API tools faq
paste
Advertisement
Guest User

Untitled

a guest
May 25th, 2016
206
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 123.76 KB | None | 0 0
raw download clone embed print report
  1. I0114 17:30:17.014989 86948 caffe.cpp:184] Using GPUs 0, 1
  2. I0114 17:30:18.139459 86948 solver.cpp:48] Initializing solver from parameters:
  3. test_iter: 10
  4. test_interval: 500
  5. base_lr: 0.001
  6. display: 500
  7. max_iter: 850000
  8. lr_policy: "fixed"
  9. gamma: 0.5
  10. momentum: 0.9
  11. weight_decay: 0.0005
  12. snapshot: 5000
  13. snapshot_prefix: "models/mv16f/mv16f1_"
  14. solver_mode: GPU
  15. device_id: 0
  16. net: "models/mv16f/mv_train1.prototxt"
  17. I0114 17:30:18.139631 86948 solver.cpp:91] Creating training net from net file: models/mv16f/mv_train1.prototxt
  18. I0114 17:30:18.140243 86948 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
  19. I0114 17:30:18.140266 86948 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
  20. I0114 17:30:18.140447 86948 net.cpp:49] Initializing net from parameters:
  21. name: "mv_16f1"
  22. state {
  23. phase: TRAIN
  24. }
  25. layer {
  26. name: "data"
  27. type: "HDF5Data"
  28. top: "data"
  29. top: "label"
  30. include {
  31. phase: TRAIN
  32. }
  33. hdf5_data_param {
  34. source: "/home/fe/anilil/caffe/models/mv16f/train.txt"
  35. batch_size: 150
  36. }
  37. }
  38. layer {
  39. name: "conv1"
  40. type: "Convolution"
  41. bottom: "data"
  42. top: "conv1"
  43. param {
  44. lr_mult: 1
  45. decay_mult: 1
  46. }
  47. param {
  48. lr_mult: 1
  49. decay_mult: 1
  50. }
  51. convolution_param {
  52. num_output: 64
  53. kernel_size: 3
  54. weight_filler {
  55. type: "xavier"
  56. }
  57. bias_filler {
  58. type: "xavier"
  59. }
  60. }
  61. }
  62. layer {
  63. name: "relu1"
  64. type: "ReLU"
  65. bottom: "conv1"
  66. top: "conv1"
  67. }
  68. layer {
  69. name: "pool1"
  70. type: "Pooling"
  71. bottom: "conv1"
  72. top: "pool1"
  73. pooling_param {
  74. pool: AVE
  75. kernel_size: 2
  76. stride: 1
  77. }
  78. }
  79. layer {
  80. name: "conv2"
  81. type: "Convolution"
  82. bottom: "pool1"
  83. top: "conv2"
  84. param {
  85. lr_mult: 1
  86. decay_mult: 1
  87. }
  88. param {
  89. lr_mult: 1
  90. decay_mult: 1
  91. }
  92. convolution_param {
  93. num_output: 128
  94. kernel_size: 3
  95. stride: 1
  96. weight_filler {
  97. type: "xavier"
  98. }
  99. bias_filler {
  100. type: "xavier"
  101. }
  102. }
  103. }
  104. layer {
  105. name: "relu2"
  106. type: "ReLU"
  107. bottom: "conv2"
  108. top: "conv2"
  109. }
  110. layer {
  111. name: "pool2"
  112. type: "Pooling"
  113. bottom: "conv2"
  114. top: "pool2"
  115. pooling_param {
  116. pool: AVE
  117. kernel_size: 2
  118. stride: 2
  119. }
  120. }
  121. layer {
  122. name: "conv3"
  123. type: "Convolution"
  124. bottom: "pool2"
  125. top: "conv3"
  126. param {
  127. lr_mult: 1
  128. decay_mult: 1
  129. }
  130. param {
  131. lr_mult: 1
  132. decay_mult: 1
  133. }
  134. convolution_param {
  135. num_output: 256
  136. kernel_size: 3
  137. stride: 1
  138. weight_filler {
  139. type: "xavier"
  140. }
  141. bias_filler {
  142. type: "xavier"
  143. }
  144. }
  145. }
  146. layer {
  147. name: "relu3"
  148. type: "ReLU"
  149. bottom: "conv3"
  150. top: "conv3"
  151. }
  152. layer {
  153. name: "pool3"
  154. type: "Pooling"
  155. bottom: "conv3"
  156. top: "pool3"
  157. pooling_param {
  158. pool: AVE
  159. kernel_size: 2
  160. stride: 2
  161. }
  162. }
  163. layer {
  164. name: "conv4"
  165. type: "Convolution"
  166. bottom: "pool3"
  167. top: "conv4"
  168. param {
  169. lr_mult: 1
  170. decay_mult: 1
  171. }
  172. param {
  173. lr_mult: 1
  174. decay_mult: 1
  175. }
  176. convolution_param {
  177. num_output: 256
  178. kernel_size: 3
  179. stride: 1
  180. weight_filler {
  181. type: "xavier"
  182. }
  183. bias_filler {
  184. type: "xavier"
  185. }
  186. }
  187. }
  188. layer {
  189. name: "relu4"
  190. type: "ReLU"
  191. bottom: "conv4"
  192. top: "conv4"
  193. }
  194. layer {
  195. name: "pool4"
  196. type: "Pooling"
  197. bottom: "conv4"
  198. top: "pool4"
  199. pooling_param {
  200. pool: AVE
  201. kernel_size: 2
  202. stride: 2
  203. }
  204. }
  205. layer {
  206. name: "conv5"
  207. type: "Convolution"
  208. bottom: "pool4"
  209. top: "conv5"
  210. param {
  211. lr_mult: 1
  212. decay_mult: 1
  213. }
  214. param {
  215. lr_mult: 1
  216. decay_mult: 1
  217. }
  218. convolution_param {
  219. num_output: 256
  220. kernel_size: 3
  221. stride: 1
  222. weight_filler {
  223. type: "xavier"
  224. }
  225. bias_filler {
  226. type: "xavier"
  227. }
  228. }
  229. }
  230. layer {
  231. name: "relu5"
  232. type: "ReLU"
  233. bottom: "conv5"
  234. top: "conv5"
  235. }
  236. layer {
  237. name: "pool5"
  238. type: "Pooling"
  239. bottom: "conv5"
  240. top: "pool5"
  241. pooling_param {
  242. pool: AVE
  243. kernel_size: 2
  244. stride: 2
  245. }
  246. }
  247. layer {
  248. name: "fc6"
  249. type: "InnerProduct"
  250. bottom: "pool5"
  251. top: "fc6"
  252. param {
  253. lr_mult: 1
  254. decay_mult: 1
  255. }
  256. param {
  257. lr_mult: 2
  258. decay_mult: 0
  259. }
  260. inner_product_param {
  261. num_output: 2048
  262. weight_filler {
  263. type: "xavier"
  264. }
  265. bias_filler {
  266. type: "xavier"
  267. }
  268. }
  269. }
  270. layer {
  271. name: "relu6"
  272. type: "ReLU"
  273. bottom: "fc6"
  274. top: "fc6"
  275. }
  276. layer {
  277. name: "fc7"
  278. type: "InnerProduct"
  279. bottom: "fc6"
  280. top: "fc7"
  281. param {
  282. lr_mult: 1
  283. decay_mult: 1
  284. }
  285. param {
  286. lr_mult: 2
  287. decay_mult: 0
  288. }
  289. inner_product_param {
  290. num_output: 2048
  291. weight_filler {
  292. type: "xavier"
  293. }
  294. bias_filler {
  295. type: "xavier"
  296. }
  297. }
  298. }
  299. layer {
  300. name: "relu7"
  301. type: "ReLU"
  302. bottom: "fc7"
  303. top: "fc7"
  304. }
  305. layer {
  306. name: "fc8"
  307. type: "InnerProduct"
  308. bottom: "fc7"
  309. top: "fc8"
  310. param {
  311. lr_mult: 1
  312. decay_mult: 1
  313. }
  314. param {
  315. lr_mult: 2
  316. decay_mult: 0
  317. }
  318. inner_product_param {
  319. num_output: 101
  320. weight_filler {
  321. type: "xavier"
  322. }
  323. bias_filler {
  324. type: "xavier"
  325. }
  326. }
  327. }
  328. layer {
  329. name: "loss"
  330. type: "SoftmaxWithLoss"
  331. bottom: "fc8"
  332. bottom: "label"
  333. top: "loss"
  334. }
  335. I0114 17:30:18.140599 86948 layer_factory.hpp:77] Creating layer data
  336. I0114 17:30:18.140625 86948 net.cpp:106] Creating Layer data
  337. I0114 17:30:18.140630 86948 net.cpp:411] data -> data
  338. I0114 17:30:18.140650 86948 net.cpp:411] data -> label
  339. I0114 17:30:18.140666 86948 hdf5_data_layer.cpp:79] Loading list of HDF5 filenames from: /home/fe/anilil/caffe/models/mv16f/train.txt
