API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Aug 25th, 2017
3,787
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 111.59 KB | None | 0 0
raw download clone embed print report
  1. model.fcn16 __init__:13 :3.9 Mb super().__init__(n_classes)
  2. start maybe_cuda:85 :1370.9 Mb return x.cuda() if torch.cuda.is_available() else x
  3. __main__ run_adaptation:314 :1370.9 Mb shift = -3
  4. __main__ run_adaptation:316 :1370.9 Mb setup, conf, stdout_tee = setup_config.get_setup_conf_log_dump(
  5. __main__ run_adaptation:317 :1370.9 Mb distance_name, path='./logs/'
  6. setup_config get_setup_conf_log_dump:72 :1370.9 Mb if '___' in setup_name:
  7. setup_config get_setup_conf_log_dump:75 :1370.9 Mb load_dict_name, log_fn_format = setup_name, setup_name
  8. setup_config get_setup_conf_log_dump:77 :1370.9 Mb setup_w_dump = get_setup_and_dump(load_dict_name, override)
  9. setup_config get_setup_and_dump:59 :1370.9 Mb override = override or {}
  10. setup_config get_setup_and_dump:60 :1370.9 Mb setup_dict = {'data': {}} # defaults
  11. setup_config get_setup_and_dump:61 :1370.9 Mb setup_dict = deep_dict_update(setup_dict, distance_config_[setup_name])
  12. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  13. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  14. setup_config deep_dict_update:20 :1370.9 Mb d[k] = u[k]
  15. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  16. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  17. setup_config deep_dict_update:17 :1370.9 Mb r = deep_dict_update(d.get(k, {}), v)
  18. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  19. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  20. setup_config deep_dict_update:20 :1370.9 Mb d[k] = u[k]
  21. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  22. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  23. setup_config deep_dict_update:20 :1370.9 Mb d[k] = u[k]
  24. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  25. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  26. setup_config deep_dict_update:20 :1370.9 Mb d[k] = u[k]
  27. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  28. setup_config deep_dict_update:21 :1370.9 Mb return d
  29. setup_config deep_dict_update:18 :1370.9 Mb d[k] = r
  30. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  31. setup_config deep_dict_update:16 :1370.9 Mb if isinstance(v, collections.Mapping):
  32. setup_config deep_dict_update:20 :1370.9 Mb d[k] = u[k]
  33. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  34. setup_config deep_dict_update:21 :1370.9 Mb return d
  35. setup_config get_setup_and_dump:62 :1370.9 Mb setup_dict = deep_dict_update(setup_dict, override)
  36. setup_config deep_dict_update:15 :1370.9 Mb for k, v in u.items():
  37. setup_config deep_dict_update:21 :1370.9 Mb return d
  38. setup_config get_setup_and_dump:64 :1370.9 Mb 'optimizer': _exec_config(setup_dict.get('optimizer', None)),
  39. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  40. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  41. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  42. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  43. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  44. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  45. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  46. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  47. setup_config _exec_config:46 :1370.9 Mb return packed_object
  48. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  49. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  50. setup_config _exec_config:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  51. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  52. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  53. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  54. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  55. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  56. setup_config _exec_config:46 :1370.9 Mb return packed_object
  57. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  58. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  59. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  60. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  61. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  62. setup_config _exec_config:46 :1370.9 Mb return packed_object
  63. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  64. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  65. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  66. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  67. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  68. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  69. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  70. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  71. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  72. setup_config _exec_config:46 :1370.9 Mb return packed_object
  73. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  74. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  75. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  76. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  77. setup_config _exec_config:46 :1370.9 Mb return packed_object
  78. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  79. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  80. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  81. setup_config _exec_config:46 :1370.9 Mb return packed_object
  82. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  83. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  84. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  85. setup_config _exec_config:46 :1370.9 Mb return packed_object
  86. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  87. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  88. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  89. setup_config _exec_config:39 :1370.9 Mb and type(packed_object[1]) is dict):
  90. setup_config _exec_config:40 :1370.9 Mb repacked_object = partial(packed_object[0], **packed_object[1])
  91. setup_config _exec_config:41 :1370.9 Mb if isinstance(packed_object[0], types.FunctionType):
  92. setup_config _exec_config:44 :1370.9 Mb return repacked_object
  93. setup_config get_setup_and_dump:65 :1370.9 Mb 'distance': _exec_config(setup_dict.get('distance', None)),
  94. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  95. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  96. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  97. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  98. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  99. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  100. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  101. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  102. setup_config _exec_config:46 :1370.9 Mb return packed_object
  103. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  104. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  105. setup_config _exec_config:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  106. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  107. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  108. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  109. setup_config _exec_config:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  110. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  111. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  112. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  113. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  114. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  115. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  116. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  117. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  118. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  119. setup_config _exec_config:46 :1370.9 Mb return packed_object
  120. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  121. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  122. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  123. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  124. setup_config _exec_config:46 :1370.9 Mb return packed_object
  125. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  126. setup_config _exec_config:46 :1370.9 Mb return packed_object
  127. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  128. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  129. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  130. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  131. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  132. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  133. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  134. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  135. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  136. setup_config _exec_config:46 :1370.9 Mb return packed_object
  137. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  138. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  139. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  140. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  141. setup_config _exec_config:46 :1370.9 Mb return packed_object
  142. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  143. setup_config _exec_config:46 :1370.9 Mb return packed_object
  144. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  145. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  146. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  147. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  148. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  149. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  150. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  151. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  152. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  153. setup_config _exec_config:46 :1370.9 Mb return packed_object
  154. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  155. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  156. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  157. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  158. setup_config _exec_config:46 :1370.9 Mb return packed_object
