training times for various n_jobs (k=3,train_size=0.9,dataset_size=25000,distance=euclidean)')\n",
  856. "figs[2].title.set_text('Prediction times for various n_jobs (k=3,train_size=0.9,dataset_size=25000,distance=euclidean)')\n",
  857. "figs[0].set_xlabel('n_jobs')\n",
  858. "figs[1].set_xlabel('n_jobs')\n",
  859. "figs[2].set_xlabel('n_jobs')\n",
  860. "figs[0].set_ylabel('Score')\n",
  861. "figs[1].set_ylabel('Times (in ms)')\n",
  862. "figs[2].set_ylabel('Times (in ms)')"
  863. ]
  864. },
  865. {
  866. "cell_type": "code",
  867. "execution_count": 26,
  868. "id": "98107e41",
  869. "metadata": {},
  870. "outputs": [
  871. {
  872. "name": "stdout",
  873. "output_type": "stream",
  874. "text": [
  875. "Métriques pour K-NN :\n",
  876. "Paramètres : (n_neighbors=3,p=2,n_jobs=1)\n",
  877. "Taille de l'échantillon : 10000\n",
  878. "Proportion des datasets : 90%\n",
  879. "Temps d'entraînement (secondes) : 0.01596\n",
  880. "Temps de prédiction (secondes) : 0.30718\n",
  881. "Précision pour chaque classe : [0.942, 0.891, 0.962, 0.959, 0.988, 0.944, 0.961, 0.97, 0.989, 0.918]\n",
  882. "Précision : 0.95\n",
  883. "Erreur : 0.05\n",
  884. "Matrice de confusion :\n",
  885. " [[ 98 0 1 0 0 0 1 0 1 0]\n",
  886. " [ 0 114 0 0 0 0 0 0 0 0]\n",
  887. " [ 2 2 102 0 0 0 0 0 0 0]\n",
  888. " [ 0 1 1 93 0 3 0 1 0 0]\n",
  889. " [ 1 5 0 0 82 0 0 0 0 5]\n",
  890. " [ 0 1 1 1 0 84 3 0 0 1]\n",
  891. " [ 0 0 0 0 0 0 99 0 0 0]\n",
  892. " [ 0 3 0 0 0 0 0 97 0 2]\n",
  893. " [ 2 1 1 3 1 2 0 0 92 0]\n",
  894. " [ 1 1 0 0 0 0 0 2 0 89]]\n"
  895. ]
  896. }
  897. ],
  898. "source": [
  899. "### Create vector of 5000 random indexes\n",
  900. "rand_indexes = np.random.randint(70000, size=10000)\n",
  901. "### Load data with the previous vector\n",
  902. "data = mnist.data[rand_indexes]\n",
  903. "target = mnist.target[rand_indexes]\n",
  904. "# Split the dataset\n",
  905. "xtrain, xtest, ytrain, ytest = model_selection.train_test_split(data, target,train_size=0.9)\n",
  906. "\n",
  907. "clf = neighbors.KNeighborsClassifier(n_neighbors=3,p=2,n_jobs=1)\n",
  908. "# Training on xtrain,ytrain\n",
  909. "t1 = time.time()\n",
  910. "clf.fit(xtrain, ytrain)\n",
  911. "t2 = time.time()\n",
  912. "# Predicting on xtest\n",
  913. "pred = clf.predict(xtest)\n",
  914. "t3 = time.time()\n",
  915. "#Calcul de différentes metrics\n",
  916. "precisions = [round(i,3) for i in metrics.precision_score(ytest, pred,average=None)]\n",
  917. "\n",
  918. "print(\"Métriques pour K-NN :\")\n",
  919. "print(\"Paramètres : (n_neighbors=3,p=2,n_jobs=1)\")\n",
  920. "print(\"Taille de l'échantillon :\", 10000)\n",
  921. "print(\"Proportion des datasets :\", \"90%\")\n",
  922. "print(\"Temps d'entraînement (secondes) :\", round(t2-t1,5))\n",
  923. "print(\"Temps de prédiction (secondes) :\", round(t3-t2,5))\n",
  924. "print(\"Précision pour chaque classe :\", precisions)\n",
  925. "print(\"Précision :\", clf.score(xtest, ytest))\n",
  926. "print(\"Erreur :\", round(metrics.zero_one_loss(ytest, pred),5))\n",
  927. "print(\"Matrice de confusion :\\n\", metrics.confusion_matrix(ytest, pred))"
  928. ]
  929. },
  930. {
  931. "cell_type": "code",
  932. "execution_count": null,
  933. "id": "a45453b9",
  934. "metadata": {},
  935. "outputs": [],
  936. "source": []
  937. }
  938. ],
  939. "metadata": {
  940. "kernelspec": {
  941. "display_name": "Python 3",
  942. "language": "python",
  943. "name": "python3"
  944. },
  945. "language_info": {
  946. "codemirror_mode": {
  947. "name": "ipython",
  948. "version": 3
  949. },
  950. "file_extension": ".py",
  951. "mimetype": "text/x-python",
  952. "name": "python",
  953. "nbconvert_exporter": "python",
  954. "pygments_lexer": "ipython3",
  955. "version": "3.8.8"
  956. }
  957. },
  958. "nbformat": 4,
  959. "nbformat_minor": 5
  960. }