  340. I0114 17:30:18.140861 86948 hdf5_data_layer.cpp:93] Number of HDF5 files: 477
  341. I0114 17:30:18.141713 86948 hdf5.cpp:32] Datatype class: H5T_FLOAT
  342. I0114 17:30:19.121565 86948 net.cpp:150] Setting up data
  343. I0114 17:30:19.121650 86948 net.cpp:157] Top shape: 150 48 58 58 (24220800)
  344. I0114 17:30:19.121657 86948 net.cpp:157] Top shape: 150 1 (150)
  345. I0114 17:30:19.121660 86948 net.cpp:165] Memory required for data: 96883800
  346. I0114 17:30:19.121668 86948 layer_factory.hpp:77] Creating layer conv1
  347. I0114 17:30:19.121704 86948 net.cpp:106] Creating Layer conv1
  348. I0114 17:30:19.121721 86948 net.cpp:454] conv1 <- data
  349. I0114 17:30:19.121743 86948 net.cpp:411] conv1 -> conv1
  350. I0114 17:30:19.253258 86948 net.cpp:150] Setting up conv1
  351. I0114 17:30:19.253304 86948 net.cpp:157] Top shape: 150 64 56 56 (30105600)
  352. I0114 17:30:19.253307 86948 net.cpp:165] Memory required for data: 217306200
  353. I0114 17:30:19.253322 86948 layer_factory.hpp:77] Creating layer relu1
  354. I0114 17:30:19.253340 86948 net.cpp:106] Creating Layer relu1
  355. I0114 17:30:19.253345 86948 net.cpp:454] relu1 <- conv1
  356. I0114 17:30:19.253350 86948 net.cpp:397] relu1 -> conv1 (in-place)
  357. I0114 17:30:19.253569 86948 net.cpp:150] Setting up relu1
  358. I0114 17:30:19.253577 86948 net.cpp:157] Top shape: 150 64 56 56 (30105600)
  359. I0114 17:30:19.253592 86948 net.cpp:165] Memory required for data: 337728600
  360. I0114 17:30:19.253595 86948 layer_factory.hpp:77] Creating layer pool1
  361. I0114 17:30:19.253607 86948 net.cpp:106] Creating Layer pool1
  362. I0114 17:30:19.253612 86948 net.cpp:454] pool1 <- conv1
  363. I0114 17:30:19.253615 86948 net.cpp:411] pool1 -> pool1
  364. I0114 17:30:19.253981 86948 net.cpp:150] Setting up pool1
  365. I0114 17:30:19.253993 86948 net.cpp:157] Top shape: 150 64 55 55 (29040000)
  366. I0114 17:30:19.253995 86948 net.cpp:165] Memory required for data: 453888600
  367. I0114 17:30:19.253998 86948 layer_factory.hpp:77] Creating layer conv2
  368. I0114 17:30:19.254022 86948 net.cpp:106] Creating Layer conv2
  369. I0114 17:30:19.254025 86948 net.cpp:454] conv2 <- pool1
  370. I0114 17:30:19.254031 86948 net.cpp:411] conv2 -> conv2
  371. I0114 17:30:19.256091 86948 net.cpp:150] Setting up conv2
  372. I0114 17:30:19.256105 86948 net.cpp:157] Top shape: 150 128 53 53 (53932800)
  373. I0114 17:30:19.256108 86948 net.cpp:165] Memory required for data: 669619800
  374. I0114 17:30:19.256117 86948 layer_factory.hpp:77] Creating layer relu2
  375. I0114 17:30:19.256122 86948 net.cpp:106] Creating Layer relu2
  376. I0114 17:30:19.256125 86948 net.cpp:454] relu2 <- conv2
  377. I0114 17:30:19.256131 86948 net.cpp:397] relu2 -> conv2 (in-place)
  378. I0114 17:30:19.256299 86948 net.cpp:150] Setting up relu2
  379. I0114 17:30:19.256310 86948 net.cpp:157] Top shape: 150 128 53 53 (53932800)
  380. I0114 17:30:19.256325 86948 net.cpp:165] Memory required for data: 885351000
  381. I0114 17:30:19.256327 86948 layer_factory.hpp:77] Creating layer pool2
  382. I0114 17:30:19.256363 86948 net.cpp:106] Creating Layer pool2
  383. I0114 17:30:19.256367 86948 net.cpp:454] pool2 <- conv2
  384. I0114 17:30:19.256372 86948 net.cpp:411] pool2 -> pool2
  385. I0114 17:30:19.256733 86948 net.cpp:150] Setting up pool2
  386. I0114 17:30:19.256744 86948 net.cpp:157] Top shape: 150 128 27 27 (13996800)
  387. I0114 17:30:19.256747 86948 net.cpp:165] Memory required for data: 941338200
  388. I0114 17:30:19.256750 86948 layer_factory.hpp:77] Creating layer conv3
  389. I0114 17:30:19.256760 86948 net.cpp:106] Creating Layer conv3
  390. I0114 17:30:19.256763 86948 net.cpp:454] conv3 <- pool2
  391. I0114 17:30:19.256770 86948 net.cpp:411] conv3 -> conv3
  392. I0114 17:30:19.260486 86948 net.cpp:150] Setting up conv3
  393. I0114 17:30:19.260499 86948 net.cpp:157] Top shape: 150 256 25 25 (24000000)
  394. I0114 17:30:19.260501 86948 net.cpp:165] Memory required for data: 1037338200
  395. I0114 17:30:19.260509 86948 layer_factory.hpp:77] Creating layer relu3
  396. I0114 17:30:19.260517 86948 net.cpp:106] Creating Layer relu3
  397. I0114 17:30:19.260520 86948 net.cpp:454] relu3 <- conv3
  398. I0114 17:30:19.260524 86948 net.cpp:397] relu3 -> conv3 (in-place)
  399. I0114 17:30:19.260691 86948 net.cpp:150] Setting up relu3
  400. I0114 17:30:19.260699 86948 net.cpp:157] Top shape: 150 256 25 25 (24000000)
  401. I0114 17:30:19.260713 86948 net.cpp:165] Memory required for data: 1133338200
  402. I0114 17:30:19.260716 86948 layer_factory.hpp:77] Creating layer pool3
  403. I0114 17:30:19.260725 86948 net.cpp:106] Creating Layer pool3
  404. I0114 17:30:19.260728 86948 net.cpp:454] pool3 <- conv3
  405. I0114 17:30:19.260733 86948 net.cpp:411] pool3 -> pool3
  406. I0114 17:30:19.261076 86948 net.cpp:150] Setting up pool3
  407. I0114 17:30:19.261086 86948 net.cpp:157] Top shape: 150 256 13 13 (6489600)
  408. I0114 17:30:19.261090 86948 net.cpp:165] Memory required for data: 1159296600
  409. I0114 17:30:19.261093 86948 layer_factory.hpp:77] Creating layer conv4
  410. I0114 17:30:19.261103 86948 net.cpp:106] Creating Layer conv4
  411. I0114 17:30:19.261106 86948 net.cpp:454] conv4 <- pool3
  412. I0114 17:30:19.261113 86948 net.cpp:411] conv4 -> conv4
  413. I0114 17:30:19.266593 86948 net.cpp:150] Setting up conv4
  414. I0114 17:30:19.266607 86948 net.cpp:157] Top shape: 150 256 11 11 (4646400)
  415. I0114 17:30:19.266610 86948 net.cpp:165] Memory required for data: 1177882200
  416. I0114 17:30:19.266616 86948 layer_factory.hpp:77] Creating layer relu4
  417. I0114 17:30:19.266621 86948 net.cpp:106] Creating Layer relu4
  418. I0114 17:30:19.266624 86948 net.cpp:454] relu4 <- conv4
  419. I0114 17:30:19.266630 86948 net.cpp:397] relu4 -> conv4 (in-place)
  420. I0114 17:30:19.266991 86948 net.cpp:150] Setting up relu4
  421. I0114 17:30:19.267014 86948 net.cpp:157] Top shape: 150 256 11 11 (4646400)
  422. I0114 17:30:19.267016 86948 net.cpp:165] Memory required for data: 1196467800
  423. I0114 17:30:19.267019 86948 layer_factory.hpp:77] Creating layer pool4
  424. I0114 17:30:19.267027 86948 net.cpp:106] Creating Layer pool4
  425. I0114 17:30:19.267041 86948 net.cpp:454] pool4 <- conv4
  426. I0114 17:30:19.267045 86948 net.cpp:411] pool4 -> pool4
  427. I0114 17:30:19.267223 86948 net.cpp:150] Setting up pool4
  428. I0114 17:30:19.267231 86948 net.cpp:157] Top shape: 150 256 6 6 (1382400)
  429. I0114 17:30:19.267246 86948 net.cpp:165] Memory required for data: 1201997400
  430. I0114 17:30:19.267248 86948 layer_factory.hpp:77] Creating layer conv5
  431. I0114 17:30:19.267258 86948 net.cpp:106] Creating Layer conv5
  432. I0114 17:30:19.267273 86948 net.cpp:454] conv5 <- pool4
  433. I0114 17:30:19.267280 86948 net.cpp:411] conv5 -> conv5
  434. I0114 17:30:19.272622 86948 net.cpp:150] Setting up conv5
  435. I0114 17:30:19.272635 86948 net.cpp:157] Top shape: 150 256 4 4 (614400)
  436. I0114 17:30:19.272639 86948 net.cpp:165] Memory required for data: 1204455000
  437. I0114 17:30:19.272647 86948 layer_factory.hpp:77] Creating layer relu5
  438. I0114 17:30:19.272652 86948 net.cpp:106] Creating Layer relu5
  439. I0114 17:30:19.272655 86948 net.cpp:454] relu5 <- conv5
  440. I0114 17:30:19.272660 86948 net.cpp:397] relu5 -> conv5 (in-place)
  441. I0114 17:30:19.272987 86948 net.cpp:150] Setting up relu5
  442. I0114 17:30:19.273010 86948 net.cpp:157] Top shape: 150 256 4 4 (614400)
  443. I0114 17:30:19.273011 86948 net.cpp:165] Memory required for data: 1206912600
  444. I0114 17:30:19.273026 86948 layer_factory.hpp:77] Creating layer pool5
  445. I0114 17:30:19.273056 86948 net.cpp:106] Creating Layer pool5
  446. I0114 17:30:19.273061 86948 net.cpp:454] pool5 <- conv5
  447. I0114 17:30:19.273064 86948 net.cpp:411] pool5 -> pool5
  448. I0114 17:30:19.273246 86948 net.cpp:150] Setting up pool5
  449. I0114 17:30:19.273253 86948 net.cpp:157] Top shape: 150 256 2 2 (153600)
  450. I0114 17:30:19.273267 86948 net.cpp:165] Memory required for data: 1207527000
  451. I0114 17:30:19.273270 86948 layer_factory.hpp:77] Creating layer fc6
  452. I0114 17:30:19.273298 86948 net.cpp:106] Creating Layer fc6
  453. I0114 17:30:19.273301 86948 net.cpp:454] fc6 <- pool5
  454. I0114 17:30:19.273306 86948 net.cpp:411] fc6 -> fc6
  455. I0114 17:30:19.289192 86948 net.cpp:150] Setting up fc6
  456. I0114 17:30:19.289223 86948 net.cpp:157] Top shape: 150 2048 (307200)
  457. I0114 17:30:19.289237 86948 net.cpp:165] Memory required for data: 1208755800
  458. I0114 17:30:19.289244 86948 layer_factory.hpp:77] Creating layer relu6
  459. I0114 17:30:19.289252 86948 net.cpp:106] Creating Layer relu6
  460. I0114 17:30:19.289257 86948 net.cpp:454] relu6 <- fc6
  461. I0114 17:30:19.289260 86948 net.cpp:397] relu6 -> fc6 (in-place)
  462. I0114 17:30:19.289674 86948 net.cpp:150] Setting up relu6
  463. I0114 17:30:19.289696 86948 net.cpp:157] Top shape: 150 2048 (307200)
  464. I0114 17:30:19.289700 86948 net.cpp:165] Memory required for data: 1209984600
  465. I0114 17:30:19.289702 86948 layer_factory.hpp:77] Creating layer fc7
  466. I0114 17:30:19.289722 86948 net.cpp:106] Creating Layer fc7
  467. I0114 17:30:19.289726 86948 net.cpp:454] fc7 <- fc6
  468. I0114 17:30:19.289732 86948 net.cpp:411] fc7 -> fc7
  469. I0114 17:30:19.322726 86948 net.cpp:150] Setting up fc7
  470. I0114 17:30:19.322765 86948 net.cpp:157] Top shape: 150 2048 (307200)
  471. I0114 17:30:19.322769 86948 net.cpp:165] Memory required for data: 1211213400
  472. I0114 17:30:19.322777 86948 layer_factory.hpp:77] Creating layer relu7
  473. I0114 17:30:19.322787 86948 net.cpp:106] Creating Layer relu7
  474. I0114 17:30:19.322790 86948 net.cpp:454] relu7 <- fc7
  475. I0114 17:30:19.322796 86948 net.cpp:397] relu7 -> fc7 (in-place)