  159. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  160. setup_config _exec_config:46 :1370.9 Mb return packed_object
  161. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  162. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  163. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  164. setup_config _exec_config:46 :1370.9 Mb return packed_object
  165. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  166. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  167. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  168. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  169. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  170. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  171. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  172. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  173. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  174. setup_config _exec_config:46 :1370.9 Mb return packed_object
  175. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  176. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  177. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  178. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  179. setup_config _exec_config:46 :1370.9 Mb return packed_object
  180. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  181. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  182. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  183. setup_config _exec_config:46 :1370.9 Mb return packed_object
  184. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  185. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  186. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  187. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  188. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  189. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  190. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  191. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  192. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  193. setup_config _exec_config:46 :1370.9 Mb return packed_object
  194. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  195. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  196. setup_config _exec_config:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  197. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  198. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  199. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  200. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  201. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  202. setup_config _exec_config:46 :1370.9 Mb return packed_object
  203. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  204. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  205. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  206. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  207. setup_config _exec_config:34 :1370.9 Mb packed_object = type(packed_object)(map(_exec_config, packed_object))
  208. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  209. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  210. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  211. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  212. setup_config _exec_config:46 :1370.9 Mb return packed_object
  213. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  214. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  215. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  216. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  217. setup_config _exec_config:46 :1370.9 Mb return packed_object
  218. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  219. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  220. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  221. setup_config _exec_config:46 :1370.9 Mb return packed_object
  222. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  223. setup_config _exec_config:25 :1370.9 Mb if packed_object is None:
  224. setup_config _exec_config:30 :1370.9 Mb if type(packed_object) is dict:
  225. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  226. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  227. setup_config _exec_config:46 :1370.9 Mb return packed_object
  228. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  229. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  230. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  231. setup_config _exec_config:46 :1370.9 Mb return packed_object
  232. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  233. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  234. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  235. setup_config _exec_config:39 :1370.9 Mb and type(packed_object[1]) is dict):
  236. setup_config _exec_config:40 :1370.9 Mb repacked_object = partial(packed_object[0], **packed_object[1])
  237. setup_config _exec_config:41 :1370.9 Mb if isinstance(packed_object[0], types.FunctionType):
  238. setup_config _exec_config:44 :1370.9 Mb return repacked_object
  239. setup_config <dictcomp>:31 :1370.9 Mb packed_object = {k: _exec_config(v) for k, v in packed_object.items()}
  240. setup_config _exec_config:33 :1370.9 Mb if isinstance(packed_object, (list, tuple)):
  241. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  242. setup_config _exec_config:46 :1370.9 Mb return packed_object
  243. setup_config _exec_config:36 :1370.9 Mb if (type(packed_object) is tuple
  244. setup_config _exec_config:37 :1370.9 Mb and len(packed_object) == 2
  245. setup_config _exec_config:38 :1370.9 Mb and callable(packed_object[0])
  246. setup_config _exec_config:39 :1370.9 Mb and type(packed_object[1]) is dict):
  247. setup_config _exec_config:40 :1370.9 Mb repacked_object = partial(packed_object[0], **packed_object[1])
  248. setup_config _exec_config:41 :1370.9 Mb if isinstance(packed_object[0], types.FunctionType):
  249. setup_config _exec_config:44 :1370.9 Mb return repacked_object
  250. setup_config get_setup_and_dump:66 :1370.9 Mb 'data': lambda: read_pytorch(**setup_dict['data']),
  251. setup_config get_setup_and_dump:67 :1370.9 Mb }, setup_dict)
  252. setup_config get_setup_conf_log_dump:78 :1370.9 Mb time_str = datetime.datetime.now().__format__(time_fmt)
  253. setup_config get_setup_conf_log_dump:79 :1370.9 Mb os.makedirs(path, exist_ok=True)
  254. setup_config get_setup_conf_log_dump:80 :1370.9 Mb fn = os.path.join(path, fn_fmt.format(setup=log_fn_format, now=time_str))
  255. setup_config get_setup_conf_log_dump:81 :1370.9 Mb with open(fn, 'w') as f:
  256. setup_config get_setup_conf_log_dump:82 :1370.9 Mb f.write(custom_json.dumps(setup_w_dump[1], indent=2))
  257. utils.custom_json default:14 :1370.9 Mb if isinstance(obj, type):
  258. utils.custom_json default:15 :1370.9 Mb return {'_type': (obj.__module__, obj.__name__)}
  259. utils.custom_json default:14 :1370.9 Mb if isinstance(obj, type):
  260. utils.custom_json default:15 :1370.9 Mb return {'_type': (obj.__module__, obj.__name__)}
  261. utils.custom_json default:14 :1370.9 Mb if isinstance(obj, type):
  262. utils.custom_json default:15 :1370.9 Mb return {'_type': (obj.__module__, obj.__name__)}
  263. setup_config get_setup_conf_log_dump:83 :1370.9 Mb tee_context_manager = tee_stdout(fn+'.log')
  264. setup_config get_setup_conf_log_dump:84 :1370.9 Mb return setup_w_dump[0], setup_w_dump[1], tee_context_manager
  265. __main__ run_adaptation:320 :1370.9 Mb distance_function = setup['distance']().cuda()
  266. distances.mlp __init__:16 :1370.9 Mb super().__init__(shapes, *argv, **kwargs)
  267. distances.mlp_base __init__:18 :1370.9 Mb super().__init__()
  268. distances.mlp_base __init__:19 :1370.9 Mb self.full_data_pretrain_n = full_data_pretrain_n
  269. distances.mlp_base __init__:20 :1370.9 Mb self._inited = False
  270. distances.mlp_base __init__:21 :1370.9 Mb self._step_every_repeat_n = step_every_repeat_n
  271. distances.mlp_base __init__:22 :1370.9 Mb self._shapes = shapes
  272. distances.mlp_base __init__:23 :1370.9 Mb self._optimizer_builder = optimizer_builder
  273. distances.mlp_base __init__:24 :1370.9 Mb self._initializer = initializer
  274. distances.mlp_base __init__:25 :1370.9 Mb self._attempt_on_forward = attempt_update_on_forward
  275. distances.mlp_base __init__:26 :1370.9 Mb self._update_attempt_counter = 0
  276. distances.mlp_base __init__:27 :1370.9 Mb self._start_from_n_updates = start_from_n_updates
  277. distances.mlp_base __init__:28 :1370.9 Mb self._sub_init(kwargs)
  278. distances.mlp_base _sub_init:50 :1370.9 Mb pass
  279. distances.mlp_base __init__:29 :1370.9 Mb self.draw_method_list.append(self.draw_contour)
  280. distances.mlp __init__:17 :1370.9 Mb self.stored_y = None
  281. __main__ run_adaptation:321 :1370.9 Mb optimizer = setup['optimizer'](target_model.parameters())
  282. __main__ run_adaptation:322 :1370.9 Mb if share_embedding:
  283. __main__ run_adaptation:323 :1370.9 Mb base_optimizer = setup['optimizer'](base_model.parameters())
  284. __main__ run_adaptation:325 :1370.9 Mb train_loaders = setup['data']()
  285. setup_config <lambda>:66 :1370.9 Mb 'data': lambda: read_pytorch(**setup_dict['data']),
  286. read_data read_pytorch:107 :1370.9 Mb if dataset == 'svhn2mnist':
  287. read_data read_pytorch:109 :1370.9 Mb elif dataset == 'gta2city':
  288. read_data read_pytorch:110 :1370.9 Mb return read_gta2city(batch_size, cuda, same_batch_n, shuffle, **kwargs)
  289. read_data read_gta2city:210 :1370.9 Mb kwargs = {'num_workers': workers, 'pin_memory': pin_memory} if cuda else {}
  290. read_data read_gta2city:212 :1370.9 Mb assert not (batch_size > 0 and same_batch_n > 1), \
  291. read_data read_gta2city:215 :1370.9 Mb if batch_size > 0 and same_batch_n == 1:
  292. read_data read_gta2city:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  293. read_data read_gta2city:224 :1370.9 Mb for split_name in ['train', 'test']
  294. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  295. read_data <dictcomp>:224 :1370.9 Mb for split_name in ['train', 'test']
