API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Aug 18th, 2019
90
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 9.36 KB | None | 0 0
raw download clone embed print report
  1. {
  2. "nbformat_minor": 1,
  3. "cells": [
  4. {
  5. "execution_count": null,
  6. "cell_type": "code",
  7. "metadata": {},
  8. "outputs": [],
  9. "source": "from sklearn.model_selection import train_test_split\nfrom sklearn.ensemble import ExtraTreesClassifier,GradientBoostingClassifier\nfrom sklearn.svm import SVC\nfrom sklearn.ensemble import VotingClassifier\nfrom sklearn.metrics import accuracy_score\nimport numpy as np\nimport pandas as pd"
  10. },
  11. {
  12. "execution_count": null,
  13. "cell_type": "code",
  14. "metadata": {
  15. "scrolled": true
  16. },
  17. "outputs": [],
  18. "source": "# The code was removed by Watson Studio for sharing."
  19. },
  20. {
  21. "execution_count": null,
  22. "cell_type": "code",
  23. "metadata": {},
  24. "outputs": [],
  25. "source": "df_data_1['address']=df_data_1['addr_state'].astype(str)+df_data_1['zip_code'].astype(str)"
  26. },
  27. {
  28. "execution_count": null,
  29. "cell_type": "code",
  30. "metadata": {},
  31. "outputs": [],
  32. "source": "removal_cols=['mths_since_last_major_derog',\n 'zip_code',\n 'application_type',\n 'mths_since_last_record',\n 'addr_state',\n 'mths_since_last_delinq']\npreserve_cols=[\"emp_title\",\"emp_length\",\"address\",\"last_week_pay\"]\nstring_cols=[\"term\",\"batch_enrolled\",\"grade\",\"sub_grade\"]"
  33. },
  34. {
  35. "execution_count": null,
  36. "cell_type": "code",
  37. "metadata": {},
  38. "outputs": [],
  39. "source": "df_final=df_data_1[preserve_cols].copy(deep=True)\nremoval_cols+=preserve_cols"
  40. },
  41. {
  42. "execution_count": null,
  43. "cell_type": "code",
  44. "metadata": {},
  45. "outputs": [],
  46. "source": "df_data_1.drop(removal_cols,axis=1,inplace=True)"
  47. },
  48. {
  49. "execution_count": null,
  50. "cell_type": "code",
  51. "metadata": {},
  52. "outputs": [],
  53. "source": "df_data_1.head()"
  54. },
  55. {
  56. "execution_count": null,
  57. "cell_type": "code",
  58. "metadata": {},
  59. "outputs": [],
  60. "source": "def replace_missing_values(df):\n return df_data_1.apply(lambda x: x.fillna(0) if x.dtype.kind in 'biufc' else x.fillna(' '))\n "
  61. },
  62. {
  63. "execution_count": null,
  64. "cell_type": "code",
  65. "metadata": {},
  66. "outputs": [],
  67. "source": "df_data_1=df_data_1[df_data_1['batch_enrolled']!=' ']"
  68. },
  69. {
  70. "execution_count": null,
  71. "cell_type": "code",
  72. "metadata": {},
  73. "outputs": [],
  74. "source": "df=replace_missing_values(df_data_1)"
  75. },
  76. {
  77. "execution_count": null,
  78. "cell_type": "code",
  79. "metadata": {},
  80. "outputs": [],
  81. "source": "df['no_funded_amnt']=df['loan_amnt']==df['funded_amnt']"
  82. },
  83. {
  84. "execution_count": null,
  85. "cell_type": "code",
  86. "metadata": {},
  87. "outputs": [],
  88. "source": "df=pd.get_dummies(df,columns=string_cols)"
  89. },
  90. {
  91. "execution_count": null,
  92. "cell_type": "code",
  93. "metadata": {},
  94. "outputs": [],
  95. "source": "target_col=df['home_ownership']\ndf.drop('home_ownership',axis=1,inplace=True)"
  96. },
  97. {
  98. "execution_count": null,
  99. "cell_type": "code",
  100. "metadata": {},
  101. "outputs": [],
  102. "source": "xtrain,xtest,ytrain,ytest=train_test_split(df,target_col,test_size=0.20, random_state=42)"
  103. },
  104. {
  105. "execution_count": null,
  106. "cell_type": "code",
  107. "metadata": {},
  108. "outputs": [],
  109. "source": "xg_clf=ExtraTreesClassifier(n_estimators=20)\ng_clf=GradientBoostingClassifier(n_estimators=40,learning_rate=0.001)\nsvm_clf = SVC(kernel=\"rbf\")\nv_clf=VotingClassifier(estimators=[('xg',xg_clf),('gb',g_clf),('svc',svm_clf)],voting='hard')"
  110. },
  111. {
  112. "execution_count": null,
  113. "cell_type": "code",
  114. "metadata": {
  115. "scrolled": true
  116. },
  117. "outputs": [],
  118. "source": "v_clf.fit(xtrain,ytrain)"
  119. },
  120. {
  121. "execution_count": null,