  476. I0114 17:30:19.323101 86948 net.cpp:150] Setting up relu7
  477. I0114 17:30:19.323110 86948 net.cpp:157] Top shape: 150 2048 (307200)
  478. I0114 17:30:19.323124 86948 net.cpp:165] Memory required for data: 1212442200
  479. I0114 17:30:19.323127 86948 layer_factory.hpp:77] Creating layer fc8
  480. I0114 17:30:19.323142 86948 net.cpp:106] Creating Layer fc8
  481. I0114 17:30:19.323144 86948 net.cpp:454] fc8 <- fc7
  482. I0114 17:30:19.323149 86948 net.cpp:411] fc8 -> fc8
  483. I0114 17:30:19.325206 86948 net.cpp:150] Setting up fc8
  484. I0114 17:30:19.325217 86948 net.cpp:157] Top shape: 150 101 (15150)
  485. I0114 17:30:19.325220 86948 net.cpp:165] Memory required for data: 1212502800
  486. I0114 17:30:19.325227 86948 layer_factory.hpp:77] Creating layer loss
  487. I0114 17:30:19.325237 86948 net.cpp:106] Creating Layer loss
  488. I0114 17:30:19.325240 86948 net.cpp:454] loss <- fc8
  489. I0114 17:30:19.325244 86948 net.cpp:454] loss <- label
  490. I0114 17:30:19.325250 86948 net.cpp:411] loss -> loss
  491. I0114 17:30:19.325264 86948 layer_factory.hpp:77] Creating layer loss
  492. I0114 17:30:19.326247 86948 net.cpp:150] Setting up loss
  493. I0114 17:30:19.326261 86948 net.cpp:157] Top shape: (1)
  494. I0114 17:30:19.326263 86948 net.cpp:160] with loss weight 1
  495. I0114 17:30:19.326287 86948 net.cpp:165] Memory required for data: 1212502804
  496. I0114 17:30:19.326289 86948 net.cpp:226] loss needs backward computation.
  497. I0114 17:30:19.326292 86948 net.cpp:226] fc8 needs backward computation.
  498. I0114 17:30:19.326295 86948 net.cpp:226] relu7 needs backward computation.
  499. I0114 17:30:19.326297 86948 net.cpp:226] fc7 needs backward computation.
  500. I0114 17:30:19.326300 86948 net.cpp:226] relu6 needs backward computation.
  501. I0114 17:30:19.326303 86948 net.cpp:226] fc6 needs backward computation.
  502. I0114 17:30:19.326305 86948 net.cpp:226] pool5 needs backward computation.
  503. I0114 17:30:19.326309 86948 net.cpp:226] relu5 needs backward computation.
  504. I0114 17:30:19.326311 86948 net.cpp:226] conv5 needs backward computation.
  505. I0114 17:30:19.326314 86948 net.cpp:226] pool4 needs backward computation.
  506. I0114 17:30:19.326318 86948 net.cpp:226] relu4 needs backward computation.
  507. I0114 17:30:19.326320 86948 net.cpp:226] conv4 needs backward computation.
  508. I0114 17:30:19.326339 86948 net.cpp:226] pool3 needs backward computation.
  509. I0114 17:30:19.326342 86948 net.cpp:226] relu3 needs backward computation.
  510. I0114 17:30:19.326344 86948 net.cpp:226] conv3 needs backward computation.
  511. I0114 17:30:19.326347 86948 net.cpp:226] pool2 needs backward computation.
  512. I0114 17:30:19.326350 86948 net.cpp:226] relu2 needs backward computation.
  513. I0114 17:30:19.326352 86948 net.cpp:226] conv2 needs backward computation.
  514. I0114 17:30:19.326355 86948 net.cpp:226] pool1 needs backward computation.
  515. I0114 17:30:19.326359 86948 net.cpp:226] relu1 needs backward computation.
  516. I0114 17:30:19.326361 86948 net.cpp:226] conv1 needs backward computation.
  517. I0114 17:30:19.326364 86948 net.cpp:228] data does not need backward computation.
  518. I0114 17:30:19.326366 86948 net.cpp:270] This network produces output loss
  519. I0114 17:30:19.326382 86948 net.cpp:283] Network initialization done.
  520. I0114 17:30:19.327203 86948 solver.cpp:181] Creating test net (#0) specified by net file: models/mv16f/mv_train1.prototxt
  521. I0114 17:30:19.327262 86948 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
  522. I0114 17:30:19.327458 86948 net.cpp:49] Initializing net from parameters:
  523. name: "mv_16f1"
  524. state {
  525. phase: TEST
  526. }
  527. layer {
  528. name: "data"
  529. type: "HDF5Data"
  530. top: "data"
  531. top: "label"
  532. include {
  533. phase: TEST
  534. }
  535. hdf5_data_param {
  536. source: "/home/fe/anilil/caffe/models/mv16f/test.txt"
  537. batch_size: 50
  538. }
  539. }
  540. layer {
  541. name: "conv1"
  542. type: "Convolution"
  543. bottom: "data"
  544. top: "conv1"
  545. param {
  546. lr_mult: 1
  547. decay_mult: 1
  548. }
  549. param {
  550. lr_mult: 1
  551. decay_mult: 1
  552. }
  553. convolution_param {
  554. num_output: 64
  555. kernel_size: 3
  556. weight_filler {
  557. type: "xavier"
  558. }
  559. bias_filler {
  560. type: "xavier"
  561. }
  562. }
  563. }
  564. layer {
  565. name: "relu1"
  566. type: "ReLU"
  567. bottom: "conv1"
  568. top: "conv1"
  569. }
  570. layer {
  571. name: "pool1"
  572. type: "Pooling"
  573. bottom: "conv1"
  574. top: "pool1"
  575. pooling_param {
  576. pool: AVE
  577. kernel_size: 2
  578. stride: 1
  579. }
  580. }
  581. layer {
  582. name: "conv2"
  583. type: "Convolution"
  584. bottom: "pool1"
  585. top: "conv2"
  586. param {
  587. lr_mult: 1
  588. decay_mult: 1
  589. }
  590. param {
  591. lr_mult: 1
  592. decay_mult: 1
  593. }
  594. convolution_param {
  595. num_output: 128
  596. kernel_size: 3
  597. stride: 1
  598. weight_filler {
  599. type: "xavier"
  600. }
  601. bias_filler {
  602. type: "xavier"
  603. }
  604. }
  605. }
  606. layer {
  607. name: "relu2"
  608. type: "ReLU"
  609. bottom: "conv2"
  610. top: "conv2"
  611. }
  612. layer {
  613. name: "pool2"
  614. type: "Pooling"
  615. bottom: "conv2"
  616. top: "pool2"
  617. pooling_param {
  618. pool: AVE
  619. kernel_size: 2
  620. stride: 2
  621. }
  622. }
  623. layer {
  624. name: "conv3"
  625. type: "Convolution"
  626. bottom: "pool2"
  627. top: "conv3"
  628. param {
  629. lr_mult: 1
  630. decay_mult: 1
  631. }
  632. param {
  633. lr_mult: 1
  634. decay_mult: 1
  635. }
  636. convolution_param {
  637. num_output: 256
  638. kernel_size: 3
  639. stride: 1
  640. weight_filler {
  641. type: "xavier"
  642. }
  643. bias_filler {
  644. type: "xavier"
  645. }
  646. }
  647. }
  648. layer {
  649. name: "relu3"
  650. type: "ReLU"
  651. bottom: "conv3"
  652. top: "conv3"
  653. }
  654. layer {
  655. name: "pool3"
  656. type: "Pooling"
  657. bottom: "conv3"
  658. top: "pool3"
  659. pooling_param {
  660. pool: AVE
  661. kernel_size: 2
  662. stride: 2
  663. }
  664. }
  665. layer {
  666. name: "conv4"
  667. type: "Convolution"
  668. bottom: "pool3"
  669. top: "conv4"
  670. param {
  671. lr_mult: 1
  672. decay_mult: 1
  673. }
  674. param {
  675. lr_mult: 1
  676. decay_mult: 1
  677. }
  678. convolution_param {
  679. num_output: 256
  680. kernel_size: 3
  681. stride: 1
  682. weight_filler {
  683. type: "xavier"
  684. }
  685. bias_filler {
  686. type: "xavier"
  687. }
  688. }
  689. }
  690. layer {
  691. name: "relu4"
  692. type: "ReLU"
  693. bottom: "conv4"
  694. top: "conv4"
  695. }
  696. layer {
  697. name: "pool4"
  698. type: "Pooling"
  699. bottom: "conv4"
  700. top: "pool4"
  701. pooling_param {
  702. pool: AVE
  703. kernel_size: 2
  704. stride: 2
  705. }
  706. }
  707. layer {
  708. name: "conv5"
  709. type: "Convolution"
  710. bottom: "pool4"
  711. top: "conv5"
  712. param {
  713. lr_mult: 1
  714. decay_mult: 1
  715. }
  716. param {
  717. lr_mult: 1
  718. decay_mult: 1
  719. }
  720. convolution_param {
  721. num_output: 256
  722. kernel_size: 3
  723. stride: 1
  724. weight_filler {
  725. type: "xavier"
  726. }
  727. bias_filler {
  728. type: "xavier"
  729. }
  730. }
  731. }
  732. layer {
  733. name: "relu5"
  734. type: "ReLU"
  735. bottom: "conv5"
  736. top: "conv5"
  737. }
  738. layer {
  739. name: "pool5"
  740. type: "Pooling"
  741. bottom: "conv5"
  742. top: "pool5"
  743. pooling_param {
  744. pool: AVE
  745. kernel_size: 2
  746. stride: 2
  747. }
  748. }
  749. layer {
  750. name: "fc6"
  751. type: "InnerProduct"
  752. bottom: "pool5"
  753. top: "fc6"
  754. param {
  755. lr_mult: 1
  756. decay_mult: 1
  757. }
  758. param {
  759. lr_mult: 2
  760. decay_mult: 0
  761. }
  762. inner_product_param {
  763. num_output: 2048
  764. weight_filler {
  765. type: "xavier"
  766. }
  767. bias_filler {
  768. type: "xavier"
  769. }
  770. }
  771. }
  772. layer {
  773. name: "relu6"
  774. type: "ReLU"
  775. bottom: "fc6"
  776. top: "fc6"
  777. }
  778. layer {
  779. name: "fc7"
  780. type: "InnerProduct"
  781. bottom: "fc6"
  782. top: "fc7"
  783. param {
  784. lr_mult: 1
  785. decay_mult: 1
  786. }
  787. param {
  788. lr_mult: 2
  789. decay_mult: 0
  790. }
  791. inner_product_param {
  792. num_output: 2048
  793. weight_filler {
  794. type: "xavier"
  795. }
  796. bias_filler {
  797. type: "xavier"
  798. }
  799. }
  800. }
  801. layer {
  802. name: "relu7"
  803. type: "ReLU"
  804. bottom: "fc7"
  805. top: "fc7"
  806. }
  807. layer {
  808. name: "fc8"
  809. type: "InnerProduct"
  810. bottom: "fc7"
  811. top: "fc8"
  812. param {
  813. lr_mult: 1
  814. decay_mult: 1
  815. }
  816. param {
  817. lr_mult: 2
  818. decay_mult: 0
  819. }
  820. inner_product_param {
  821. num_output: 101
  822. weight_filler {
  823. type: "xavier"
  824. }
  825. bias_filler {
  826. type: "xavier"
  827. }
  828. }
  829. }
  830. layer {
  831. name: "accuracy"
  832. type: "Accuracy"
  833. bottom: "fc8"
  834. bottom: "label"
  835. top: "accuracy"
  836. include {
  837. phase: TEST
  838. }
  839. }
  840. layer {
  841. name: "loss"
  842. type: "SoftmaxWithLoss"
  843. bottom: "fc8"
  844. bottom: "label"
  845. top: "loss"
  846. }
  847. I0114 17:30:19.327607 86948 layer_factory.hpp:77] Creating layer data
  848. I0114 17:30:19.327620 86948 net.cpp:106] Creating Layer data
  849. I0114 17:30:19.327623 86948 net.cpp:411] data -> data
  850. I0114 17:30:19.327630 86948 net.cpp:411] data -> label
  851. I0114 17:30:19.327636 86948 hdf5_data_layer.cpp:79] Loading list of HDF5 filenames from: /home/fe/anilil/caffe/models/mv16f/test.txt