  296. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  297. read_data <dictcomp>:224 :1370.9 Mb for split_name in ['train', 'test']
  298. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  299. read_data read_gta2city:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  300. read_data read_gta2city:236 :1370.9 Mb for split_name in ['train', 'test']
  301. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  302. read_data <dictcomp>:236 :1370.9 Mb for split_name in ['train', 'test']
  303. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  304. read_data <dictcomp>:236 :1370.9 Mb for split_name in ['train', 'test']
  305. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  306. read_data read_gta2city:241 :1370.9 Mb return source_loader, target_loader
  307. __main__ run_adaptation:326 :1370.9 Mb eval_loaders = setup['data']()
  308. setup_config <lambda>:66 :1370.9 Mb 'data': lambda: read_pytorch(**setup_dict['data']),
  309. read_data read_pytorch:107 :1370.9 Mb if dataset == 'svhn2mnist':
  310. read_data read_pytorch:109 :1370.9 Mb elif dataset == 'gta2city':
  311. read_data read_pytorch:110 :1370.9 Mb return read_gta2city(batch_size, cuda, same_batch_n, shuffle, **kwargs)
  312. read_data read_gta2city:210 :1370.9 Mb kwargs = {'num_workers': workers, 'pin_memory': pin_memory} if cuda else {}
  313. read_data read_gta2city:212 :1370.9 Mb assert not (batch_size > 0 and same_batch_n > 1), \
  314. read_data read_gta2city:215 :1370.9 Mb if batch_size > 0 and same_batch_n == 1:
  315. read_data read_gta2city:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  316. read_data read_gta2city:224 :1370.9 Mb for split_name in ['train', 'test']
  317. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  318. read_data <dictcomp>:224 :1370.9 Mb for split_name in ['train', 'test']
  319. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  320. read_data <dictcomp>:224 :1370.9 Mb for split_name in ['train', 'test']
  321. read_data <dictcomp>:216 :1370.9 Mb source_loader: Mapping[str, torch.utils.data.DataLoader] = {
  322. read_data read_gta2city:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  323. read_data read_gta2city:236 :1370.9 Mb for split_name in ['train', 'test']
  324. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  325. read_data <dictcomp>:236 :1370.9 Mb for split_name in ['train', 'test']
  326. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  327. read_data <dictcomp>:236 :1370.9 Mb for split_name in ['train', 'test']
  328. read_data <dictcomp>:227 :1370.9 Mb target_loader: Mapping[str, torch.utils.data.DataLoader] = {
  329. read_data read_gta2city:241 :1370.9 Mb return source_loader, target_loader
  330. __main__ run_adaptation:328 :1370.9 Mb with stdout_tee:
  331. utils.tee_stdout tee_stdout:22 :1370.9 Mb print('teeing std to', f_names)
  332. utils.tee_stdout tee_stdout:23 :1370.9 Mb f_objects = [open(f_name, 'w') for f_name in f_names]
  333. utils.tee_stdout <listcomp>:23 :1370.9 Mb f_objects = [open(f_name, 'w') for f_name in f_names]
  334. utils.tee_stdout <listcomp>:23 :1370.9 Mb f_objects = [open(f_name, 'w') for f_name in f_names]
  335. utils.tee_stdout tee_stdout:24 :1370.9 Mb with contextlib.redirect_stdout(MergedIO(sys.stdout, *f_objects)):
  336. utils.tee_stdout __init__:9 :1370.9 Mb self._file_objects = file_objects
  337. utils.tee_stdout tee_stdout:25 :1370.9 Mb yield
  338. __main__ run_adaptation:329 :1370.9 Mb print(conf)
  339. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  340. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  341. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  342. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  343. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  344. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  345. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  346. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  347. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  348. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  349. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  350. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  351. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  352. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  353. __main__ run_adaptation:330 :1370.9 Mb print(snapshot_name)
  354. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  355. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  356. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  357. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  358. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  359. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  360. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  361. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  362. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  363. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  364. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  365. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  366. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  367. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  368. __main__ run_adaptation:332 :1370.9 Mb run_iterations(distance_function, distance_name, eval_loaders, shift,
  369. __main__ run_adaptation:333 :1370.9 Mb train_loaders)
  370. __main__ run_iterations:340 :1370.9 Mb for sub_epoch_i in range(total_epochs):
  371. __main__ run_iterations:348 :1370.9 Mb if eval_on_first_iter is True and sub_epoch_i == 0:
  372. __main__ run_iterations:351 :1370.9 Mb print(f'{sub_epoch_i:>3} updating')
  373. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  374. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  375. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  376. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  377. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  378. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  379. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  380. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  381. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  382. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  383. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  384. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  385. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  386. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  387. __main__ run_iterations:353 :1370.9 Mb if sub_epoch_i == 0:
  388. __main__ run_iterations:354 :1370.9 Mb repeat_learning_i = getattr(distance_function, 'full_data_pretrain_n', 1)
  389. __main__ run_iterations:358 :1370.9 Mb for sub_repeat in range(repeat_learning_i):
  390. __main__ run_iterations:359 :1370.9 Mb pretraining = repeat_learning_i > 1
  391. __main__ run_iterations:360 :1370.9 Mb sub_match_iterator = match_source_target(
  392. __main__ run_iterations:361 :1370.9 Mb sub_epoch_i, train_loaders, distance_function, shift=shift,
  393. __main__ run_iterations:362 :1370.9 Mb discriminator_only=pretraining
  394. __main__ run_iterations:365 :1370.9 Mb print('supervised update')
  395. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  396. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  397. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  398. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  399. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  400. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  401. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  402. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  403. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  404. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  405. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  406. utils.tee_stdout write:13 :1370.9 Mb f.write(string)
  407. utils.tee_stdout write:14 :1370.9 Mb f.flush()
  408. utils.tee_stdout write:12 :1370.9 Mb for f in self._file_objects:
  409. __main__ run_iterations:367 :1370.9 Mb if share_embedding is True:
  410. __main__ run_iterations:368 :1370.9 Mb sub_base_iterator = supervised_source_update(train_loaders[0]['train'])
  411. __main__ run_iterations:369 :1370.9 Mb sub_iterator = zip(sub_match_iterator, sub_base_iterator)
  412. __main__ run_iterations:373 :1370.9 Mb for __ in sub_iterator:
  413. __main__ match_source_target:107 :1370.9 Mb dist_func.train(True)
  414. __main__ match_source_target:109 :1370.9 Mb discriminator_needs_update = hasattr(dist_func, 'attempt_update_d')
  415. __main__ match_source_target:111 :1370.9 Mb if not discriminator_needs_update and discriminator_only:
  416. __main__ match_source_target:115 :1370.9 Mb if discriminator_only is True:
  417. __main__ match_source_target:118 :1370.9 Mb data_len = len(loaders[0]['train'])
  418. __main__ match_source_target:119 :1370.9 Mb yield_every = data_len // update_report_per_epoch
  419. __main__ match_source_target:120 :1370.9 Mb adaptation_iterator = enumerate(zip(loaders[0]['train'], loaders[1]['train']))
  420. __main__ match_source_target:121 :1370.9 Mb tqdm_adaptation_iterator = tqdm(adaptation_iterator, total=data_len, desc='match')
  421. __main__ match_source_target:122 :1370.9 Mb for batch_idx, ((source_data, s_y), (target_data, ___)) in tqdm_adaptation_iterator:
  422. __main__ match_source_target:123 :1370.9 Mb if source_data.numel == 0:
  423. __main__ match_source_target:127 :1370.9 Mb dist_func.current_pos = (epoch_i, batch_idx, batch_idx)
  424. __main__ match_source_target:128 :1370.9 Mb dist_func.y_a = s_y
  425. __main__ match_source_target:130 :1370.9 Mb data_s = Variable(maybe_cuda(source_data))
  426. start maybe_cuda:85 :1370.9 Mb return x.cuda() if torch.cuda.is_available() else x
  427. __main__ match_source_target:131 :1370.9 Mb data_t = Variable(maybe_cuda(target_data))
  428. start maybe_cuda:85 :1370.9 Mb return x.cuda() if torch.cuda.is_available() else x
  429. __main__ match_source_target:134 :1370.9 Mb if discriminator_needs_update is True:
  430. __main__ match_source_target:136 :1370.9 Mb feature_apply(partial(dist_func.attempt_update_d, epoch_i=epoch_i),
  431. __main__ match_source_target:137 :1370.9 Mb base_model, target_model, data_s, data_t, shift)