  122. "cell_type": "code",
  123. "metadata": {
  124. "scrolled": true
  125. },
  126. "outputs": [],
  127. "source": "for clf in (xg_clf, g_clf, svm_clf, v_clf):\n clf.fit(xtrain, ytrain)\n ypred = clf.predict(xtest)\n print(clf.__class__.__name__, accuracy_score(ytest, ypred))"
  128. },
  129. {
  130. "execution_count": null,
  131. "cell_type": "code",
  132. "metadata": {},
  133. "outputs": [],
  134. "source": ""
  135. },
  136. {
  137. "execution_count": null,
  138. "cell_type": "code",
  139. "metadata": {},
  140. "outputs": [],
  141. "source": ""
  142. },
  143. {
  144. "execution_count": null,
  145. "cell_type": "code",
  146. "metadata": {},
  147. "outputs": [],
  148. "source": ""
  149. },
  150. {
  151. "execution_count": null,
  152. "cell_type": "code",
  153. "metadata": {},
  154. "outputs": [],
  155. "source": ""
  156. },
  157. {
  158. "execution_count": null,
  159. "cell_type": "code",
  160. "metadata": {},
  161. "outputs": [],
  162. "source": ""
  163. },
  164. {
  165. "execution_count": null,
  166. "cell_type": "code",
  167. "metadata": {},
  168. "outputs": [],
  169. "source": ""
  170. },
  171. {
  172. "execution_count": null,
  173. "cell_type": "code",
  174. "metadata": {},
  175. "outputs": [],
  176. "source": ""
  177. },
  178. {
  179. "execution_count": null,
  180. "cell_type": "code",
  181. "metadata": {},
  182. "outputs": [],
  183. "source": ""
  184. },
  185. {
  186. "execution_count": null,
  187. "cell_type": "code",
  188. "metadata": {},
  189. "outputs": [],
  190. "source": ""
  191. },
  192. {
  193. "execution_count": null,
  194. "cell_type": "code",
  195. "metadata": {},
  196. "outputs": [],
  197. "source": ""
  198. },
  199. {
  200. "execution_count": null,
  201. "cell_type": "code",
  202. "metadata": {},
  203. "outputs": [],
  204. "source": ""
  205. },
  206. {
  207. "execution_count": null,
  208. "cell_type": "code",
  209. "metadata": {},
  210. "outputs": [],
  211. "source": ""
  212. },
  213. {
  214. "execution_count": null,
  215. "cell_type": "code",
  216. "metadata": {},
  217. "outputs": [],
  218. "source": ""
  219. },
  220. {
  221. "execution_count": null,
  222. "cell_type": "code",
  223. "metadata": {},
  224. "outputs": [],
  225. "source": ""
  226. },
  227. {
  228. "execution_count": null,
  229. "cell_type": "code",
  230. "metadata": {},
  231. "outputs": [],
  232. "source": ""
  233. },
  234. {
  235. "execution_count": null,
  236. "cell_type": "code",
  237. "metadata": {},
  238. "outputs": [],
  239. "source": ""
  240. },
  241. {
  242. "execution_count": null,
  243. "cell_type": "code",
  244. "metadata": {},
  245. "outputs": [],
  246. "source": ""
  247. },
  248. {
  249. "execution_count": null,
  250. "cell_type": "code",
  251. "metadata": {},
  252. "outputs": [],
  253. "source": ""
  254. },
  255. {
  256. "execution_count": null,
  257. "cell_type": "code",
  258. "metadata": {},
  259. "outputs": [],
  260. "source": ""
  261. },
  262. {
  263. "execution_count": null,
  264. "cell_type": "code",
  265. "metadata": {},
  266. "outputs": [],
  267. "source": ""
  268. },
  269. {
  270. "execution_count": null,
  271. "cell_type": "code",
  272. "metadata": {},
  273. "outputs": [],
  274. "source": ""
  275. },
  276. {
  277. "execution_count": null,
  278. "cell_type": "code",
  279. "metadata": {},
  280. "outputs": [],
  281. "source": ""
  282. },
  283. {
  284. "execution_count": null,
  285. "cell_type": "code",
  286. "metadata": {},
  287. "outputs": [],
  288. "source": ""
  289. }
  290. ],
  291. "metadata": {
  292. "kernelspec": {
  293. "display_name": "Python 3.6",
  294. "name": "python3",
  295. "language": "python"
  296. },
  297. "language_info": {
  298. "mimetype": "text/x-python",
  299. "nbconvert_exporter": "python",
  300. "version": "3.6.8",
  301. "name": "python",
  302. "file_extension": ".py",
  303. "pygments_lexer": "ipython3",
  304. "codemirror_mode": {
  305. "version": 3,
  306. "name": "ipython"
  307. }
  308. }
  309. },
  310. "nbformat": 4
  311. }
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!