  852. I0114 17:30:19.327733 86948 hdf5_data_layer.cpp:93] Number of HDF5 files: 205
  853. I0114 17:30:20.281930 86948 net.cpp:150] Setting up data
  854. I0114 17:30:20.281970 86948 net.cpp:157] Top shape: 50 48 58 58 (8073600)
  855. I0114 17:30:20.281988 86948 net.cpp:157] Top shape: 50 1 (50)
  856. I0114 17:30:20.281991 86948 net.cpp:165] Memory required for data: 32294600
  857. I0114 17:30:20.281996 86948 layer_factory.hpp:77] Creating layer label_data_1_split
  858. I0114 17:30:20.282017 86948 net.cpp:106] Creating Layer label_data_1_split
  859. I0114 17:30:20.282021 86948 net.cpp:454] label_data_1_split <- label
  860. I0114 17:30:20.282027 86948 net.cpp:411] label_data_1_split -> label_data_1_split_0
  861. I0114 17:30:20.282037 86948 net.cpp:411] label_data_1_split -> label_data_1_split_1
  862. I0114 17:30:20.282089 86948 net.cpp:150] Setting up label_data_1_split
  863. I0114 17:30:20.282109 86948 net.cpp:157] Top shape: 50 1 (50)
  864. I0114 17:30:20.282112 86948 net.cpp:157] Top shape: 50 1 (50)
  865. I0114 17:30:20.282126 86948 net.cpp:165] Memory required for data: 32295000
  866. I0114 17:30:20.282130 86948 layer_factory.hpp:77] Creating layer conv1
  867. I0114 17:30:20.282142 86948 net.cpp:106] Creating Layer conv1
  868. I0114 17:30:20.282146 86948 net.cpp:454] conv1 <- data
  869. I0114 17:30:20.282151 86948 net.cpp:411] conv1 -> conv1
  870. I0114 17:30:20.283648 86948 net.cpp:150] Setting up conv1
  871. I0114 17:30:20.283661 86948 net.cpp:157] Top shape: 50 64 56 56 (10035200)
  872. I0114 17:30:20.283664 86948 net.cpp:165] Memory required for data: 72435800
  873. I0114 17:30:20.283674 86948 layer_factory.hpp:77] Creating layer relu1
  874. I0114 17:30:20.283679 86948 net.cpp:106] Creating Layer relu1
  875. I0114 17:30:20.283682 86948 net.cpp:454] relu1 <- conv1
  876. I0114 17:30:20.283686 86948 net.cpp:397] relu1 -> conv1 (in-place)
  877. I0114 17:30:20.283843 86948 net.cpp:150] Setting up relu1
  878. I0114 17:30:20.283851 86948 net.cpp:157] Top shape: 50 64 56 56 (10035200)
  879. I0114 17:30:20.283865 86948 net.cpp:165] Memory required for data: 112576600
  880. I0114 17:30:20.283869 86948 layer_factory.hpp:77] Creating layer pool1
  881. I0114 17:30:20.283875 86948 net.cpp:106] Creating Layer pool1
  882. I0114 17:30:20.283879 86948 net.cpp:454] pool1 <- conv1
  883. I0114 17:30:20.283882 86948 net.cpp:411] pool1 -> pool1
  884. I0114 17:30:20.284235 86948 net.cpp:150] Setting up pool1
  885. I0114 17:30:20.284245 86948 net.cpp:157] Top shape: 50 64 55 55 (9680000)
  886. I0114 17:30:20.284247 86948 net.cpp:165] Memory required for data: 151296600
  887. I0114 17:30:20.284250 86948 layer_factory.hpp:77] Creating layer conv2
  888. I0114 17:30:20.284260 86948 net.cpp:106] Creating Layer conv2
  889. I0114 17:30:20.284262 86948 net.cpp:454] conv2 <- pool1
  890. I0114 17:30:20.284266 86948 net.cpp:411] conv2 -> conv2
  891. I0114 17:30:20.285594 86948 net.cpp:150] Setting up conv2
  892. I0114 17:30:20.285606 86948 net.cpp:157] Top shape: 50 128 53 53 (17977600)
  893. I0114 17:30:20.285609 86948 net.cpp:165] Memory required for data: 223207000
  894. I0114 17:30:20.285617 86948 layer_factory.hpp:77] Creating layer relu2
  895. I0114 17:30:20.285621 86948 net.cpp:106] Creating Layer relu2
  896. I0114 17:30:20.285624 86948 net.cpp:454] relu2 <- conv2
  897. I0114 17:30:20.285629 86948 net.cpp:397] relu2 -> conv2 (in-place)
  898. I0114 17:30:20.285780 86948 net.cpp:150] Setting up relu2
  899. I0114 17:30:20.285787 86948 net.cpp:157] Top shape: 50 128 53 53 (17977600)
  900. I0114 17:30:20.285801 86948 net.cpp:165] Memory required for data: 295117400
  901. I0114 17:30:20.285804 86948 layer_factory.hpp:77] Creating layer pool2
  902. I0114 17:30:20.285809 86948 net.cpp:106] Creating Layer pool2
  903. I0114 17:30:20.285812 86948 net.cpp:454] pool2 <- conv2
  904. I0114 17:30:20.285816 86948 net.cpp:411] pool2 -> pool2
  905. I0114 17:30:20.286202 86948 net.cpp:150] Setting up pool2
  906. I0114 17:30:20.286224 86948 net.cpp:157] Top shape: 50 128 27 27 (4665600)
  907. I0114 17:30:20.286226 86948 net.cpp:165] Memory required for data: 313779800
  908. I0114 17:30:20.286231 86948 layer_factory.hpp:77] Creating layer conv3
  909. I0114 17:30:20.286250 86948 net.cpp:106] Creating Layer conv3
  910. I0114 17:30:20.286253 86948 net.cpp:454] conv3 <- pool2
  911. I0114 17:30:20.286259 86948 net.cpp:411] conv3 -> conv3
  912. I0114 17:30:20.289516 86948 net.cpp:150] Setting up conv3
  913. I0114 17:30:20.289530 86948 net.cpp:157] Top shape: 50 256 25 25 (8000000)
  914. I0114 17:30:20.289532 86948 net.cpp:165] Memory required for data: 345779800
  915. I0114 17:30:20.289541 86948 layer_factory.hpp:77] Creating layer relu3
  916. I0114 17:30:20.289547 86948 net.cpp:106] Creating Layer relu3
  917. I0114 17:30:20.289551 86948 net.cpp:454] relu3 <- conv3
  918. I0114 17:30:20.289554 86948 net.cpp:397] relu3 -> conv3 (in-place)
  919. I0114 17:30:20.289873 86948 net.cpp:150] Setting up relu3
  920. I0114 17:30:20.289896 86948 net.cpp:157] Top shape: 50 256 25 25 (8000000)
  921. I0114 17:30:20.289898 86948 net.cpp:165] Memory required for data: 377779800
  922. I0114 17:30:20.289901 86948 layer_factory.hpp:77] Creating layer pool3
  923. I0114 17:30:20.289919 86948 net.cpp:106] Creating Layer pool3
  924. I0114 17:30:20.289922 86948 net.cpp:454] pool3 <- conv3
  925. I0114 17:30:20.289927 86948 net.cpp:411] pool3 -> pool3
  926. I0114 17:30:20.290241 86948 net.cpp:150] Setting up pool3
  927. I0114 17:30:20.290263 86948 net.cpp:157] Top shape: 50 256 13 13 (2163200)
  928. I0114 17:30:20.290266 86948 net.cpp:165] Memory required for data: 386432600
  929. I0114 17:30:20.290269 86948 layer_factory.hpp:77] Creating layer conv4
  930. I0114 17:30:20.290292 86948 net.cpp:106] Creating Layer conv4
  931. I0114 17:30:20.290294 86948 net.cpp:454] conv4 <- pool3
  932. I0114 17:30:20.290299 86948 net.cpp:411] conv4 -> conv4
  933. I0114 17:30:20.296136 86948 net.cpp:150] Setting up conv4
  934. I0114 17:30:20.296149 86948 net.cpp:157] Top shape: 50 256 11 11 (1548800)
  935. I0114 17:30:20.296152 86948 net.cpp:165] Memory required for data: 392627800
  936. I0114 17:30:20.296159 86948 layer_factory.hpp:77] Creating layer relu4
  937. I0114 17:30:20.296165 86948 net.cpp:106] Creating Layer relu4
  938. I0114 17:30:20.296169 86948 net.cpp:454] relu4 <- conv4
  939. I0114 17:30:20.296172 86948 net.cpp:397] relu4 -> conv4 (in-place)
  940. I0114 17:30:20.296542 86948 net.cpp:150] Setting up relu4
  941. I0114 17:30:20.296553 86948 net.cpp:157] Top shape: 50 256 11 11 (1548800)
  942. I0114 17:30:20.296567 86948 net.cpp:165] Memory required for data: 398823000
  943. I0114 17:30:20.296571 86948 layer_factory.hpp:77] Creating layer pool4
  944. I0114 17:30:20.296589 86948 net.cpp:106] Creating Layer pool4
  945. I0114 17:30:20.296592 86948 net.cpp:454] pool4 <- conv4
  946. I0114 17:30:20.296598 86948 net.cpp:411] pool4 -> pool4
  947. I0114 17:30:20.296790 86948 net.cpp:150] Setting up pool4
  948. I0114 17:30:20.296799 86948 net.cpp:157] Top shape: 50 256 6 6 (460800)
  949. I0114 17:30:20.296813 86948 net.cpp:165] Memory required for data: 400666200
  950. I0114 17:30:20.296816 86948 layer_factory.hpp:77] Creating layer conv5
  951. I0114 17:30:20.296838 86948 net.cpp:106] Creating Layer conv5
  952. I0114 17:30:20.296841 86948 net.cpp:454] conv5 <- pool4
  953. I0114 17:30:20.296847 86948 net.cpp:411] conv5 -> conv5
  954. I0114 17:30:20.302322 86948 net.cpp:150] Setting up conv5
  955. I0114 17:30:20.302346 86948 net.cpp:157] Top shape: 50 256 4 4 (204800)
  956. I0114 17:30:20.302350 86948 net.cpp:165] Memory required for data: 401485400
  957. I0114 17:30:20.302369 86948 layer_factory.hpp:77] Creating layer relu5
  958. I0114 17:30:20.302376 86948 net.cpp:106] Creating Layer relu5
  959. I0114 17:30:20.302379 86948 net.cpp:454] relu5 <- conv5
  960. I0114 17:30:20.302383 86948 net.cpp:397] relu5 -> conv5 (in-place)
  961. I0114 17:30:20.302727 86948 net.cpp:150] Setting up relu5
  962. I0114 17:30:20.302738 86948 net.cpp:157] Top shape: 50 256 4 4 (204800)
  963. I0114 17:30:20.302741 86948 net.cpp:165] Memory required for data: 402304600
  964. I0114 17:30:20.302743 86948 layer_factory.hpp:77] Creating layer pool5
  965. I0114 17:30:20.302752 86948 net.cpp:106] Creating Layer pool5
  966. I0114 17:30:20.302755 86948 net.cpp:454] pool5 <- conv5
  967. I0114 17:30:20.302762 86948 net.cpp:411] pool5 -> pool5
  968. I0114 17:30:20.302942 86948 net.cpp:150] Setting up pool5
  969. I0114 17:30:20.302963 86948 net.cpp:157] Top shape: 50 256 2 2 (51200)
  970. I0114 17:30:20.302965 86948 net.cpp:165] Memory required for data: 402509400
  971. I0114 17:30:20.302968 86948 layer_factory.hpp:77] Creating layer fc6
  972. I0114 17:30:20.302974 86948 net.cpp:106] Creating Layer fc6
  973. I0114 17:30:20.302978 86948 net.cpp:454] fc6 <- pool5
  974. I0114 17:30:20.302995 86948 net.cpp:411] fc6 -> fc6
  975. I0114 17:30:20.331744 86948 net.cpp:150] Setting up fc6
  976. I0114 17:30:20.331787 86948 net.cpp:157] Top shape: 50 2048 (102400)
  977. I0114 17:30:20.331796 86948 net.cpp:165] Memory required for data: 402919000
  978. I0114 17:30:20.331814 86948 layer_factory.hpp:77] Creating layer relu6
  979. I0114 17:30:20.331830 86948 net.cpp:106] Creating Layer relu6
  980. I0114 17:30:20.331838 86948 net.cpp:454] relu6 <- fc6
  981. I0114 17:30:20.331852 86948 net.cpp:397] relu6 -> fc6 (in-place)
  982. I0114 17:30:20.332582 86948 net.cpp:150] Setting up relu6
  983. I0114 17:30:20.332607 86948 net.cpp:157] Top shape: 50 2048 (102400)
  984. I0114 17:30:20.332613 86948 net.cpp:165] Memory required for data: 403328600
  985. I0114 17:30:20.332620 86948 layer_factory.hpp:77] Creating layer fc7
  986. I0114 17:30:20.332638 86948 net.cpp:106] Creating Layer fc7
  987. I0114 17:30:20.332645 86948 net.cpp:454] fc7 <- fc6
  988. I0114 17:30:20.332656 86948 net.cpp:411] fc7 -> fc7
  989. I0114 17:30:20.391887 86948 net.cpp:150] Setting up fc7
  990. I0114 17:30:20.391923 86948 net.cpp:157] Top shape: 50 2048 (102400)
  991. I0114 17:30:20.391929 86948 net.cpp:165] Memory required for data: 403738200
  992. I0114 17:30:20.391943 86948 layer_factory.hpp:77] Creating layer relu7
  993. I0114 17:30:20.391959 86948 net.cpp:106] Creating Layer relu7
  994. I0114 17:30:20.391966 86948 net.cpp:454] relu7 <- fc7
  995. I0114 17:30:20.391979 86948 net.cpp:397] relu7 -> fc7 (in-place)
  996. I0114 17:30:20.392313 86948 net.cpp:150] Setting up relu7
  997. I0114 17:30:20.392329 86948 net.cpp:157] Top shape: 50 2048 (102400)
  998. I0114 17:30:20.392334 86948 net.cpp:165] Memory required for data: 404147800
  999. I0114 17:30:20.392339 86948 layer_factory.hpp:77] Creating layer fc8
  1000. I0114 17:30:20.392351 86948 net.cpp:106] Creating Layer fc8
  1001. I0114 17:30:20.392356 86948 net.cpp:454] fc8 <- fc7
  1002. I0114 17:30:20.392366 86948 net.cpp:411] fc8 -> fc8
  1003. I0114 17:30:20.395522 86948 net.cpp:150] Setting up fc8
  1004. I0114 17:30:20.395540 86948 net.cpp:157] Top shape: 50 101 (5050)
  1005. I0114 17:30:20.395545 86948 net.cpp:165] Memory required for data: 404168000