  432. distances.utils feature_apply:103 :1370.9 Mb a_phi = model_a.features_at(input_a, shift)
  433. model.fcn16 features_at:16 :1370.9 Mb assert shift == -3, "!= -3 not implemented, not needed"
  434. model.fcn16 features_at:18 :1370.9 Mb h = x
  435. model.fcn16 features_at:19 :1370.9 Mb h = self.relu1_1(self.conv1_1(h))
  436. model.fcn16 features_at:20 :2118.9 Mb h = self.relu1_2(self.conv1_2(h))
  437. model.fcn16 features_at:21 :2764.9 Mb h = self.pool1(h)
  438. model.fcn16 features_at:23 :3088.9 Mb h = self.relu2_1(self.conv2_1(h))
  439. model.fcn16 features_at:24 :3412.9 Mb h = self.relu2_2(self.conv2_2(h))
  440. model.fcn16 features_at:25 :3736.9 Mb h = self.pool2(h)
  441. model.fcn16 features_at:27 :3898.9 Mb h = self.relu3_1(self.conv3_1(h))
  442. model.fcn16 features_at:28 :4060.9 Mb h = self.relu3_2(self.conv3_2(h))
  443. model.fcn16 features_at:29 :4222.9 Mb h = self.relu3_3(self.conv3_3(h))
  444. model.fcn16 features_at:30 :4384.9 Mb h = self.pool3(h)
  445. model.fcn16 features_at:32 :4384.9 Mb h = self.relu4_1(self.conv4_1(h))
  446. model.fcn16 features_at:33 :4466.9 Mb h = self.relu4_2(self.conv4_2(h))
  447. model.fcn16 features_at:34 :4548.9 Mb h = self.relu4_3(self.conv4_3(h))
  448. model.fcn16 features_at:35 :4630.9 Mb h = self.pool4(h)
  449. model.fcn16 features_at:37 :4672.9 Mb h = self.relu5_1(self.conv5_1(h))
  450. model.fcn16 features_at:38 :4694.9 Mb h = self.relu5_2(self.conv5_2(h))
  451. model.fcn16 features_at:39 :4716.9 Mb h = self.relu5_3(self.conv5_3(h))
  452. model.fcn16 features_at:40 :4738.9 Mb h = self.pool5(h)
  453. model.fcn16 features_at:42 :4738.9 Mb h = self.relu6(self.fc6(h))
  454. model.fcn16 features_at:43 :4966.9 Mb h = self.drop6(h)
  455. model.fcn16 features_at:45 :4966.9 Mb h = self.relu7(self.fc7(h))
  456. model.fcn16 features_at:46 :4966.9 Mb h = self.drop7(h)
  457. model.fcn16 features_at:48 :4966.9 Mb h = self.score_fr(h)
  458. model.fcn16 features_at:50 :4966.9 Mb return h
  459. distances.utils feature_apply:104 :4966.9 Mb b_phi = model_b.features_at(input_b, shift)
  460. model.fcn16 features_at:16 :4966.9 Mb assert shift == -3, "!= -3 not implemented, not needed"
  461. model.fcn16 features_at:18 :4966.9 Mb h = x
  462. model.fcn16 features_at:19 :4966.9 Mb h = self.relu1_1(self.conv1_1(h))
  463. model.fcn16 features_at:20 :5612.9 Mb h = self.relu1_2(self.conv1_2(h))
  464. model.fcn16 features_at:21 :6258.9 Mb h = self.pool1(h)
  465. model.fcn16 features_at:23 :6744.9 Mb h = self.relu2_1(self.conv2_1(h))
  466. model.fcn16 features_at:24 :7068.9 Mb h = self.relu2_2(self.conv2_2(h))
  467. model.fcn16 features_at:25 :7392.9 Mb h = self.pool2(h)
  468. model.fcn16 features_at:27 :7554.9 Mb h = self.relu3_1(self.conv3_1(h))
  469. model.fcn16 features_at:28 :7716.9 Mb h = self.relu3_2(self.conv3_2(h))
  470. model.fcn16 features_at:29 :7878.9 Mb h = self.relu3_3(self.conv3_3(h))
  471. model.fcn16 features_at:30 :8040.9 Mb h = self.pool3(h)
  472. model.fcn16 features_at:32 :8164.9 Mb h = self.relu4_1(self.conv4_1(h))
  473. model.fcn16 features_at:33 :8246.9 Mb h = self.relu4_2(self.conv4_2(h))
  474. model.fcn16 features_at:34 :8328.9 Mb h = self.relu4_3(self.conv4_3(h))
  475. model.fcn16 features_at:35 :8410.9 Mb h = self.pool4(h)
  476. model.fcn16 features_at:37 :8474.9 Mb h = self.relu5_1(self.conv5_1(h))
  477. model.fcn16 features_at:38 :8496.9 Mb h = self.relu5_2(self.conv5_2(h))
  478. model.fcn16 features_at:39 :8518.9 Mb h = self.relu5_3(self.conv5_3(h))
  479. model.fcn16 features_at:40 :8540.9 Mb h = self.pool5(h)
  480. model.fcn16 features_at:42 :8540.9 Mb h = self.relu6(self.fc6(h))
  481. model.fcn16 features_at:43 :8768.9 Mb h = self.drop6(h)
  482. model.fcn16 features_at:45 :8768.9 Mb h = self.relu7(self.fc7(h))
  483. model.fcn16 features_at:46 :8768.9 Mb h = self.drop7(h)
  484. model.fcn16 features_at:48 :8768.9 Mb h = self.score_fr(h)
  485. model.fcn16 features_at:50 :8768.9 Mb return h
  486. distances.utils feature_apply:105 :8768.9 Mb dist_val = apply_func(a_phi, b_phi)
  487. distances.mlp_base attempt_update_d:62 :8768.9 Mb self._check_inputs(features_a, features_b)
  488. distances.mlp_fcn _check_inputs:32 :8768.9 Mb self._maybe_init(features_a)
  489. distances.mlp_fcn _maybe_init:28 :8768.9 Mb if not self._inited:
  490. distances.mlp_fcn _maybe_init:29 :8768.9 Mb self._init_net(some_feature_input.size()[1:])
  491. distances.mlp_fcn _init_net:37 :8768.9 Mb if self._shapes is not None:
  492. distances.mlp_fcn _init_net:38 :8768.9 Mb used_kern_sizes = (n_features[0],) + tuple(self._shapes['channels'])
  493. distances.mlp_fcn _init_net:39 :8768.9 Mb self.cnn_net = self.classifier = nn.Sequential(
  494. distances.mlp_fcn _init_net:40 :8768.9 Mb *chain.from_iterable(
  495. distances.mlp_fcn _init_net:41 :8768.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  496. distances.mlp_fcn _init_net:45 :8768.9 Mb for i in range(len(used_kern_sizes)-1)
  497. distances.mlp_fcn _init_net:47 :8768.9 Mb Flatten()
  498. distances.mlp_fcn <genexpr>:41 :8768.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  499. distances.mlp_fcn <genexpr>:45 :8768.9 Mb for i in range(len(used_kern_sizes)-1)
  500. distances.mlp_fcn <genexpr>:41 :8768.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  501. distances.mlp_fcn <genexpr>:45 :8768.9 Mb for i in range(len(used_kern_sizes)-1)
  502. distances.mlp_fcn <genexpr>:41 :8768.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  503. distances.mlp_fcn _init_net:50 :8768.9 Mb test_pass_var = Variable(torch.zeros(1, *n_features), volatile=True).cuda()
  504. distances.mlp_fcn _init_net:51 :8768.9 Mb feature_n = self.cnn_net(test_pass_var).size(1) # because [1, F*H*W]
  505. distances.mlp_fcn forward:17 :8768.9 Mb return X.view(X.size(0), -1)
  506. distances.mlp_fcn _init_net:52 :8768.9 Mb self.net = nn.Sequential(
  507. distances.mlp_fcn _init_net:53 :8768.9 Mb self.cnn_net,
  508. distances.mlp_fcn _init_net:54 :8768.9 Mb nn.Linear(feature_n, 1)
  509. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  510. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  511. distances.mlp_fcn _init_net:59 :8768.9 Mb self._initializer(w)
  512. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  513. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  514. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  515. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  516. distances.mlp_fcn _init_net:59 :8768.9 Mb self._initializer(w)
  517. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  518. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  519. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  520. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  521. distances.mlp_fcn _init_net:59 :8768.9 Mb self._initializer(w)
  522. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  523. distances.mlp_fcn _init_net:58 :8768.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  524. distances.mlp_fcn _init_net:57 :8768.9 Mb for w in self.parameters():
  525. distances.mlp_fcn _init_net:61 :8768.9 Mb self._d_optimizer = self._optimizer_builder(self.parameters())
  526. distances.mlp_fcn _init_net:62 :8768.9 Mb self._n_features = n_features
  527. distances.mlp_fcn _init_net:63 :8768.9 Mb self.__inited = True
  528. distances.mlp_fcn _check_inputs:33 :8768.9 Mb assert features_a.size()[1:] == features_b.size()[1:] == self._n_features
  529. distances.mlp_base attempt_update_d:64 :8768.9 Mb if self._start_from_n_updates > 0 and epoch_i == 0:
  530. distances.mlp_base attempt_update_d:73 :8768.9 Mb if (self._step_every_repeat_n[0] > 0
  531. distances.mlp_base attempt_update_d:74 :8768.9 Mb and self._update_attempt_counter % self._step_every_repeat_n[0] == 0):
  532. distances.mlp_base attempt_update_d:75 :8768.9 Mb for _ in range(self._step_every_repeat_n[1]):
  533. distances.mlp_base attempt_update_d:76 :8768.9 Mb self.d_update_step(features_a, features_b)
  534. distances.mlp_base d_update_step:54 :8768.9 Mb cross_ent = self.objective(features_a, features_b)
  535. distances.mlp objective:21 :8768.9 Mb full_x = torch.cat([features_a, features_b], 0)
  536. distances.mlp objective:23 :8768.9 Mb if self.stored_y is None:
  537. distances.mlp objective:24 :8768.9 Mb to_concat = [torch.ones(features_a.size(0)), torch.zeros(features_b.size(0))]
  538. distances.mlp objective:25 :8768.9 Mb _y = Variable(torch.cat(to_concat).cuda())
  539. distances.mlp objective:26 :8768.9 Mb self.stored_y = _y
  540. distances.mlp objective:30 :8768.9 Mb full_y = _y if not inverse_labels else (1 - _y)
  541. distances.mlp objective:31 :8768.9 Mb output = self.network_forward(full_x)
  542. distances.mlp_base network_forward:37 :8768.9 Mb return self.net(x)
  543. distances.mlp_fcn forward:17 :8768.9 Mb return X.view(X.size(0), -1)
  544. distances.mlp objective:32 :8798.9 Mb cross_ent = F.binary_cross_entropy(F.sigmoid(output.view(-1)), full_y.view(-1))
  545. distances.mlp objective:33 :8818.9 Mb assert not has_nan(cross_ent)
  546. distances.utils has_nan:83 :8818.9 Mb if type(x) is float:
  547. distances.utils has_nan:86 :8818.9 Mb x = x.data if type(x) is Variable else x
  548. distances.utils has_nan:87 :8818.9 Mb return ((float(torch.sum(x != x)) > 0)
  549. distances.utils has_nan:88 :8818.9 Mb or float(torch.sum(x == np.inf)) > 0)
  550. distances.mlp objective:34 :8818.9 Mb return cross_ent
  551. distances.mlp_base d_update_step:56 :8818.9 Mb self._d_optimizer.zero_grad()
  552. distances.mlp_base d_update_step:58 :8818.9 Mb cross_ent.backward()
  553. distances.mlp_base d_update_step:59 :6284.9 Mb self._d_optimizer.step()
  554. distances.mlp_base attempt_update_d:77 :6284.9 Mb self._update_attempt_counter += 1
  555. distances.mlp_base attempt_update_d:75 :6284.9 Mb for _ in range(self._step_every_repeat_n[1]):
  556. distances.mlp_base attempt_update_d:78 :6284.9 Mb return
  557. distances.utils feature_apply:106 :6284.9 Mb return dist_val
  558. __main__ match_source_target:139 :6284.9 Mb del data_s, data_t
  559. __main__ match_source_target:140 :6284.9 Mb data_s = Variable(maybe_cuda(source_data))
  560. start maybe_cuda:85 :6284.9 Mb return x.cuda() if torch.cuda.is_available() else x