  1006. I0114 17:30:20.395555 86948 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
  1007. I0114 17:30:20.395565 86948 net.cpp:106] Creating Layer fc8_fc8_0_split
  1008. I0114 17:30:20.395570 86948 net.cpp:454] fc8_fc8_0_split <- fc8
  1009. I0114 17:30:20.395576 86948 net.cpp:411] fc8_fc8_0_split -> fc8_fc8_0_split_0
  1010. I0114 17:30:20.395611 86948 net.cpp:411] fc8_fc8_0_split -> fc8_fc8_0_split_1
  1011. I0114 17:30:20.395666 86948 net.cpp:150] Setting up fc8_fc8_0_split
  1012. I0114 17:30:20.395676 86948 net.cpp:157] Top shape: 50 101 (5050)
  1013. I0114 17:30:20.395681 86948 net.cpp:157] Top shape: 50 101 (5050)
  1014. I0114 17:30:20.395685 86948 net.cpp:165] Memory required for data: 404208400
  1015. I0114 17:30:20.395689 86948 layer_factory.hpp:77] Creating layer accuracy
  1016. I0114 17:30:20.395707 86948 net.cpp:106] Creating Layer accuracy
  1017. I0114 17:30:20.395712 86948 net.cpp:454] accuracy <- fc8_fc8_0_split_0
  1018. I0114 17:30:20.395719 86948 net.cpp:454] accuracy <- label_data_1_split_0
  1019. I0114 17:30:20.395727 86948 net.cpp:411] accuracy -> accuracy
  1020. I0114 17:30:20.395741 86948 net.cpp:150] Setting up accuracy
  1021. I0114 17:30:20.395746 86948 net.cpp:157] Top shape: (1)
  1022. I0114 17:30:20.395750 86948 net.cpp:165] Memory required for data: 404208404
  1023. I0114 17:30:20.395756 86948 layer_factory.hpp:77] Creating layer loss
  1024. I0114 17:30:20.395762 86948 net.cpp:106] Creating Layer loss
  1025. I0114 17:30:20.395766 86948 net.cpp:454] loss <- fc8_fc8_0_split_1
  1026. I0114 17:30:20.395772 86948 net.cpp:454] loss <- label_data_1_split_1
  1027. I0114 17:30:20.395781 86948 net.cpp:411] loss -> loss
  1028. I0114 17:30:20.395792 86948 layer_factory.hpp:77] Creating layer loss
  1029. I0114 17:30:20.396368 86948 net.cpp:150] Setting up loss
  1030. I0114 17:30:20.396384 86948 net.cpp:157] Top shape: (1)
  1031. I0114 17:30:20.396389 86948 net.cpp:160] with loss weight 1
  1032. I0114 17:30:20.396401 86948 net.cpp:165] Memory required for data: 404208408
  1033. I0114 17:30:20.396406 86948 net.cpp:226] loss needs backward computation.
  1034. I0114 17:30:20.396412 86948 net.cpp:228] accuracy does not need backward computation.
  1035. I0114 17:30:20.396417 86948 net.cpp:226] fc8_fc8_0_split needs backward computation.
  1036. I0114 17:30:20.396421 86948 net.cpp:226] fc8 needs backward computation.
  1037. I0114 17:30:20.396425 86948 net.cpp:226] relu7 needs backward computation.
  1038. I0114 17:30:20.396430 86948 net.cpp:226] fc7 needs backward computation.
  1039. I0114 17:30:20.396433 86948 net.cpp:226] relu6 needs backward computation.
  1040. I0114 17:30:20.396437 86948 net.cpp:226] fc6 needs backward computation.
  1041. I0114 17:30:20.396442 86948 net.cpp:226] pool5 needs backward computation.
  1042. I0114 17:30:20.396446 86948 net.cpp:226] relu5 needs backward computation.
  1043. I0114 17:30:20.396451 86948 net.cpp:226] conv5 needs backward computation.
  1044. I0114 17:30:20.396456 86948 net.cpp:226] pool4 needs backward computation.
  1045. I0114 17:30:20.396461 86948 net.cpp:226] relu4 needs backward computation.
  1046. I0114 17:30:20.396466 86948 net.cpp:226] conv4 needs backward computation.
  1047. I0114 17:30:20.396469 86948 net.cpp:226] pool3 needs backward computation.
  1048. I0114 17:30:20.396474 86948 net.cpp:226] relu3 needs backward computation.
  1049. I0114 17:30:20.396478 86948 net.cpp:226] conv3 needs backward computation.
  1050. I0114 17:30:20.396482 86948 net.cpp:226] pool2 needs backward computation.
  1051. I0114 17:30:20.396487 86948 net.cpp:226] relu2 needs backward computation.
  1052. I0114 17:30:20.396492 86948 net.cpp:226] conv2 needs backward computation.
  1053. I0114 17:30:20.396495 86948 net.cpp:226] pool1 needs backward computation.
  1054. I0114 17:30:20.396500 86948 net.cpp:226] relu1 needs backward computation.
  1055. I0114 17:30:20.396504 86948 net.cpp:226] conv1 needs backward computation.
  1056. I0114 17:30:20.396510 86948 net.cpp:228] label_data_1_split does not need backward computation.
  1057. I0114 17:30:20.396517 86948 net.cpp:228] data does not need backward computation.
  1058. I0114 17:30:20.396520 86948 net.cpp:270] This network produces output accuracy
  1059. I0114 17:30:20.396525 86948 net.cpp:270] This network produces output loss
  1060. I0114 17:30:20.396551 86948 net.cpp:283] Network initialization done.
  1061. I0114 17:30:20.396718 86948 solver.cpp:60] Solver scaffolding done.
  1062. I0114 17:30:20.397513 86948 caffe.cpp:128] Finetuning from models/mv16f/mv16f1__iter_5000.caffemodel
  1063. I0114 17:30:31.195835 86948 parallel.cpp:391] GPUs pairs 0:1
  1064. I0114 17:30:31.513195 86948 net.cpp:99] Sharing layer data from root net
  1065. I0114 17:30:31.514991 86948 net.cpp:143] Created top blob 0 (shape: 150 48 58 58 (24220800)) for shared layer data
  1066. I0114 17:30:31.515151 86948 net.cpp:143] Created top blob 1 (shape: 150 1 (150)) for shared layer data
  1067. I0114 17:30:31.810956 86948 parallel.cpp:419] Starting Optimization
  1068. I0114 17:30:31.811024 86948 solver.cpp:288] Solving mv_16f1
  1069. I0114 17:30:31.811029 86948 solver.cpp:289] Learning Rate Policy: fixed
  1070. I0114 17:30:31.811134 86948 solver.cpp:341] Iteration 0, Testing net (#0)
  1071. I0114 17:30:32.401675 86948 solver.cpp:409] Test net output #0: accuracy = 0.03
  1072. I0114 17:30:32.401713 86948 solver.cpp:409] Test net output #1: loss = 6.22612 (* 1 = 6.22612 loss)
  1073. I0114 17:30:32.615201 86948 solver.cpp:237] Iteration 0, loss = 6.00407
  1074. I0114 17:30:32.615250 86948 solver.cpp:253] Train net output #0: loss = 6.00407 (* 1 = 6.00407 loss)
  1075. I0114 17:30:32.930352 86948 sgd_solver.cpp:106] Iteration 0, lr = 0.001
  1076. I0114 17:53:19.297849 86948 solver.cpp:341] Iteration 500, Testing net (#0)
  1077. I0114 17:53:20.085278 86948 solver.cpp:409] Test net output #0: accuracy = 0.08
  1078. I0114 17:53:20.085350 86948 solver.cpp:409] Test net output #1: loss = 4.21846 (* 1 = 4.21846 loss)
  1079. I0114 17:53:33.010169 86948 solver.cpp:237] Iteration 500, loss = 3.1388
  1080. I0114 17:53:33.010236 86948 solver.cpp:253] Train net output #0: loss = 3.1388 (* 1 = 3.1388 loss)
  1081. I0114 17:53:33.311209 86948 sgd_solver.cpp:106] Iteration 500, lr = 0.001
  1082. I0114 18:15:45.039989 86948 solver.cpp:341] Iteration 1000, Testing net (#0)
  1083. I0114 18:15:45.830772 86948 solver.cpp:409] Test net output #0: accuracy = 0.106
  1084. I0114 18:15:45.830831 86948 solver.cpp:409] Test net output #1: loss = 4.29256 (* 1 = 4.29256 loss)
  1085. I0114 18:16:07.472872 86948 solver.cpp:237] Iteration 1000, loss = 2.89553
  1086. I0114 18:16:07.472937 86948 solver.cpp:253] Train net output #0: loss = 2.89553 (* 1 = 2.89553 loss)
  1087. I0114 18:16:07.806648 86948 sgd_solver.cpp:106] Iteration 1000, lr = 0.001
  1088. I0114 18:38:54.700490 86948 solver.cpp:341] Iteration 1500, Testing net (#0)
  1089. I0114 18:38:55.191038 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1090. I0114 18:38:55.191095 86948 solver.cpp:409] Test net output #1: loss = 4.38897 (* 1 = 4.38897 loss)
  1091. I0114 18:39:08.405194 86948 solver.cpp:237] Iteration 1500, loss = 2.40281
  1092. I0114 18:39:08.405258 86948 solver.cpp:253] Train net output #0: loss = 2.40281 (* 1 = 2.40281 loss)
  1093. I0114 18:39:08.707820 86948 sgd_solver.cpp:106] Iteration 1500, lr = 0.001
  1094. I0114 19:01:43.483299 86948 solver.cpp:341] Iteration 2000, Testing net (#0)
  1095. I0114 19:01:57.033280 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1096. I0114 19:01:57.033339 86948 solver.cpp:409] Test net output #1: loss = 4.58599 (* 1 = 4.58599 loss)
  1097. I0114 19:02:15.622921 86948 solver.cpp:237] Iteration 2000, loss = 2.65678
  1098. I0114 19:02:15.623080 86948 solver.cpp:253] Train net output #0: loss = 2.65678 (* 1 = 2.65678 loss)
  1099. I0114 19:02:16.000957 86948 sgd_solver.cpp:106] Iteration 2000, lr = 0.001
  1100. I0114 19:25:49.882880 86948 solver.cpp:341] Iteration 2500, Testing net (#0)
  1101. I0114 19:25:50.668056 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1102. I0114 19:25:50.668131 86948 solver.cpp:409] Test net output #1: loss = 4.66895 (* 1 = 4.66895 loss)
  1103. I0114 19:26:04.158541 86948 solver.cpp:237] Iteration 2500, loss = 2.29065
  1104. I0114 19:26:04.158601 86948 solver.cpp:253] Train net output #0: loss = 2.29065 (* 1 = 2.29065 loss)
  1105. I0114 19:26:04.158622 86948 sgd_solver.cpp:106] Iteration 2500, lr = 0.001
  1106. I0114 19:48:19.053063 86948 solver.cpp:341] Iteration 3000, Testing net (#0)
  1107. I0114 19:48:19.546254 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1108. I0114 19:48:19.546349 86948 solver.cpp:409] Test net output #1: loss = 5.19176 (* 1 = 5.19176 loss)
  1109. I0114 19:48:32.406949 86948 solver.cpp:237] Iteration 3000, loss = 1.92223
  1110. I0114 19:48:32.407004 86948 solver.cpp:253] Train net output #0: loss = 1.92223 (* 1 = 1.92223 loss)
  1111. I0114 19:48:32.708853 86948 sgd_solver.cpp:106] Iteration 3000, lr = 0.001
  1112. I0114 20:11:15.198561 86948 solver.cpp:341] Iteration 3500, Testing net (#0)
  1113. I0114 20:11:15.988100 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1114. I0114 20:11:15.988169 86948 solver.cpp:409] Test net output #1: loss = 4.85761 (* 1 = 4.85761 loss)
  1115. I0114 20:11:29.490351 86948 solver.cpp:237] Iteration 3500, loss = 1.88295
  1116. I0114 20:11:29.490416 86948 solver.cpp:253] Train net output #0: loss = 1.88295 (* 1 = 1.88295 loss)
  1117. I0114 20:11:29.864094 86948 sgd_solver.cpp:106] Iteration 3500, lr = 0.001
  1118. I0114 20:33:36.204324 86948 solver.cpp:341] Iteration 4000, Testing net (#0)
  1119. I0114 20:33:49.203364 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1120. I0114 20:33:49.203444 86948 solver.cpp:409] Test net output #1: loss = 5.02526 (* 1 = 5.02526 loss)
  1121. I0114 20:34:01.464068 86948 solver.cpp:237] Iteration 4000, loss = 1.94949
  1122. I0114 20:34:01.464118 86948 solver.cpp:253] Train net output #0: loss = 1.94949 (* 1 = 1.94949 loss)
  1123. I0114 20:34:01.813761 86948 sgd_solver.cpp:106] Iteration 4000, lr = 0.001
  1124. I0114 20:56:38.518110 86948 solver.cpp:341] Iteration 4500, Testing net (#0)
  1125. I0114 20:56:39.302815 86948 solver.cpp:409] Test net output #0: accuracy = 0.146
  1126. I0114 20:56:39.302886 86948 solver.cpp:409] Test net output #1: loss = 5.19903 (* 1 = 5.19903 loss)
  1127. I0114 20:56:56.997396 86948 solver.cpp:237] Iteration 4500, loss = 1.50638
  1128. I0114 20:56:56.997449 86948 solver.cpp:253] Train net output #0: loss = 1.50638 (* 1 = 1.50638 loss)
  1129. I0114 20:56:57.367223 86948 sgd_solver.cpp:106] Iteration 4500, lr = 0.001
  1130. I0114 21:19:51.173880 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_5000.caffemodel
  1131. I0114 21:19:54.054301 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_5000.solverstate
  1132. I0114 21:19:54.105180 86948 solver.cpp:341] Iteration 5000, Testing net (#0)
  1133. I0114 21:19:54.591950 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1134. I0114 21:19:54.592030 86948 solver.cpp:409] Test net output #1: loss = 5.67052 (* 1 = 5.67052 loss)
  1135. I0114 21:20:07.774837 86948 solver.cpp:237] Iteration 5000, loss = 1.57842
  1136. I0114 21:20:07.774885 86948 solver.cpp:253] Train net output #0: loss = 1.57842 (* 1 = 1.57842 loss)
  1137. I0114 21:20:08.038023 86948 sgd_solver.cpp:106] Iteration 5000, lr = 0.001
  1138. I0114 21:43:29.756516 86948 solver.cpp:341] Iteration 5500, Testing net (#0)
  1139. I0114 21:43:30.251178 86948 solver.cpp:409] Test net output #0: accuracy = 0.134
  1140. I0114 21:43:30.251236 86948 solver.cpp:409] Test net output #1: loss = 6.00113 (* 1 = 6.00113 loss)
  1141. I0114 21:43:42.861708 86948 solver.cpp:237] Iteration 5500, loss = 1.49909
  1142. I0114 21:43:42.861762 86948 solver.cpp:253] Train net output #0: loss = 1.49909 (* 1 = 1.49909 loss)
  1143. I0114 21:43:43.164937 86948 sgd_solver.cpp:106] Iteration 5500, lr = 0.001
  1144. I0114 22:05:53.412967 86948 solver.cpp:341] Iteration 6000, Testing net (#0)
  1145. I0114 22:06:06.204192 86948 solver.cpp:409] Test net output #0: accuracy = 0.134
  1146. I0114 22:06:06.204249 86948 solver.cpp:409] Test net output #1: loss = 6.14714 (* 1 = 6.14714 loss)
  1147. I0114 22:06:19.697239 86948 solver.cpp:237] Iteration 6000, loss = 1.08655
  1148. I0114 22:06:19.697300 86948 solver.cpp:253] Train net output #0: loss = 1.08655 (* 1 = 1.08655 loss)
  1149. I0114 22:06:19.999053 86948 sgd_solver.cpp:106] Iteration 6000, lr = 0.001
  1150. I0114 22:28:44.623858 86948 solver.cpp:341] Iteration 6500, Testing net (#0)
  1151. I0114 22:28:45.117502 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1152. I0114 22:28:45.117586 86948 solver.cpp:409] Test net output #1: loss = 6.22176 (* 1 = 6.22176 loss)
  1153. I0114 22:29:02.504175 86948 solver.cpp:237] Iteration 6500, loss = 1.09802
  1154. I0114 22:29:02.504230 86948 solver.cpp:253] Train net output #0: loss = 1.09802 (* 1 = 1.09802 loss)
  1155. I0114 22:29:02.504263 86948 sgd_solver.cpp:106] Iteration 6500, lr = 0.001
  1156. I0114 22:51:08.072340 86948 solver.cpp:341] Iteration 7000, Testing net (#0)
  1157. I0114 22:51:08.570003 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1158. I0114 22:51:08.570077 86948 solver.cpp:409] Test net output #1: loss = 7.06356 (* 1 = 7.06356 loss)
  1159. I0114 22:51:21.792124 86948 solver.cpp:237] Iteration 7000, loss = 1.38146
  1160. I0114 22:51:21.792181 86948 solver.cpp:253] Train net output #0: loss = 1.38146 (* 1 = 1.38146 loss)
  1161. I0114 22:51:22.154963 86948 sgd_solver.cpp:106] Iteration 7000, lr = 0.001
  1162. I0114 23:13:57.562847 86948 solver.cpp:341] Iteration 7500, Testing net (#0)
  1163. I0114 23:13:58.056743 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1164. I0114 23:13:58.056813 86948 solver.cpp:409] Test net output #1: loss = 6.39689 (* 1 = 6.39689 loss)
  1165. I0114 23:14:19.756793 86948 solver.cpp:237] Iteration 7500, loss = 1.23898
  1166. I0114 23:14:19.756858 86948 solver.cpp:253] Train net output #0: loss = 1.23898 (* 1 = 1.23898 loss)
  1167. I0114 23:14:20.112123 86948 sgd_solver.cpp:106] Iteration 7500, lr = 0.001
  1168. I0114 23:37:58.058513 86948 solver.cpp:341] Iteration 8000, Testing net (#0)
  1169. I0114 23:38:11.531455 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1170. I0114 23:38:11.531529 86948 solver.cpp:409] Test net output #1: loss = 6.81792 (* 1 = 6.81792 loss)
  1171. I0114 23:38:25.274824 86948 solver.cpp:237] Iteration 8000, loss = 1.02914
  1172. I0114 23:38:25.274878 86948 solver.cpp:253] Train net output #0: loss = 1.02914 (* 1 = 1.02914 loss)
  1173. I0114 23:38:25.577234 86948 sgd_solver.cpp:106] Iteration 8000, lr = 0.001
  1174. I0115 00:05:17.607841 86948 solver.cpp:341] Iteration 8500, Testing net (#0)
  1175. I0115 00:05:18.392987 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1176. I0115 00:05:18.393034 86948 solver.cpp:409] Test net output #1: loss = 6.75926 (* 1 = 6.75926 loss)
  1177. I0115 00:05:45.961122 86948 solver.cpp:237] Iteration 8500, loss = 1.06676
  1178. I0115 00:05:45.961179 86948 solver.cpp:253] Train net output #0: loss = 1.06676 (* 1 = 1.06676 loss)
  1179. I0115 00:05:46.262384 86948 sgd_solver.cpp:106] Iteration 8500, lr = 0.001
  1180. I0115 00:32:18.132803 86948 solver.cpp:341] Iteration 9000, Testing net (#0)
  1181. I0115 00:32:18.625879 86948 solver.cpp:409] Test net output #0: accuracy = 0.134
  1182. I0115 00:32:18.625929 86948 solver.cpp:409] Test net output #1: loss = 7.14524 (* 1 = 7.14524 loss)
  1183. I0115 00:32:30.816737 86948 solver.cpp:237] Iteration 9000, loss = 0.860762
  1184. I0115 00:32:30.816799 86948 solver.cpp:253] Train net output #0: loss = 0.860762 (* 1 = 0.860762 loss)
  1185. I0115 00:32:31.143739 86948 sgd_solver.cpp:106] Iteration 9000, lr = 0.001
  1186. I0115 01:00:46.037446 86948 solver.cpp:341] Iteration 9500, Testing net (#0)
  1187. I0115 01:00:46.825520 86948 solver.cpp:409] Test net output #0: accuracy = 0.148
  1188. I0115 01:00:46.825567 86948 solver.cpp:409] Test net output #1: loss = 7.651 (* 1 = 7.651 loss)
  1189. I0115 01:01:12.079197 86948 solver.cpp:237] Iteration 9500, loss = 0.76086
  1190. I0115 01:01:12.079242 86948 solver.cpp:253] Train net output #0: loss = 0.76086 (* 1 = 0.76086 loss)
  1191. I0115 01:01:12.424202 86948 sgd_solver.cpp:106] Iteration 9500, lr = 0.001
  1192. I0115 01:23:29.878836 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_10000.caffemodel
  1193. I0115 01:23:31.686995 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_10000.solverstate
  1194. I0115 01:23:31.741595 86948 solver.cpp:341] Iteration 10000, Testing net (#0)
  1195. I0115 01:23:53.724689 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1196. I0115 01:23:53.724733 86948 solver.cpp:409] Test net output #1: loss = 7.54584 (* 1 = 7.54584 loss)
  1197. I0115 01:24:14.223534 86948 solver.cpp:237] Iteration 10000, loss = 0.774985
  1198. I0115 01:24:14.223733 86948 solver.cpp:253] Train net output #0: loss = 0.774985 (* 1 = 0.774985 loss)
  1199. I0115 01:24:14.602334 86948 sgd_solver.cpp:106] Iteration 10000, lr = 0.001
  1200. I0115 01:47:00.340564 86948 solver.cpp:341] Iteration 10500, Testing net (#0)
  1201. I0115 01:47:01.126672 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1202. I0115 01:47:01.126751 86948 solver.cpp:409] Test net output #1: loss = 7.71585 (* 1 = 7.71585 loss)
  1203. I0115 01:47:13.787405 86948 solver.cpp:237] Iteration 10500, loss = 0.845633
  1204. I0115 01:47:13.787449 86948 solver.cpp:253] Train net output #0: loss = 0.845633 (* 1 = 0.845633 loss)
  1205. I0115 01:47:13.787469 86948 sgd_solver.cpp:106] Iteration 10500, lr = 0.001
  1206. I0115 02:15:35.751904 86948 solver.cpp:341] Iteration 11000, Testing net (#0)
  1207. I0115 02:15:36.241739 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1208. I0115 02:15:36.241788 86948 solver.cpp:409] Test net output #1: loss = 9.06797 (* 1 = 9.06797 loss)
  1209. I0115 02:15:48.996397 86948 solver.cpp:237] Iteration 11000, loss = 0.546424
  1210. I0115 02:15:48.996453 86948 solver.cpp:253] Train net output #0: loss = 0.546424 (* 1 = 0.546424 loss)
  1211. I0115 02:15:48.996476 86948 sgd_solver.cpp:106] Iteration 11000, lr = 0.001
  1212. I0115 02:43:31.419049 86948 solver.cpp:341] Iteration 11500, Testing net (#0)
  1213. I0115 02:43:32.210687 86948 solver.cpp:409] Test net output #0: accuracy = 0.104
  1214. I0115 02:43:32.210752 86948 solver.cpp:409] Test net output #1: loss = 8.80403 (* 1 = 8.80403 loss)
  1215. I0115 02:43:44.821365 86948 solver.cpp:237] Iteration 11500, loss = 0.706103
  1216. I0115 02:43:44.821415 86948 solver.cpp:253] Train net output #0: loss = 0.706103 (* 1 = 0.706103 loss)
  1217. I0115 02:43:45.123746 86948 sgd_solver.cpp:106] Iteration 11500, lr = 0.001
  1218. I0115 03:14:23.524358 86948 solver.cpp:341] Iteration 12000, Testing net (#0)
  1219. I0115 03:14:46.023252 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1220. I0115 03:14:46.023300 86948 solver.cpp:409] Test net output #1: loss = 8.83962 (* 1 = 8.83962 loss)
  1221. I0115 03:14:59.884909 86948 solver.cpp:237] Iteration 12000, loss = 0.786133
  1222. I0115 03:14:59.885141 86948 solver.cpp:253] Train net output #0: loss = 0.786133 (* 1 = 0.786133 loss)
  1223. I0115 03:15:00.243067 86948 sgd_solver.cpp:106] Iteration 12000, lr = 0.001
  1224. I0115 03:41:47.587093 86948 solver.cpp:341] Iteration 12500, Testing net (#0)
  1225. I0115 03:41:48.073796 86948 solver.cpp:409] Test net output #0: accuracy = 0.098
  1226. I0115 03:41:48.073854 86948 solver.cpp:409] Test net output #1: loss = 9.32967 (* 1 = 9.32967 loss)
  1227. I0115 03:42:09.332177 86948 solver.cpp:237] Iteration 12500, loss = 0.707963
  1228. I0115 03:42:09.332263 86948 solver.cpp:253] Train net output #0: loss = 0.707963 (* 1 = 0.707963 loss)
  1229. I0115 03:42:09.676595 86948 sgd_solver.cpp:106] Iteration 12500, lr = 0.001
  1230. I0115 04:06:02.889426 86948 solver.cpp:341] Iteration 13000, Testing net (#0)
  1231. I0115 04:06:03.378060 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1232. I0115 04:06:03.378123 86948 solver.cpp:409] Test net output #1: loss = 9.17314 (* 1 = 9.17314 loss)
  1233. I0115 04:06:22.668329 86948 solver.cpp:237] Iteration 13000, loss = 0.509333
  1234. I0115 04:06:22.668419 86948 solver.cpp:253] Train net output #0: loss = 0.509333 (* 1 = 0.509333 loss)
  1235. I0115 04:06:22.970981 86948 sgd_solver.cpp:106] Iteration 13000, lr = 0.001
  1236. I0115 04:32:01.955811 86948 solver.cpp:341] Iteration 13500, Testing net (#0)
  1237. I0115 04:32:02.444028 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1238. I0115 04:32:02.444092 86948 solver.cpp:409] Test net output #1: loss = 9.12479 (* 1 = 9.12479 loss)
  1239. I0115 04:32:20.073575 86948 solver.cpp:237] Iteration 13500, loss = 0.500275
  1240. I0115 04:32:20.073634 86948 solver.cpp:253] Train net output #0: loss = 0.500275 (* 1 = 0.500275 loss)
  1241. I0115 04:32:20.435401 86948 sgd_solver.cpp:106] Iteration 13500, lr = 0.001
  1242. I0115 04:57:03.939975 86948 solver.cpp:341] Iteration 14000, Testing net (#0)
  1243. I0115 04:57:24.463392 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1244. I0115 04:57:24.463464 86948 solver.cpp:409] Test net output #1: loss = 9.76042 (* 1 = 9.76042 loss)
  1245. I0115 04:57:48.684335 86948 solver.cpp:237] Iteration 14000, loss = 0.390494
  1246. I0115 04:57:48.684521 86948 solver.cpp:253] Train net output #0: loss = 0.390494 (* 1 = 0.390494 loss)
  1247. I0115 04:57:49.035643 86948 sgd_solver.cpp:106] Iteration 14000, lr = 0.001
  1248. I0115 05:26:57.706825 86948 solver.cpp:341] Iteration 14500, Testing net (#0)
  1249. I0115 05:26:58.193374 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1250. I0115 05:26:58.193425 86948 solver.cpp:409] Test net output #1: loss = 10.1548 (* 1 = 10.1548 loss)
  1251. I0115 05:27:13.510231 86948 solver.cpp:237] Iteration 14500, loss = 0.692382
  1252. I0115 05:27:13.510294 86948 solver.cpp:253] Train net output #0: loss = 0.692382 (* 1 = 0.692382 loss)
  1253. I0115 05:27:13.812265 86948 sgd_solver.cpp:106] Iteration 14500, lr = 0.001
  1254. I0115 05:55:32.336146 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_15000.caffemodel
  1255. I0115 05:55:34.037616 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_15000.solverstate