  561. __main__ match_source_target:141 :6284.9 Mb data_t = Variable(maybe_cuda(target_data))
  562. start maybe_cuda:85 :6284.9 Mb return x.cuda() if torch.cuda.is_available() else x
  563. __main__ match_source_target:145 :6284.9 Mb if not discriminator_only:
  564. __main__ match_source_target:146 :6284.9 Mb dist_val = feature_apply(dist_func,
  565. __main__ match_source_target:147 :6284.9 Mb base_model, target_model, data_s, data_t, shift)
  566. distances.utils feature_apply:103 :6284.9 Mb a_phi = model_a.features_at(input_a, shift)
  567. model.fcn16 features_at:16 :6284.9 Mb assert shift == -3, "!= -3 not implemented, not needed"
  568. model.fcn16 features_at:18 :6284.9 Mb h = x
  569. model.fcn16 features_at:19 :6284.9 Mb h = self.relu1_1(self.conv1_1(h))
  570. model.fcn16 features_at:20 :6284.9 Mb h = self.relu1_2(self.conv1_2(h))
  571. model.fcn16 features_at:21 :6284.9 Mb h = self.pool1(h)
  572. model.fcn16 features_at:23 :6284.9 Mb h = self.relu2_1(self.conv2_1(h))
  573. model.fcn16 features_at:24 :6284.9 Mb h = self.relu2_2(self.conv2_2(h))
  574. model.fcn16 features_at:25 :6284.9 Mb h = self.pool2(h)
  575. model.fcn16 features_at:27 :6284.9 Mb h = self.relu3_1(self.conv3_1(h))
  576. model.fcn16 features_at:28 :6284.9 Mb h = self.relu3_2(self.conv3_2(h))
  577. model.fcn16 features_at:29 :6284.9 Mb h = self.relu3_3(self.conv3_3(h))
  578. model.fcn16 features_at:30 :6284.9 Mb h = self.pool3(h)
  579. model.fcn16 features_at:32 :6284.9 Mb h = self.relu4_1(self.conv4_1(h))
  580. model.fcn16 features_at:33 :6284.9 Mb h = self.relu4_2(self.conv4_2(h))
  581. model.fcn16 features_at:34 :6284.9 Mb h = self.relu4_3(self.conv4_3(h))
  582. model.fcn16 features_at:35 :6284.9 Mb h = self.pool4(h)
  583. model.fcn16 features_at:37 :6284.9 Mb h = self.relu5_1(self.conv5_1(h))
  584. model.fcn16 features_at:38 :6284.9 Mb h = self.relu5_2(self.conv5_2(h))
  585. model.fcn16 features_at:39 :6284.9 Mb h = self.relu5_3(self.conv5_3(h))
  586. model.fcn16 features_at:40 :6284.9 Mb h = self.pool5(h)
  587. model.fcn16 features_at:42 :6284.9 Mb h = self.relu6(self.fc6(h))
  588. model.fcn16 features_at:43 :6284.9 Mb h = self.drop6(h)
  589. model.fcn16 features_at:45 :6284.9 Mb h = self.relu7(self.fc7(h))
  590. model.fcn16 features_at:46 :6284.9 Mb h = self.drop7(h)
  591. model.fcn16 features_at:48 :6284.9 Mb h = self.score_fr(h)
  592. model.fcn16 features_at:50 :6284.9 Mb return h
  593. distances.utils feature_apply:104 :6284.9 Mb b_phi = model_b.features_at(input_b, shift)
  594. model.fcn16 features_at:16 :6284.9 Mb assert shift == -3, "!= -3 not implemented, not needed"
  595. model.fcn16 features_at:18 :6284.9 Mb h = x
  596. model.fcn16 features_at:19 :6284.9 Mb h = self.relu1_1(self.conv1_1(h))
  597. model.fcn16 features_at:20 :6284.9 Mb h = self.relu1_2(self.conv1_2(h))
  598. model.fcn16 features_at:21 :6930.9 Mb h = self.pool1(h)
  599. model.fcn16 features_at:23 :7254.9 Mb h = self.relu2_1(self.conv2_1(h))
  600. model.fcn16 features_at:24 :7578.9 Mb h = self.relu2_2(self.conv2_2(h))
  601. model.fcn16 features_at:25 :7902.9 Mb h = self.pool2(h)
  602. model.fcn16 features_at:27 :8146.9 Mb h = self.relu3_1(self.conv3_1(h))
  603. model.fcn16 features_at:28 :8308.9 Mb h = self.relu3_2(self.conv3_2(h))
  604. model.fcn16 features_at:29 :8470.9 Mb h = self.relu3_3(self.conv3_3(h))
  605. model.fcn16 features_at:30 :8632.9 Mb h = self.pool3(h)
  606. model.fcn16 features_at:32 :8756.9 Mb h = self.relu4_1(self.conv4_1(h))
  607. model.fcn16 features_at:33 :8838.9 Mb h = self.relu4_2(self.conv4_2(h))
  608. model.fcn16 features_at:34 :8920.9 Mb h = self.relu4_3(self.conv4_3(h))
  609. model.fcn16 features_at:35 :9002.9 Mb h = self.pool4(h)
  610. model.fcn16 features_at:37 :9044.9 Mb h = self.relu5_1(self.conv5_1(h))
  611. model.fcn16 features_at:38 :9044.9 Mb h = self.relu5_2(self.conv5_2(h))
  612. model.fcn16 features_at:39 :9066.9 Mb h = self.relu5_3(self.conv5_3(h))
  613. model.fcn16 features_at:40 :9088.9 Mb h = self.pool5(h)
  614. model.fcn16 features_at:42 :9088.9 Mb h = self.relu6(self.fc6(h))
  615. model.fcn16 features_at:43 :9316.9 Mb h = self.drop6(h)
  616. model.fcn16 features_at:45 :9316.9 Mb h = self.relu7(self.fc7(h))
  617. model.fcn16 features_at:46 :9316.9 Mb h = self.drop7(h)
  618. model.fcn16 features_at:48 :9316.9 Mb h = self.score_fr(h)
  619. model.fcn16 features_at:50 :9316.9 Mb return h
  620. distances.utils feature_apply:105 :9316.9 Mb dist_val = apply_func(a_phi, b_phi)
  621. distances.mlp_base forward:40 :9316.9 Mb self._check_inputs(features_a, features_b)
  622. distances.mlp_fcn _check_inputs:32 :9316.9 Mb self._maybe_init(features_a)
  623. distances.mlp_fcn _maybe_init:28 :9316.9 Mb if not self._inited:
  624. distances.mlp_fcn _maybe_init:29 :9316.9 Mb self._init_net(some_feature_input.size()[1:])
  625. distances.mlp_fcn _init_net:37 :9316.9 Mb if self._shapes is not None:
  626. distances.mlp_fcn _init_net:38 :9316.9 Mb used_kern_sizes = (n_features[0],) + tuple(self._shapes['channels'])
  627. distances.mlp_fcn _init_net:39 :9316.9 Mb self.cnn_net = self.classifier = nn.Sequential(
  628. distances.mlp_fcn _init_net:40 :9316.9 Mb *chain.from_iterable(
  629. distances.mlp_fcn _init_net:41 :9316.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  630. distances.mlp_fcn _init_net:45 :9316.9 Mb for i in range(len(used_kern_sizes)-1)
  631. distances.mlp_fcn _init_net:47 :9316.9 Mb Flatten()
  632. distances.mlp_fcn <genexpr>:41 :9316.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  633. distances.mlp_fcn <genexpr>:45 :9316.9 Mb for i in range(len(used_kern_sizes)-1)
  634. distances.mlp_fcn <genexpr>:41 :9316.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  635. distances.mlp_fcn <genexpr>:45 :9316.9 Mb for i in range(len(used_kern_sizes)-1)
  636. distances.mlp_fcn <genexpr>:41 :9316.9 Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  637. distances.mlp_fcn _init_net:50 :9316.9 Mb test_pass_var = Variable(torch.zeros(1, *n_features), volatile=True).cuda()
  638. distances.mlp_fcn _init_net:51 :9316.9 Mb feature_n = self.cnn_net(test_pass_var).size(1) # because [1, F*H*W]
  639. distances.mlp_fcn forward:17 :9316.9 Mb return X.view(X.size(0), -1)
  640. distances.mlp_fcn _init_net:52 :9316.9 Mb self.net = nn.Sequential(
  641. distances.mlp_fcn _init_net:53 :9316.9 Mb self.cnn_net,
  642. distances.mlp_fcn _init_net:54 :9316.9 Mb nn.Linear(feature_n, 1)
  643. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  644. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  645. distances.mlp_fcn _init_net:59 :9316.9 Mb self._initializer(w)
  646. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  647. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  648. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  649. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  650. distances.mlp_fcn _init_net:59 :9316.9 Mb self._initializer(w)
  651. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  652. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  653. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  654. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  655. distances.mlp_fcn _init_net:59 :9316.9 Mb self._initializer(w)
  656. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  657. distances.mlp_fcn _init_net:58 :9316.9 Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  658. distances.mlp_fcn _init_net:57 :9316.9 Mb for w in self.parameters():
  659. distances.mlp_fcn _init_net:61 :9316.9 Mb self._d_optimizer = self._optimizer_builder(self.parameters())
  660. distances.mlp_fcn _init_net:62 :9316.9 Mb self._n_features = n_features
  661. distances.mlp_fcn _init_net:63 :9316.9 Mb self.__inited = True
  662. distances.mlp_fcn _check_inputs:33 :9316.9 Mb assert features_a.size()[1:] == features_b.size()[1:] == self._n_features
  663. distances.mlp_base forward:42 :9316.9 Mb if self._attempt_on_forward is True:
  664. distances.mlp_base forward:46 :9316.9 Mb inv_label_cross_ent = self.objective(features_a, features_b, inverse_labels=inv)
  665. distances.mlp objective:21 :9316.9 Mb full_x = torch.cat([features_a, features_b], 0)
  666. distances.mlp objective:23 :9316.9 Mb if self.stored_y is None:
  667. distances.mlp objective:28 :9316.9 Mb _y = self.stored_y
  668. distances.mlp objective:30 :9316.9 Mb full_y = _y if not inverse_labels else (1 - _y)
  669. distances.mlp objective:31 :9316.9 Mb output = self.network_forward(full_x)
  670. distances.mlp_base network_forward:37 :9316.9 Mb return self.net(x)
  671. distances.mlp_fcn forward:17 :9316.9 Mb return X.view(X.size(0), -1)
  672. distances.mlp objective:32 :9316.9 Mb cross_ent = F.binary_cross_entropy(F.sigmoid(output.view(-1)), full_y.view(-1))
  673. distances.mlp objective:33 :9316.9 Mb assert not has_nan(cross_ent)
  674. distances.utils has_nan:83 :9316.9 Mb if type(x) is float:
  675. distances.utils has_nan:86 :9316.9 Mb x = x.data if type(x) is Variable else x
  676. distances.utils has_nan:87 :9316.9 Mb return ((float(torch.sum(x != x)) > 0)
  677. distances.utils has_nan:88 :9316.9 Mb or float(torch.sum(x == np.inf)) > 0)
  678. distances.mlp objective:34 :9316.9 Mb return cross_ent
  679. distances.mlp_base forward:47 :9316.9 Mb return inv_label_cross_ent
  680. distances.utils feature_apply:106 :9316.9 Mb return dist_val
  681. __main__ match_source_target:149 :9316.9 Mb optimizer.zero_grad()
  682. __main__ match_source_target:150 :9316.9 Mb dist_val.backward()
  683. __main__ match_source_target:151 :10620.9Mb optimizer.step()
  684. __main__ match_source_target:154 :10620.9Mb if (batch_idx + 1) % yield_every == 0:
  685. __main__ match_source_target:122 :10620.9Mb for batch_idx, ((source_data, s_y), (target_data, ___)) in tqdm_adaptation_iterator:
  686. __main__ match_source_target:123 :10620.9Mb if source_data.numel == 0:
  687. __main__ match_source_target:127 :10620.9Mb dist_func.current_pos = (epoch_i, batch_idx, batch_idx)
  688. __main__ match_source_target:128 :10620.9Mb dist_func.y_a = s_y
  689. __main__ match_source_target:130 :10620.9Mb data_s = Variable(maybe_cuda(source_data))
  690. start maybe_cuda:85 :10620.9Mb return x.cuda() if torch.cuda.is_available() else x
  691. __main__ match_source_target:131 :10620.9Mb data_t = Variable(maybe_cuda(target_data))
  692. start maybe_cuda:85 :10620.9Mb return x.cuda() if torch.cuda.is_available() else x
  693. __main__ match_source_target:134 :10620.9Mb if discriminator_needs_update is True:
  694. __main__ match_source_target:136 :10620.9Mb feature_apply(partial(dist_func.attempt_update_d, epoch_i=epoch_i),
  695. __main__ match_source_target:137 :10620.9Mb base_model, target_model, data_s, data_t, shift)
  696. distances.utils feature_apply:103 :10620.9Mb a_phi = model_a.features_at(input_a, shift)
  697. model.fcn16 features_at:16 :10620.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  698. model.fcn16 features_at:18 :10620.9Mb h = x
  699. model.fcn16 features_at:19 :10620.9Mb h = self.relu1_1(self.conv1_1(h))
  700. model.fcn16 features_at:20 :10620.9Mb h = self.relu1_2(self.conv1_2(h))
  701. model.fcn16 features_at:21 :10620.9Mb h = self.pool1(h)
  702. model.fcn16 features_at:23 :10620.9Mb h = self.relu2_1(self.conv2_1(h))
  703. model.fcn16 features_at:24 :10620.9Mb h = self.relu2_2(self.conv2_2(h))
  704. model.fcn16 features_at:25 :10620.9Mb h = self.pool2(h)
  705. model.fcn16 features_at:27 :10620.9Mb h = self.relu3_1(self.conv3_1(h))
  706. model.fcn16 features_at:28 :10620.9Mb h = self.relu3_2(self.conv3_2(h))
  707. model.fcn16 features_at:29 :10620.9Mb h = self.relu3_3(self.conv3_3(h))
  708. model.fcn16 features_at:30 :10620.9Mb h = self.pool3(h)
  709. model.fcn16 features_at:32 :10620.9Mb h = self.relu4_1(self.conv4_1(h))
  710. model.fcn16 features_at:33 :10620.9Mb h = self.relu4_2(self.conv4_2(h))
  711. model.fcn16 features_at:34 :10620.9Mb h = self.relu4_3(self.conv4_3(h))
  712. model.fcn16 features_at:35 :10620.9Mb h = self.pool4(h)
  713. model.fcn16 features_at:37 :10620.9Mb h = self.relu5_1(self.conv5_1(h))
  714. model.fcn16 features_at:38 :10620.9Mb h = self.relu5_2(self.conv5_2(h))
  715. model.fcn16 features_at:39 :10620.9Mb h = self.relu5_3(self.conv5_3(h))
  716. model.fcn16 features_at:40 :10620.9Mb h = self.pool5(h)
  717. model.fcn16 features_at:42 :10620.9Mb h = self.relu6(self.fc6(h))
  718. model.fcn16 features_at:43 :10620.9Mb h = self.drop6(h)
  719. model.fcn16 features_at:45 :10620.9Mb h = self.relu7(self.fc7(h))
  720. model.fcn16 features_at:46 :10620.9Mb h = self.drop7(h)
  721. model.fcn16 features_at:48 :10620.9Mb h = self.score_fr(h)
  722. model.fcn16 features_at:50 :10620.9Mb return h
  723. distances.utils feature_apply:104 :10620.9Mb b_phi = model_b.features_at(input_b, shift)
  724. model.fcn16 features_at:16 :10620.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  725. model.fcn16 features_at:18 :10620.9Mb h = x
  726. model.fcn16 features_at:19 :10620.9Mb h = self.relu1_1(self.conv1_1(h))
  727. model.fcn16 features_at:20 :10620.9Mb h = self.relu1_2(self.conv1_2(h))
  728. model.fcn16 features_at:21 :10620.9Mb h = self.pool1(h)
  729. model.fcn16 features_at:23 :10620.9Mb h = self.relu2_1(self.conv2_1(h))
  730. model.fcn16 features_at:24 :10620.9Mb h = self.relu2_2(self.conv2_2(h))
  731. model.fcn16 features_at:25 :10620.9Mb h = self.pool2(h)
  732. model.fcn16 features_at:27 :10620.9Mb h = self.relu3_1(self.conv3_1(h))
  733. model.fcn16 features_at:28 :10620.9Mb h = self.relu3_2(self.conv3_2(h))
  734. model.fcn16 features_at:29 :10620.9Mb h = self.relu3_3(self.conv3_3(h))
  735. model.fcn16 features_at:30 :10620.9Mb h = self.pool3(h)
  736. model.fcn16 features_at:32 :10620.9Mb h = self.relu4_1(self.conv4_1(h))
  737. model.fcn16 features_at:33 :10620.9Mb h = self.relu4_2(self.conv4_2(h))
  738. model.fcn16 features_at:34 :10620.9Mb h = self.relu4_3(self.conv4_3(h))
  739. model.fcn16 features_at:35 :10620.9Mb h = self.pool4(h)
  740. model.fcn16 features_at:37 :10620.9Mb h = self.relu5_1(self.conv5_1(h))
  741. model.fcn16 features_at:38 :10620.9Mb h = self.relu5_2(self.conv5_2(h))
  742. model.fcn16 features_at:39 :10620.9Mb h = self.relu5_3(self.conv5_3(h))
  743. model.fcn16 features_at:40 :10620.9Mb h = self.pool5(h)
  744. model.fcn16 features_at:42 :10620.9Mb h = self.relu6(self.fc6(h))
  745. model.fcn16 features_at:43 :10620.9Mb h = self.drop6(h)
  746. model.fcn16 features_at:45 :10620.9Mb h = self.relu7(self.fc7(h))
  747. model.fcn16 features_at:46 :10620.9Mb h = self.drop7(h)
  748. model.fcn16 features_at:48 :10620.9Mb h = self.score_fr(h)
  749. model.fcn16 features_at:50 :10620.9Mb return h
  750. distances.utils feature_apply:105 :10620.9Mb dist_val = apply_func(a_phi, b_phi)
  751. distances.mlp_base attempt_update_d:62 :10620.9Mb self._check_inputs(features_a, features_b)
  752. distances.mlp_fcn _check_inputs:32 :10620.9Mb self._maybe_init(features_a)
  753. distances.mlp_fcn _maybe_init:28 :10620.9Mb if not self._inited:
  754. distances.mlp_fcn _maybe_init:29 :10620.9Mb self._init_net(some_feature_input.size()[1:])
  755. distances.mlp_fcn _init_net:37 :10620.9Mb if self._shapes is not None:
  756. distances.mlp_fcn _init_net:38 :10620.9Mb used_kern_sizes = (n_features[0],) + tuple(self._shapes['channels'])
  757. distances.mlp_fcn _init_net:39 :10620.9Mb self.cnn_net = self.classifier = nn.Sequential(
  758. distances.mlp_fcn _init_net:40 :10620.9Mb *chain.from_iterable(
  759. distances.mlp_fcn _init_net:41 :10620.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  760. distances.mlp_fcn _init_net:45 :10620.9Mb for i in range(len(used_kern_sizes)-1)
  761. distances.mlp_fcn _init_net:47 :10620.9Mb Flatten()
  762. distances.mlp_fcn <genexpr>:41 :10620.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  763. distances.mlp_fcn <genexpr>:45 :10620.9Mb for i in range(len(used_kern_sizes)-1)
  764. distances.mlp_fcn <genexpr>:41 :10620.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  765. distances.mlp_fcn <genexpr>:45 :10620.9Mb for i in range(len(used_kern_sizes)-1)
  766. distances.mlp_fcn <genexpr>:41 :10620.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  767. distances.mlp_fcn _init_net:50 :10620.9Mb test_pass_var = Variable(torch.zeros(1, *n_features), volatile=True).cuda()
  768. distances.mlp_fcn _init_net:51 :10620.9Mb feature_n = self.cnn_net(test_pass_var).size(1) # because [1, F*H*W]
  769. distances.mlp_fcn forward:17 :10620.9Mb return X.view(X.size(0), -1)
  770. distances.mlp_fcn _init_net:52 :10620.9Mb self.net = nn.Sequential(
  771. distances.mlp_fcn _init_net:53 :10620.9Mb self.cnn_net,
  772. distances.mlp_fcn _init_net:54 :10620.9Mb nn.Linear(feature_n, 1)
  773. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  774. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  775. distances.mlp_fcn _init_net:59 :10620.9Mb self._initializer(w)
  776. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  777. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  778. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  779. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  780. distances.mlp_fcn _init_net:59 :10620.9Mb self._initializer(w)
  781. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  782. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  783. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  784. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  785. distances.mlp_fcn _init_net:59 :10620.9Mb self._initializer(w)
  786. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  787. distances.mlp_fcn _init_net:58 :10620.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  788. distances.mlp_fcn _init_net:57 :10620.9Mb for w in self.parameters():
  789. distances.mlp_fcn _init_net:61 :10620.9Mb self._d_optimizer = self._optimizer_builder(self.parameters())
  790. distances.mlp_fcn _init_net:62 :10620.9Mb self._n_features = n_features
  791. distances.mlp_fcn _init_net:63 :10620.9Mb self.__inited = True
  792. distances.mlp_fcn _check_inputs:33 :10620.9Mb assert features_a.size()[1:] == features_b.size()[1:] == self._n_features
  793. distances.mlp_base attempt_update_d:64 :10620.9Mb if self._start_from_n_updates > 0 and epoch_i == 0:
  794. distances.mlp_base attempt_update_d:73 :10620.9Mb if (self._step_every_repeat_n[0] > 0
  795. distances.mlp_base attempt_update_d:74 :10620.9Mb and self._update_attempt_counter % self._step_every_repeat_n[0] == 0):
  796. distances.mlp_base attempt_update_d:75 :10620.9Mb for _ in range(self._step_every_repeat_n[1]):
  797. distances.mlp_base attempt_update_d:76 :10620.9Mb self.d_update_step(features_a, features_b)
  798. distances.mlp_base d_update_step:54 :10620.9Mb cross_ent = self.objective(features_a, features_b)
  799. distances.mlp objective:21 :10620.9Mb full_x = torch.cat([features_a, features_b], 0)
  800. distances.mlp objective:23 :10620.9Mb if self.stored_y is None:
  801. distances.mlp objective:28 :10620.9Mb _y = self.stored_y
  802. distances.mlp objective:30 :10620.9Mb full_y = _y if not inverse_labels else (1 - _y)
  803. distances.mlp objective:31 :10620.9Mb output = self.network_forward(full_x)
  804. distances.mlp_base network_forward:37 :10620.9Mb return self.net(x)
  805. distances.mlp_fcn forward:17 :10620.9Mb return X.view(X.size(0), -1)
  806. distances.mlp objective:32 :10620.9Mb cross_ent = F.binary_cross_entropy(F.sigmoid(output.view(-1)), full_y.view(-1))
  807. distances.mlp objective:33 :10620.9Mb assert not has_nan(cross_ent)
  808. distances.utils has_nan:83 :10620.9Mb if type(x) is float:
  809. distances.utils has_nan:86 :10620.9Mb x = x.data if type(x) is Variable else x
  810. distances.utils has_nan:87 :10620.9Mb return ((float(torch.sum(x != x)) > 0)
  811. distances.utils has_nan:88 :10620.9Mb or float(torch.sum(x == np.inf)) > 0)
  812. distances.mlp objective:34 :10620.9Mb return cross_ent
  813. distances.mlp_base d_update_step:56 :10620.9Mb self._d_optimizer.zero_grad()
  814. distances.mlp_base d_update_step:58 :10620.9Mb cross_ent.backward()
  815. distances.mlp_base d_update_step:59 :11404.9Mb self._d_optimizer.step()
  816. distances.mlp_base attempt_update_d:77 :11404.9Mb self._update_attempt_counter += 1
  817. distances.mlp_base attempt_update_d:75 :11404.9Mb for _ in range(self._step_every_repeat_n[1]):
  818. distances.mlp_base attempt_update_d:78 :11404.9Mb return
  819. distances.utils feature_apply:106 :11404.9Mb return dist_val