  1256. I0115 05:55:34.086087 86948 solver.cpp:341] Iteration 15000, Testing net (#0)
  1257. I0115 05:55:34.573166 86948 solver.cpp:409] Test net output #0: accuracy = 0.106
  1258. I0115 05:55:34.573215 86948 solver.cpp:409] Test net output #1: loss = 9.8564 (* 1 = 9.8564 loss)
  1259. I0115 05:55:55.581955 86948 solver.cpp:237] Iteration 15000, loss = 0.507724
  1260. I0115 05:55:55.582023 86948 solver.cpp:253] Train net output #0: loss = 0.507724 (* 1 = 0.507724 loss)
  1261. I0115 05:55:55.884019 86948 sgd_solver.cpp:106] Iteration 15000, lr = 0.001
  1262. I0115 06:23:44.102008 86948 solver.cpp:341] Iteration 15500, Testing net (#0)
  1263. I0115 06:23:44.889958 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1264. I0115 06:23:44.890012 86948 solver.cpp:409] Test net output #1: loss = 10.4728 (* 1 = 10.4728 loss)
  1265. I0115 06:24:01.230536 86948 solver.cpp:237] Iteration 15500, loss = 0.417666
  1266. I0115 06:24:01.230588 86948 solver.cpp:253] Train net output #0: loss = 0.417666 (* 1 = 0.417666 loss)
  1267. I0115 06:24:01.606312 86948 sgd_solver.cpp:106] Iteration 15500, lr = 0.001
  1268. I0115 06:48:03.508607 86948 solver.cpp:341] Iteration 16000, Testing net (#0)
  1269. I0115 06:48:26.399644 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1270. I0115 06:48:26.399693 86948 solver.cpp:409] Test net output #1: loss = 10.979 (* 1 = 10.979 loss)
  1271. I0115 06:48:27.674831 86948 solver.cpp:237] Iteration 16000, loss = 0.40489
  1272. I0115 06:48:27.674890 86948 solver.cpp:253] Train net output #0: loss = 0.40489 (* 1 = 0.40489 loss)
  1273. I0115 06:48:28.055383 86948 sgd_solver.cpp:106] Iteration 16000, lr = 0.001
  1274. I0115 07:15:44.285596 86948 solver.cpp:341] Iteration 16500, Testing net (#0)
  1275. I0115 07:15:45.069802 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1276. I0115 07:15:45.069869 86948 solver.cpp:409] Test net output #1: loss = 10.3104 (* 1 = 10.3104 loss)
  1277. I0115 07:16:09.445392 86948 solver.cpp:237] Iteration 16500, loss = 0.29393
  1278. I0115 07:16:09.445464 86948 solver.cpp:253] Train net output #0: loss = 0.29393 (* 1 = 0.29393 loss)
  1279. I0115 07:16:09.445511 86948 sgd_solver.cpp:106] Iteration 16500, lr = 0.001
  1280. I0115 07:38:32.618374 86948 solver.cpp:341] Iteration 17000, Testing net (#0)
  1281. I0115 07:38:33.405944 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1282. I0115 07:38:33.405997 86948 solver.cpp:409] Test net output #1: loss = 11.4087 (* 1 = 11.4087 loss)
  1283. I0115 07:39:02.668887 86948 solver.cpp:237] Iteration 17000, loss = 0.475379
  1284. I0115 07:39:02.669137 86948 solver.cpp:253] Train net output #0: loss = 0.475379 (* 1 = 0.475379 loss)
  1285. I0115 07:39:03.016024 86948 sgd_solver.cpp:106] Iteration 17000, lr = 0.001
  1286. I0115 08:07:22.610738 86948 solver.cpp:341] Iteration 17500, Testing net (#0)
  1287. I0115 08:07:23.399281 86948 solver.cpp:409] Test net output #0: accuracy = 0.098
  1288. I0115 08:07:23.399346 86948 solver.cpp:409] Test net output #1: loss = 11.3343 (* 1 = 11.3343 loss)
  1289. I0115 08:07:35.556828 86948 solver.cpp:237] Iteration 17500, loss = 0.412808
  1290. I0115 08:07:35.556901 86948 solver.cpp:253] Train net output #0: loss = 0.412808 (* 1 = 0.412808 loss)
  1291. I0115 08:07:35.929672 86948 sgd_solver.cpp:106] Iteration 17500, lr = 0.001
  1292. I0115 08:35:21.397052 86948 solver.cpp:341] Iteration 18000, Testing net (#0)
  1293. I0115 08:35:37.317168 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1294. I0115 08:35:37.317226 86948 solver.cpp:409] Test net output #1: loss = 11.6325 (* 1 = 11.6325 loss)
  1295. I0115 08:35:54.273187 86948 solver.cpp:237] Iteration 18000, loss = 0.330421
  1296. I0115 08:35:54.273433 86948 solver.cpp:253] Train net output #0: loss = 0.330421 (* 1 = 0.330421 loss)
  1297. I0115 08:35:54.644815 86948 sgd_solver.cpp:106] Iteration 18000, lr = 0.001
  1298. I0115 09:04:32.714617 86948 solver.cpp:341] Iteration 18500, Testing net (#0)
  1299. I0115 09:04:33.499600 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1300. I0115 09:04:33.499655 86948 solver.cpp:409] Test net output #1: loss = 11.2051 (* 1 = 11.2051 loss)
  1301. I0115 09:04:45.729574 86948 solver.cpp:237] Iteration 18500, loss = 0.248244
  1302. I0115 09:04:45.729639 86948 solver.cpp:253] Train net output #0: loss = 0.248244 (* 1 = 0.248244 loss)
  1303. I0115 09:04:46.056254 86948 sgd_solver.cpp:106] Iteration 18500, lr = 0.001
  1304. I0115 09:33:10.057044 86948 solver.cpp:341] Iteration 19000, Testing net (#0)
  1305. I0115 09:33:10.552685 86948 solver.cpp:409] Test net output #0: accuracy = 0.088
  1306. I0115 09:33:10.552734 86948 solver.cpp:409] Test net output #1: loss = 12.5566 (* 1 = 12.5566 loss)
  1307. I0115 09:33:27.995590 86948 solver.cpp:237] Iteration 19000, loss = 0.353737
  1308. I0115 09:33:27.995659 86948 solver.cpp:253] Train net output #0: loss = 0.353737 (* 1 = 0.353737 loss)
  1309. I0115 09:33:28.297780 86948 sgd_solver.cpp:106] Iteration 19000, lr = 0.001
  1310. I0115 09:59:55.573832 86948 solver.cpp:341] Iteration 19500, Testing net (#0)
  1311. I0115 09:59:56.362241 86948 solver.cpp:409] Test net output #0: accuracy = 0.1
  1312. I0115 09:59:56.362289 86948 solver.cpp:409] Test net output #1: loss = 12.3916 (* 1 = 12.3916 loss)
  1313. I0115 10:00:18.836007 86948 solver.cpp:237] Iteration 19500, loss = 0.343227
  1314. I0115 10:00:18.836064 86948 solver.cpp:253] Train net output #0: loss = 0.343227 (* 1 = 0.343227 loss)
  1315. I0115 10:00:19.138316 86948 sgd_solver.cpp:106] Iteration 19500, lr = 0.001
  1316. I0115 10:28:15.445359 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_20000.caffemodel
  1317. I0115 10:28:17.336303 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_20000.solverstate
  1318. I0115 10:28:17.381590 86948 solver.cpp:341] Iteration 20000, Testing net (#0)
  1319. I0115 10:28:39.768867 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1320. I0115 10:28:39.768931 86948 solver.cpp:409] Test net output #1: loss = 12.533 (* 1 = 12.533 loss)
  1321. I0115 10:28:54.931838 86948 solver.cpp:237] Iteration 20000, loss = 0.288482
  1322. I0115 10:28:54.932060 86948 solver.cpp:253] Train net output #0: loss = 0.288482 (* 1 = 0.288482 loss)
  1323. I0115 10:28:55.283051 86948 sgd_solver.cpp:106] Iteration 20000, lr = 0.001
  1324. I0115 10:57:32.207489 86948 solver.cpp:341] Iteration 20500, Testing net (#0)
  1325. I0115 10:57:32.996423 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1326. I0115 10:57:32.996495 86948 solver.cpp:409] Test net output #1: loss = 12.6864 (* 1 = 12.6864 loss)
  1327. I0115 10:57:47.032009 86948 solver.cpp:237] Iteration 20500, loss = 0.232661
  1328. I0115 10:57:47.032083 86948 solver.cpp:253] Train net output #0: loss = 0.232661 (* 1 = 0.232661 loss)
  1329. I0115 10:57:47.383950 86948 sgd_solver.cpp:106] Iteration 20500, lr = 0.001
  1330. I0115 11:25:23.241185 86948 solver.cpp:341] Iteration 21000, Testing net (#0)
  1331. I0115 11:25:24.027587 86948 solver.cpp:409] Test net output #0: accuracy = 0.088
  1332. I0115 11:25:24.027698 86948 solver.cpp:409] Test net output #1: loss = 13.6841 (* 1 = 13.6841 loss)
  1333. I0115 11:25:39.803107 86948 solver.cpp:237] Iteration 21000, loss = 0.163174
  1334. I0115 11:25:39.803169 86948 solver.cpp:253] Train net output #0: loss = 0.163174 (* 1 = 0.163174 loss)
  1335. I0115 11:25:40.157781 86948 sgd_solver.cpp:106] Iteration 21000, lr = 0.001
  1336. I0115 11:54:44.030809 86948 solver.cpp:341] Iteration 21500, Testing net (#0)
  1337. I0115 11:54:44.521996 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1338. I0115 11:54:44.522053 86948 solver.cpp:409] Test net output #1: loss = 12.5438 (* 1 = 12.5438 loss)
  1339. I0115 11:54:57.836741 86948 solver.cpp:237] Iteration 21500, loss = 0.339656
  1340. I0115 11:54:57.836797 86948 solver.cpp:253] Train net output #0: loss = 0.339656 (* 1 = 0.339656 loss)
  1341. I0115 11:54:58.139484 86948 sgd_solver.cpp:106] Iteration 21500, lr = 0.001
  1342. I0115 12:22:13.480896 86948 solver.cpp:341] Iteration 22000, Testing net (#0)
  1343. I0115 12:22:28.147697 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1344. I0115 12:22:28.147763 86948 solver.cpp:409] Test net output #1: loss = 12.3577 (* 1 = 12.3577 loss)
  1345. I0115 12:22:49.409201 86948 solver.cpp:237] Iteration 22000, loss = 0.13327
  1346. I0115 12:22:49.409493 86948 solver.cpp:253] Train net output #0: loss = 0.133271 (* 1 = 0.133271 loss)
  1347. I0115 12:22:49.774583 86948 sgd_solver.cpp:106] Iteration 22000, lr = 0.001
  1348. I0115 12:47:31.931730 86948 solver.cpp:341] Iteration 22500, Testing net (#0)
  1349. I0115 12:47:32.418171 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1350. I0115 12:47:32.418236 86948 solver.cpp:409] Test net output #1: loss = 13.1006 (* 1 = 13.1006 loss)
  1351. I0115 12:47:54.692035 86948 solver.cpp:237] Iteration 22500, loss = 0.171656
  1352. I0115 12:47:54.692101 86948 solver.cpp:253] Train net output #0: loss = 0.171656 (* 1 = 0.171656 loss)
  1353. I0115 12:47:55.034996 86948 sgd_solver.cpp:106] Iteration 22500, lr = 0.001
  1354. I0115 13:15:58.244537 86948 solver.cpp:341] Iteration 23000, Testing net (#0)
  1355. I0115 13:15:59.031538 86948 solver.cpp:409] Test net output #0: accuracy = 0.106
  1356. I0115 13:15:59.031599 86948 solver.cpp:409] Test net output #1: loss = 13.0743 (* 1 = 13.0743 loss)
  1357. I0115 13:16:18.517042 86948 solver.cpp:237] Iteration 23000, loss = 0.137712
  1358. I0115 13:16:18.517102 86948 solver.cpp:253] Train net output #0: loss = 0.137712 (* 1 = 0.137712 loss)
  1359. I0115 13:16:18.517133 86948 sgd_solver.cpp:106] Iteration 23000, lr = 0.001
  1360. I0115 13:45:23.165734 86948 solver.cpp:341] Iteration 23500, Testing net (#0)
  1361. I0115 13:45:23.656873 86948 solver.cpp:409] Test net output #0: accuracy = 0.092
  1362. I0115 13:45:23.656924 86948 solver.cpp:409] Test net output #1: loss = 13.9083 (* 1 = 13.9083 loss)
  1363. I0115 13:45:48.137435 86948 solver.cpp:237] Iteration 23500, loss = 0.265021
  1364. I0115 13:45:48.137493 86948 solver.cpp:253] Train net output #0: loss = 0.265021 (* 1 = 0.265021 loss)
  1365. I0115 13:45:48.466833 86948 sgd_solver.cpp:106] Iteration 23500, lr = 0.001
  1366. I0115 14:15:13.209718 86948 solver.cpp:341] Iteration 24000, Testing net (#0)
  1367. I0115 14:15:27.127425 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1368. I0115 14:15:27.127475 86948 solver.cpp:409] Test net output #1: loss = 13.4618 (* 1 = 13.4618 loss)
  1369. I0115 14:15:52.682905 86948 solver.cpp:237] Iteration 24000, loss = 0.0797336
  1370. I0115 14:15:52.683089 86948 solver.cpp:253] Train net output #0: loss = 0.0797339 (* 1 = 0.0797339 loss)
  1371. I0115 14:15:53.035676 86948 sgd_solver.cpp:106] Iteration 24000, lr = 0.001
  1372. I0115 14:44:07.245959 86948 solver.cpp:341] Iteration 24500, Testing net (#0)
  1373. I0115 14:44:08.031157 86948 solver.cpp:409] Test net output #0: accuracy = 0.13
  1374. I0115 14:44:08.031208 86948 solver.cpp:409] Test net output #1: loss = 13.2707 (* 1 = 13.2707 loss)
  1375. I0115 14:44:20.815029 86948 solver.cpp:237] Iteration 24500, loss = 0.194503