  820. __main__ match_source_target:139 :11404.9Mb del data_s, data_t
  821. __main__ match_source_target:140 :11404.9Mb data_s = Variable(maybe_cuda(source_data))
  822. start maybe_cuda:85 :11404.9Mb return x.cuda() if torch.cuda.is_available() else x
  823. __main__ match_source_target:141 :11404.9Mb data_t = Variable(maybe_cuda(target_data))
  824. start maybe_cuda:85 :11404.9Mb return x.cuda() if torch.cuda.is_available() else x
  825. __main__ match_source_target:145 :11404.9Mb if not discriminator_only:
  826. __main__ match_source_target:146 :11404.9Mb dist_val = feature_apply(dist_func,
  827. __main__ match_source_target:147 :11404.9Mb base_model, target_model, data_s, data_t, shift)
  828. distances.utils feature_apply:103 :11404.9Mb a_phi = model_a.features_at(input_a, shift)
  829. model.fcn16 features_at:16 :11404.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  830. model.fcn16 features_at:18 :11404.9Mb h = x
  831. model.fcn16 features_at:19 :11404.9Mb h = self.relu1_1(self.conv1_1(h))
  832. model.fcn16 features_at:20 :11404.9Mb h = self.relu1_2(self.conv1_2(h))
  833. model.fcn16 features_at:21 :11404.9Mb h = self.pool1(h)
  834. model.fcn16 features_at:23 :11404.9Mb h = self.relu2_1(self.conv2_1(h))
  835. model.fcn16 features_at:24 :11404.9Mb h = self.relu2_2(self.conv2_2(h))
  836. model.fcn16 features_at:25 :11404.9Mb h = self.pool2(h)
  837. model.fcn16 features_at:27 :11404.9Mb h = self.relu3_1(self.conv3_1(h))
  838. model.fcn16 features_at:28 :11404.9Mb h = self.relu3_2(self.conv3_2(h))
  839. model.fcn16 features_at:29 :11404.9Mb h = self.relu3_3(self.conv3_3(h))
  840. model.fcn16 features_at:30 :11404.9Mb h = self.pool3(h)
  841. model.fcn16 features_at:32 :11404.9Mb h = self.relu4_1(self.conv4_1(h))
  842. model.fcn16 features_at:33 :11404.9Mb h = self.relu4_2(self.conv4_2(h))
  843. model.fcn16 features_at:34 :11404.9Mb h = self.relu4_3(self.conv4_3(h))
  844. model.fcn16 features_at:35 :11404.9Mb h = self.pool4(h)
  845. model.fcn16 features_at:37 :11404.9Mb h = self.relu5_1(self.conv5_1(h))
  846. model.fcn16 features_at:38 :11404.9Mb h = self.relu5_2(self.conv5_2(h))
  847. model.fcn16 features_at:39 :11404.9Mb h = self.relu5_3(self.conv5_3(h))
  848. model.fcn16 features_at:40 :11404.9Mb h = self.pool5(h)
  849. model.fcn16 features_at:42 :11404.9Mb h = self.relu6(self.fc6(h))
  850. model.fcn16 features_at:43 :11404.9Mb h = self.drop6(h)
  851. model.fcn16 features_at:45 :11404.9Mb h = self.relu7(self.fc7(h))
  852. model.fcn16 features_at:46 :11404.9Mb h = self.drop7(h)
  853. model.fcn16 features_at:48 :11404.9Mb h = self.score_fr(h)
  854. model.fcn16 features_at:50 :11404.9Mb return h
  855. distances.utils feature_apply:104 :11404.9Mb b_phi = model_b.features_at(input_b, shift)
  856. model.fcn16 features_at:16 :11404.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  857. model.fcn16 features_at:18 :11404.9Mb h = x
  858. model.fcn16 features_at:19 :11404.9Mb h = self.relu1_1(self.conv1_1(h))
  859. model.fcn16 features_at:20 :11404.9Mb h = self.relu1_2(self.conv1_2(h))
  860. model.fcn16 features_at:21 :11404.9Mb h = self.pool1(h)
  861. model.fcn16 features_at:23 :11404.9Mb h = self.relu2_1(self.conv2_1(h))
  862. model.fcn16 features_at:24 :11404.9Mb h = self.relu2_2(self.conv2_2(h))
  863. model.fcn16 features_at:25 :11404.9Mb h = self.pool2(h)
  864. model.fcn16 features_at:27 :11404.9Mb h = self.relu3_1(self.conv3_1(h))
  865. model.fcn16 features_at:28 :11404.9Mb h = self.relu3_2(self.conv3_2(h))
  866. model.fcn16 features_at:29 :11404.9Mb h = self.relu3_3(self.conv3_3(h))
  867. model.fcn16 features_at:30 :11404.9Mb h = self.pool3(h)
  868. model.fcn16 features_at:32 :11404.9Mb h = self.relu4_1(self.conv4_1(h))
  869. model.fcn16 features_at:33 :11404.9Mb h = self.relu4_2(self.conv4_2(h))
  870. model.fcn16 features_at:34 :11404.9Mb h = self.relu4_3(self.conv4_3(h))
  871. model.fcn16 features_at:35 :11404.9Mb h = self.pool4(h)
  872. model.fcn16 features_at:37 :11404.9Mb h = self.relu5_1(self.conv5_1(h))
  873. model.fcn16 features_at:38 :11404.9Mb h = self.relu5_2(self.conv5_2(h))
  874. model.fcn16 features_at:39 :11404.9Mb h = self.relu5_3(self.conv5_3(h))
  875. model.fcn16 features_at:40 :11404.9Mb h = self.pool5(h)
  876. model.fcn16 features_at:42 :11404.9Mb h = self.relu6(self.fc6(h))
  877. model.fcn16 features_at:43 :11404.9Mb h = self.drop6(h)
  878. model.fcn16 features_at:45 :11404.9Mb h = self.relu7(self.fc7(h))
  879. model.fcn16 features_at:46 :11404.9Mb h = self.drop7(h)
  880. model.fcn16 features_at:48 :11404.9Mb h = self.score_fr(h)
  881. model.fcn16 features_at:50 :11404.9Mb return h
  882. distances.utils feature_apply:105 :11404.9Mb dist_val = apply_func(a_phi, b_phi)
  883. distances.mlp_base forward:40 :11404.9Mb self._check_inputs(features_a, features_b)
  884. distances.mlp_fcn _check_inputs:32 :11404.9Mb self._maybe_init(features_a)
  885. distances.mlp_fcn _maybe_init:28 :11404.9Mb if not self._inited:
  886. distances.mlp_fcn _maybe_init:29 :11404.9Mb self._init_net(some_feature_input.size()[1:])
  887. distances.mlp_fcn _init_net:37 :11404.9Mb if self._shapes is not None:
  888. distances.mlp_fcn _init_net:38 :11404.9Mb used_kern_sizes = (n_features[0],) + tuple(self._shapes['channels'])
  889. distances.mlp_fcn _init_net:39 :11404.9Mb self.cnn_net = self.classifier = nn.Sequential(
  890. distances.mlp_fcn _init_net:40 :11404.9Mb *chain.from_iterable(
  891. distances.mlp_fcn _init_net:41 :11404.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  892. distances.mlp_fcn _init_net:45 :11404.9Mb for i in range(len(used_kern_sizes)-1)
  893. distances.mlp_fcn _init_net:47 :11404.9Mb Flatten()
  894. distances.mlp_fcn <genexpr>:41 :11404.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  895. distances.mlp_fcn <genexpr>:45 :11404.9Mb for i in range(len(used_kern_sizes)-1)
  896. distances.mlp_fcn <genexpr>:41 :11404.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  897. distances.mlp_fcn <genexpr>:45 :11404.9Mb for i in range(len(used_kern_sizes)-1)
  898. distances.mlp_fcn <genexpr>:41 :11404.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  899. distances.mlp_fcn _init_net:50 :11404.9Mb test_pass_var = Variable(torch.zeros(1, *n_features), volatile=True).cuda()
  900. distances.mlp_fcn _init_net:51 :11404.9Mb feature_n = self.cnn_net(test_pass_var).size(1) # because [1, F*H*W]
  901. distances.mlp_fcn forward:17 :11404.9Mb return X.view(X.size(0), -1)
  902. distances.mlp_fcn _init_net:52 :11404.9Mb self.net = nn.Sequential(
  903. distances.mlp_fcn _init_net:53 :11404.9Mb self.cnn_net,
  904. distances.mlp_fcn _init_net:54 :11404.9Mb nn.Linear(feature_n, 1)
  905. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  906. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  907. distances.mlp_fcn _init_net:59 :11404.9Mb self._initializer(w)
  908. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  909. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  910. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  911. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  912. distances.mlp_fcn _init_net:59 :11404.9Mb self._initializer(w)
  913. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  914. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  915. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  916. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  917. distances.mlp_fcn _init_net:59 :11404.9Mb self._initializer(w)
  918. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  919. distances.mlp_fcn _init_net:58 :11404.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  920. distances.mlp_fcn _init_net:57 :11404.9Mb for w in self.parameters():
  921. distances.mlp_fcn _init_net:61 :11404.9Mb self._d_optimizer = self._optimizer_builder(self.parameters())
  922. distances.mlp_fcn _init_net:62 :11404.9Mb self._n_features = n_features
  923. distances.mlp_fcn _init_net:63 :11404.9Mb self.__inited = True
  924. distances.mlp_fcn _check_inputs:33 :11404.9Mb assert features_a.size()[1:] == features_b.size()[1:] == self._n_features
  925. distances.mlp_base forward:42 :11404.9Mb if self._attempt_on_forward is True:
  926. distances.mlp_base forward:46 :11404.9Mb inv_label_cross_ent = self.objective(features_a, features_b, inverse_labels=inv)
  927. distances.mlp objective:21 :11404.9Mb full_x = torch.cat([features_a, features_b], 0)
  928. distances.mlp objective:23 :11404.9Mb if self.stored_y is None:
  929. distances.mlp objective:28 :11404.9Mb _y = self.stored_y
  930. distances.mlp objective:30 :11404.9Mb full_y = _y if not inverse_labels else (1 - _y)
  931. distances.mlp objective:31 :11404.9Mb output = self.network_forward(full_x)
  932. distances.mlp_base network_forward:37 :11404.9Mb return self.net(x)
  933. distances.mlp_fcn forward:17 :11404.9Mb return X.view(X.size(0), -1)
  934. distances.mlp objective:32 :11404.9Mb cross_ent = F.binary_cross_entropy(F.sigmoid(output.view(-1)), full_y.view(-1))
  935. distances.mlp objective:33 :11404.9Mb assert not has_nan(cross_ent)
  936. distances.utils has_nan:83 :11404.9Mb if type(x) is float:
  937. distances.utils has_nan:86 :11404.9Mb x = x.data if type(x) is Variable else x
  938. distances.utils has_nan:87 :11404.9Mb return ((float(torch.sum(x != x)) > 0)
  939. distances.utils has_nan:88 :11404.9Mb or float(torch.sum(x == np.inf)) > 0)
  940. distances.mlp objective:34 :11404.9Mb return cross_ent
  941. distances.mlp_base forward:47 :11404.9Mb return inv_label_cross_ent
  942. distances.utils feature_apply:106 :11404.9Mb return dist_val
  943. __main__ match_source_target:149 :11404.9Mb optimizer.zero_grad()
  944. __main__ match_source_target:150 :11404.9Mb dist_val.backward()
  945. __main__ match_source_target:151 :11796.9Mb optimizer.step()
  946. __main__ match_source_target:154 :11796.9Mb if (batch_idx + 1) % yield_every == 0:
  947. __main__ match_source_target:122 :11796.9Mb for batch_idx, ((source_data, s_y), (target_data, ___)) in tqdm_adaptation_iterator:
  948. __main__ match_source_target:123 :11796.9Mb if source_data.numel == 0:
  949. __main__ match_source_target:127 :11796.9Mb dist_func.current_pos = (epoch_i, batch_idx, batch_idx)
  950. __main__ match_source_target:128 :11796.9Mb dist_func.y_a = s_y
  951. __main__ match_source_target:130 :11796.9Mb data_s = Variable(maybe_cuda(source_data))
  952. start maybe_cuda:85 :11796.9Mb return x.cuda() if torch.cuda.is_available() else x
  953. __main__ match_source_target:131 :11796.9Mb data_t = Variable(maybe_cuda(target_data))
  954. start maybe_cuda:85 :11796.9Mb return x.cuda() if torch.cuda.is_available() else x
  955. __main__ match_source_target:134 :11796.9Mb if discriminator_needs_update is True:
  956. __main__ match_source_target:136 :11796.9Mb feature_apply(partial(dist_func.attempt_update_d, epoch_i=epoch_i),
  957. __main__ match_source_target:137 :11796.9Mb base_model, target_model, data_s, data_t, shift)
  958. distances.utils feature_apply:103 :11796.9Mb a_phi = model_a.features_at(input_a, shift)
  959. model.fcn16 features_at:16 :11796.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  960. model.fcn16 features_at:18 :11796.9Mb h = x
  961. model.fcn16 features_at:19 :11796.9Mb h = self.relu1_1(self.conv1_1(h))
  962. model.fcn16 features_at:20 :11796.9Mb h = self.relu1_2(self.conv1_2(h))
  963. model.fcn16 features_at:21 :11796.9Mb h = self.pool1(h)
  964. model.fcn16 features_at:23 :11796.9Mb h = self.relu2_1(self.conv2_1(h))
  965. model.fcn16 features_at:24 :11796.9Mb h = self.relu2_2(self.conv2_2(h))
  966. model.fcn16 features_at:25 :11796.9Mb h = self.pool2(h)
  967. model.fcn16 features_at:27 :11796.9Mb h = self.relu3_1(self.conv3_1(h))
  968. model.fcn16 features_at:28 :11796.9Mb h = self.relu3_2(self.conv3_2(h))
  969. model.fcn16 features_at:29 :11796.9Mb h = self.relu3_3(self.conv3_3(h))
  970. model.fcn16 features_at:30 :11796.9Mb h = self.pool3(h)
  971. model.fcn16 features_at:32 :11796.9Mb h = self.relu4_1(self.conv4_1(h))
  972. model.fcn16 features_at:33 :11796.9Mb h = self.relu4_2(self.conv4_2(h))
  973. model.fcn16 features_at:34 :11796.9Mb h = self.relu4_3(self.conv4_3(h))
  974. model.fcn16 features_at:35 :11796.9Mb h = self.pool4(h)
  975. model.fcn16 features_at:37 :11796.9Mb h = self.relu5_1(self.conv5_1(h))
  976. model.fcn16 features_at:38 :11796.9Mb h = self.relu5_2(self.conv5_2(h))
  977. model.fcn16 features_at:39 :11796.9Mb h = self.relu5_3(self.conv5_3(h))
  978. model.fcn16 features_at:40 :11796.9Mb h = self.pool5(h)
  979. model.fcn16 features_at:42 :11796.9Mb h = self.relu6(self.fc6(h))
  980. model.fcn16 features_at:43 :11796.9Mb h = self.drop6(h)
  981. model.fcn16 features_at:45 :11796.9Mb h = self.relu7(self.fc7(h))
  982. model.fcn16 features_at:46 :11796.9Mb h = self.drop7(h)
  983. model.fcn16 features_at:48 :11796.9Mb h = self.score_fr(h)
  984. model.fcn16 features_at:50 :11796.9Mb return h
  985. distances.utils feature_apply:104 :11796.9Mb b_phi = model_b.features_at(input_b, shift)
  986. model.fcn16 features_at:16 :11796.9Mb assert shift == -3, "!= -3 not implemented, not needed"
  987. model.fcn16 features_at:18 :11796.9Mb h = x
  988. model.fcn16 features_at:19 :11796.9Mb h = self.relu1_1(self.conv1_1(h))
  989. model.fcn16 features_at:20 :11796.9Mb h = self.relu1_2(self.conv1_2(h))
  990. model.fcn16 features_at:21 :11796.9Mb h = self.pool1(h)
  991. model.fcn16 features_at:23 :11796.9Mb h = self.relu2_1(self.conv2_1(h))
  992. model.fcn16 features_at:24 :11796.9Mb h = self.relu2_2(self.conv2_2(h))
  993. model.fcn16 features_at:25 :11796.9Mb h = self.pool2(h)
  994. model.fcn16 features_at:27 :11796.9Mb h = self.relu3_1(self.conv3_1(h))
  995. model.fcn16 features_at:28 :11796.9Mb h = self.relu3_2(self.conv3_2(h))
  996. model.fcn16 features_at:29 :11796.9Mb h = self.relu3_3(self.conv3_3(h))
  997. model.fcn16 features_at:30 :11796.9Mb h = self.pool3(h)
  998. model.fcn16 features_at:32 :11796.9Mb h = self.relu4_1(self.conv4_1(h))
  999. model.fcn16 features_at:33 :11796.9Mb h = self.relu4_2(self.conv4_2(h))
  1000. model.fcn16 features_at:34 :11796.9Mb h = self.relu4_3(self.conv4_3(h))
  1001. model.fcn16 features_at:35 :11796.9Mb h = self.pool4(h)
  1002. model.fcn16 features_at:37 :11796.9Mb h = self.relu5_1(self.conv5_1(h))
  1003. model.fcn16 features_at:38 :11796.9Mb h = self.relu5_2(self.conv5_2(h))
  1004. model.fcn16 features_at:39 :11796.9Mb h = self.relu5_3(self.conv5_3(h))
  1005. model.fcn16 features_at:40 :11796.9Mb h = self.pool5(h)
  1006. model.fcn16 features_at:42 :11796.9Mb h = self.relu6(self.fc6(h))
  1007. model.fcn16 features_at:43 :11796.9Mb h = self.drop6(h)
  1008. model.fcn16 features_at:45 :11796.9Mb h = self.relu7(self.fc7(h))
  1009. model.fcn16 features_at:46 :11796.9Mb h = self.drop7(h)
  1010. model.fcn16 features_at:48 :11796.9Mb h = self.score_fr(h)
  1011. model.fcn16 features_at:50 :11796.9Mb return h
  1012. distances.utils feature_apply:105 :11796.9Mb dist_val = apply_func(a_phi, b_phi)
  1013. distances.mlp_base attempt_update_d:62 :11796.9Mb self._check_inputs(features_a, features_b)
  1014. distances.mlp_fcn _check_inputs:32 :11796.9Mb self._maybe_init(features_a)
  1015. distances.mlp_fcn _maybe_init:28 :11796.9Mb if not self._inited:
  1016. distances.mlp_fcn _maybe_init:29 :11796.9Mb self._init_net(some_feature_input.size()[1:])
  1017. distances.mlp_fcn _init_net:37 :11796.9Mb if self._shapes is not None:
  1018. distances.mlp_fcn _init_net:38 :11796.9Mb used_kern_sizes = (n_features[0],) + tuple(self._shapes['channels'])
  1019. distances.mlp_fcn _init_net:39 :11796.9Mb self.cnn_net = self.classifier = nn.Sequential(
  1020. distances.mlp_fcn _init_net:40 :11796.9Mb *chain.from_iterable(
  1021. distances.mlp_fcn _init_net:41 :11796.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  1022. distances.mlp_fcn _init_net:45 :11796.9Mb for i in range(len(used_kern_sizes)-1)
  1023. distances.mlp_fcn _init_net:47 :11796.9Mb Flatten()
  1024. distances.mlp_fcn <genexpr>:41 :11796.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  1025. distances.mlp_fcn <genexpr>:45 :11796.9Mb for i in range(len(used_kern_sizes)-1)
  1026. distances.mlp_fcn <genexpr>:41 :11796.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  1027. distances.mlp_fcn <genexpr>:45 :11796.9Mb for i in range(len(used_kern_sizes)-1)
  1028. distances.mlp_fcn <genexpr>:41 :11796.9Mb (nn.Conv2d(used_kern_sizes[i], used_kern_sizes[i+1],
  1029. distances.mlp_fcn _init_net:50 :11796.9Mb test_pass_var = Variable(torch.zeros(1, *n_features), volatile=True).cuda()
  1030. distances.mlp_fcn _init_net:51 :11796.9Mb feature_n = self.cnn_net(test_pass_var).size(1) # because [1, F*H*W]
  1031. distances.mlp_fcn forward:17 :11796.9Mb return X.view(X.size(0), -1)
  1032. distances.mlp_fcn _init_net:52 :11796.9Mb self.net = nn.Sequential(
  1033. distances.mlp_fcn _init_net:53 :11796.9Mb self.cnn_net,
  1034. distances.mlp_fcn _init_net:54 :11796.9Mb nn.Linear(feature_n, 1)
  1035. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1036. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1037. distances.mlp_fcn _init_net:59 :11796.9Mb self._initializer(w)
  1038. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1039. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1040. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1041. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1042. distances.mlp_fcn _init_net:59 :11796.9Mb self._initializer(w)
  1043. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1044. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1045. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1046. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1047. distances.mlp_fcn _init_net:59 :11796.9Mb self._initializer(w)
  1048. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1049. distances.mlp_fcn _init_net:58 :11796.9Mb if w.ndimension() >= 2 and not getattr(w, 'initialized', False):
  1050. distances.mlp_fcn _init_net:57 :11796.9Mb for w in self.parameters():
  1051. distances.mlp_fcn _init_net:61 :11796.9Mb self._d_optimizer = self._optimizer_builder(self.parameters())
  1052. distances.mlp_fcn _init_net:62 :11796.9Mb self._n_features = n_features
  1053. distances.mlp_fcn _init_net:63 :11796.9Mb self.__inited = True
  1054. distances.mlp_fcn _check_inputs:33 :11796.9Mb assert features_a.size()[1:] == features_b.size()[1:] == self._n_features
  1055. distances.mlp_base attempt_update_d:64 :11796.9Mb if self._start_from_n_updates > 0 and epoch_i == 0:
  1056. distances.mlp_base attempt_update_d:73 :11796.9Mb if (self._step_every_repeat_n[0] > 0
  1057. distances.mlp_base attempt_update_d:74 :11796.9Mb and self._update_attempt_counter % self._step_every_repeat_n[0] == 0):
  1058. distances.mlp_base attempt_update_d:75 :11796.9Mb for _ in range(self._step_every_repeat_n[1]):
  1059. distances.mlp_base attempt_update_d:76 :11796.9Mb self.d_update_step(features_a, features_b)
  1060. distances.mlp_base d_update_step:54 :11796.9Mb cross_ent = self.objective(features_a, features_b)
  1061. distances.mlp objective:21 :11796.9Mb full_x = torch.cat([features_a, features_b], 0)
  1062. distances.mlp objective:23 :11796.9Mb if self.stored_y is None:
  1063. distances.mlp objective:28 :11796.9Mb _y = self.stored_y
  1064. distances.mlp objective:30 :11796.9Mb full_y = _y if not inverse_labels else (1 - _y)
  1065. distances.mlp objective:31 :11796.9Mb output = self.network_forward(full_x)
  1066. distances.mlp_base network_forward:37 :11796.9Mb return self.net(x)
  1067. distances.mlp_fcn forward:17 :11796.9Mb return X.view(X.size(0), -1)
  1068. distances.mlp objective:32 :11796.9Mb cross_ent = F.binary_cross_entropy(F.sigmoid(output.view(-1)), full_y.view(-1))
  1069. distances.mlp objective:33 :11796.9Mb assert not has_nan(cross_ent)
  1070. distances.utils has_nan:83 :11796.9Mb if type(x) is float:
  1071. distances.utils has_nan:86 :11796.9Mb x = x.data if type(x) is Variable else x
  1072. distances.utils has_nan:87 :11796.9Mb return ((float(torch.sum(x != x)) > 0)
  1073. distances.utils has_nan:88 :11796.9Mb or float(torch.sum(x == np.inf)) > 0)
  1074. distances.mlp objective:34 :11796.9Mb return cross_ent
  1075. distances.mlp_base d_update_step:56 :11796.9Mb self._d_optimizer.zero_grad()
  1076. distances.mlp_base d_update_step:58 :11796.9Mb cross_ent.backward()
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!