  1376. I0115 14:44:20.815084 86948 solver.cpp:253] Train net output #0: loss = 0.194503 (* 1 = 0.194503 loss)
  1377. I0115 14:44:21.117302 86948 sgd_solver.cpp:106] Iteration 24500, lr = 0.001
  1378. I0115 15:13:06.487923 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_25000.caffemodel
  1379. I0115 15:13:08.724500 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_25000.solverstate
  1380. I0115 15:13:08.772209 86948 solver.cpp:341] Iteration 25000, Testing net (#0)
  1381. I0115 15:13:09.255282 86948 solver.cpp:409] Test net output #0: accuracy = 0.14
  1382. I0115 15:13:09.255313 86948 solver.cpp:409] Test net output #1: loss = 13.9014 (* 1 = 13.9014 loss)
  1383. I0115 15:13:28.586627 86948 solver.cpp:237] Iteration 25000, loss = 0.198213
  1384. I0115 15:13:28.586673 86948 solver.cpp:253] Train net output #0: loss = 0.198213 (* 1 = 0.198213 loss)
  1385. I0115 15:13:28.586694 86948 sgd_solver.cpp:106] Iteration 25000, lr = 0.001
  1386. I0115 15:39:11.728093 86948 solver.cpp:341] Iteration 25500, Testing net (#0)
  1387. I0115 15:39:12.220155 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1388. I0115 15:39:12.220222 86948 solver.cpp:409] Test net output #1: loss = 14.6033 (* 1 = 14.6033 loss)
  1389. I0115 15:39:13.502130 86948 solver.cpp:237] Iteration 25500, loss = 0.250862
  1390. I0115 15:39:13.502204 86948 solver.cpp:253] Train net output #0: loss = 0.250862 (* 1 = 0.250862 loss)
  1391. I0115 15:39:13.805196 86948 sgd_solver.cpp:106] Iteration 25500, lr = 0.001
  1392. I0115 16:04:51.572655 86948 solver.cpp:341] Iteration 26000, Testing net (#0)
  1393. I0115 16:05:10.281999 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1394. I0115 16:05:10.282049 86948 solver.cpp:409] Test net output #1: loss = 15.0046 (* 1 = 15.0046 loss)
  1395. I0115 16:05:32.037425 86948 solver.cpp:237] Iteration 26000, loss = 0.0719195
  1396. I0115 16:05:32.037680 86948 solver.cpp:253] Train net output #0: loss = 0.0719199 (* 1 = 0.0719199 loss)
  1397. I0115 16:05:32.395720 86948 sgd_solver.cpp:106] Iteration 26000, lr = 0.001
  1398. I0115 16:34:12.324597 86948 solver.cpp:341] Iteration 26500, Testing net (#0)
  1399. I0115 16:34:12.813323 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1400. I0115 16:34:12.813388 86948 solver.cpp:409] Test net output #1: loss = 14.6345 (* 1 = 14.6345 loss)
  1401. I0115 16:34:29.782642 86948 solver.cpp:237] Iteration 26500, loss = 0.0966772
  1402. I0115 16:34:29.782699 86948 solver.cpp:253] Train net output #0: loss = 0.0966777 (* 1 = 0.0966777 loss)
  1403. I0115 16:34:30.084247 86948 sgd_solver.cpp:106] Iteration 26500, lr = 0.001
  1404. I0115 17:03:03.270098 86948 solver.cpp:341] Iteration 27000, Testing net (#0)
  1405. I0115 17:03:03.758162 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1406. I0115 17:03:03.758208 86948 solver.cpp:409] Test net output #1: loss = 14.6089 (* 1 = 14.6089 loss)
  1407. I0115 17:03:24.800110 86948 solver.cpp:237] Iteration 27000, loss = 0.11815
  1408. I0115 17:03:24.800163 86948 solver.cpp:253] Train net output #0: loss = 0.118151 (* 1 = 0.118151 loss)
  1409. I0115 17:03:25.180635 86948 sgd_solver.cpp:106] Iteration 27000, lr = 0.001
  1410. I0115 17:30:54.107142 86948 solver.cpp:341] Iteration 27500, Testing net (#0)
  1411. I0115 17:30:54.893143 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1412. I0115 17:30:54.893173 86948 solver.cpp:409] Test net output #1: loss = 15.1023 (* 1 = 15.1023 loss)
  1413. I0115 17:31:15.332089 86948 solver.cpp:237] Iteration 27500, loss = 0.166894
  1414. I0115 17:31:15.332123 86948 solver.cpp:253] Train net output #0: loss = 0.166895 (* 1 = 0.166895 loss)
  1415. I0115 17:31:15.332142 86948 sgd_solver.cpp:106] Iteration 27500, lr = 0.001
  1416. I0115 17:59:33.270324 86948 solver.cpp:341] Iteration 28000, Testing net (#0)
  1417. I0115 17:59:47.750169 86948 solver.cpp:409] Test net output #0: accuracy = 0.098
  1418. I0115 17:59:47.750216 86948 solver.cpp:409] Test net output #1: loss = 15.66 (* 1 = 15.66 loss)
  1419. I0115 18:00:00.982518 86948 solver.cpp:237] Iteration 28000, loss = 0.0917245
  1420. I0115 18:00:00.982571 86948 solver.cpp:253] Train net output #0: loss = 0.0917249 (* 1 = 0.0917249 loss)
  1421. I0115 18:00:00.982594 86948 sgd_solver.cpp:106] Iteration 28000, lr = 0.001
  1422. I0115 18:27:58.396294 86948 solver.cpp:341] Iteration 28500, Testing net (#0)
  1423. I0115 18:27:58.884178 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1424. I0115 18:27:58.884222 86948 solver.cpp:409] Test net output #1: loss = 15.1945 (* 1 = 15.1945 loss)
  1425. I0115 18:28:10.991932 86948 solver.cpp:237] Iteration 28500, loss = 0.233415
  1426. I0115 18:28:10.991986 86948 solver.cpp:253] Train net output #0: loss = 0.233416 (* 1 = 0.233416 loss)
  1427. I0115 18:28:11.294566 86948 sgd_solver.cpp:106] Iteration 28500, lr = 0.001
  1428. I0115 18:53:29.752965 86948 solver.cpp:341] Iteration 29000, Testing net (#0)
  1429. I0115 18:53:30.241336 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1430. I0115 18:53:30.241379 86948 solver.cpp:409] Test net output #1: loss = 15.9175 (* 1 = 15.9175 loss)
  1431. I0115 18:53:45.822639 86948 solver.cpp:237] Iteration 29000, loss = 0.180933
  1432. I0115 18:53:45.822686 86948 solver.cpp:253] Train net output #0: loss = 0.180933 (* 1 = 0.180933 loss)
  1433. I0115 18:53:46.178261 86948 sgd_solver.cpp:106] Iteration 29000, lr = 0.001
  1434. I0115 19:20:57.301084 86948 solver.cpp:341] Iteration 29500, Testing net (#0)
  1435. I0115 19:20:58.086940 86948 solver.cpp:409] Test net output #0: accuracy = 0.108
  1436. I0115 19:20:58.086983 86948 solver.cpp:409] Test net output #1: loss = 15.0374 (* 1 = 15.0374 loss)
  1437. I0115 19:21:15.846012 86948 solver.cpp:237] Iteration 29500, loss = 0.126706
  1438. I0115 19:21:15.846066 86948 solver.cpp:253] Train net output #0: loss = 0.126706 (* 1 = 0.126706 loss)
  1439. I0115 19:21:16.226754 86948 sgd_solver.cpp:106] Iteration 29500, lr = 0.001
  1440. I0115 19:50:20.279408 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_30000.caffemodel
  1441. I0115 19:50:23.981377 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_30000.solverstate
  1442. I0115 19:50:24.027272 86948 solver.cpp:341] Iteration 30000, Testing net (#0)
  1443. I0115 19:50:41.670526 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1444. I0115 19:50:41.670569 86948 solver.cpp:409] Test net output #1: loss = 14.7405 (* 1 = 14.7405 loss)
  1445. I0115 19:51:10.223618 86948 solver.cpp:237] Iteration 30000, loss = 0.115344
  1446. I0115 19:51:10.223824 86948 solver.cpp:253] Train net output #0: loss = 0.115344 (* 1 = 0.115344 loss)
  1447. I0115 19:51:10.554566 86948 sgd_solver.cpp:106] Iteration 30000, lr = 0.001
  1448. I0115 20:19:14.663985 86948 solver.cpp:341] Iteration 30500, Testing net (#0)
  1449. I0115 20:19:15.150887 86948 solver.cpp:409] Test net output #0: accuracy = 0.108
  1450. I0115 20:19:15.150933 86948 solver.cpp:409] Test net output #1: loss = 14.5146 (* 1 = 14.5146 loss)
  1451. I0115 20:19:27.349344 86948 solver.cpp:237] Iteration 30500, loss = 0.176101
  1452. I0115 20:19:27.349388 86948 solver.cpp:253] Train net output #0: loss = 0.176101 (* 1 = 0.176101 loss)
  1453. I0115 20:19:27.715607 86948 sgd_solver.cpp:106] Iteration 30500, lr = 0.001
  1454. I0115 20:48:49.768007 86948 solver.cpp:341] Iteration 31000, Testing net (#0)
  1455. I0115 20:48:50.554081 86948 solver.cpp:409] Test net output #0: accuracy = 0.146
  1456. I0115 20:48:50.554113 86948 solver.cpp:409] Test net output #1: loss = 15.8795 (* 1 = 15.8795 loss)
  1457. I0115 20:49:03.957561 86948 solver.cpp:237] Iteration 31000, loss = 0.152509
  1458. I0115 20:49:03.957617 86948 solver.cpp:253] Train net output #0: loss = 0.15251 (* 1 = 0.15251 loss)
  1459. I0115 20:49:04.340080 86948 sgd_solver.cpp:106] Iteration 31000, lr = 0.001
  1460. I0115 21:17:05.424391 86948 solver.cpp:341] Iteration 31500, Testing net (#0)
  1461. I0115 21:17:06.212334 86948 solver.cpp:409] Test net output #0: accuracy = 0.14
  1462. I0115 21:17:06.212385 86948 solver.cpp:409] Test net output #1: loss = 15.1155 (* 1 = 15.1155 loss)
  1463. I0115 21:17:24.824414 86948 solver.cpp:237] Iteration 31500, loss = 0.106935
  1464. I0115 21:17:24.824450 86948 solver.cpp:253] Train net output #0: loss = 0.106935 (* 1 = 0.106935 loss)
  1465. I0115 21:17:24.824468 86948 sgd_solver.cpp:106] Iteration 31500, lr = 0.001
  1466. I0115 21:42:54.570103 86948 solver.cpp:341] Iteration 32000, Testing net (#0)
  1467. I0115 21:43:11.151145 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1468. I0115 21:43:11.151206 86948 solver.cpp:409] Test net output #1: loss = 15.0276 (* 1 = 15.0276 loss)
  1469. I0115 21:43:37.811069 86948 solver.cpp:237] Iteration 32000, loss = 0.105481
  1470. I0115 21:43:37.811275 86948 solver.cpp:253] Train net output #0: loss = 0.105482 (* 1 = 0.105482 loss)
  1471. I0115 21:43:38.146634 86948 sgd_solver.cpp:106] Iteration 32000, lr = 0.001
  1472. I0115 22:10:46.791887 86948 solver.cpp:341] Iteration 32500, Testing net (#0)
  1473. I0115 22:10:47.576876 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1474. I0115 22:10:47.576911 86948 solver.cpp:409] Test net output #1: loss = 16.8682 (* 1 = 16.8682 loss)
  1475. I0115 22:11:00.348250 86948 solver.cpp:237] Iteration 32500, loss = 0.136433
  1476. I0115 22:11:00.348315 86948 solver.cpp:253] Train net output #0: loss = 0.136434 (* 1 = 0.136434 loss)
  1477. I0115 22:11:00.703428 86948 sgd_solver.cpp:106] Iteration 32500, lr = 0.001
  1478. I0115 22:40:09.482702 86948 solver.cpp:341] Iteration 33000, Testing net (#0)
  1479. I0115 22:40:09.970733 86948 solver.cpp:409] Test net output #0: accuracy = 0.106
  1480. I0115 22:40:09.970774 86948 solver.cpp:409] Test net output #1: loss = 15.7237 (* 1 = 15.7237 loss)
  1481. I0115 22:40:33.196830 86948 solver.cpp:237] Iteration 33000, loss = 0.0419159
  1482. I0115 22:40:33.196878 86948 solver.cpp:253] Train net output #0: loss = 0.0419162 (* 1 = 0.0419162 loss)
  1483. I0115 22:40:33.520575 86948 sgd_solver.cpp:106] Iteration 33000, lr = 0.001
  1484. I0115 23:09:33.160050 86948 solver.cpp:341] Iteration 33500, Testing net (#0)
  1485. I0115 23:09:33.949401 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  1486. I0115 23:09:33.949445 86948 solver.cpp:409] Test net output #1: loss = 16.4421 (* 1 = 16.4421 loss)
  1487. I0115 23:09:55.919976 86948 solver.cpp:237] Iteration 33500, loss = 0.0335577
  1488. I0115 23:09:55.920012 86948 solver.cpp:253] Train net output #0: loss = 0.0335579 (* 1 = 0.0335579 loss)
  1489. I0115 23:09:55.920037 86948 sgd_solver.cpp:106] Iteration 33500, lr = 0.001
  1490. I0115 23:38:55.816653 86948 solver.cpp:341] Iteration 34000, Testing net (#0)
  1491. I0115 23:39:11.819865 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  1492. I0115 23:39:11.819916 86948 solver.cpp:409] Test net output #1: loss = 15.4734 (* 1 = 15.4734 loss)
  1493. I0115 23:39:39.288936 86948 solver.cpp:237] Iteration 34000, loss = 0.0957365
  1494. I0115 23:39:39.289160 86948 solver.cpp:253] Train net output #0: loss = 0.0957366 (* 1 = 0.0957366 loss)
  1495. I0115 23:39:39.647867 86948 sgd_solver.cpp:106] Iteration 34000, lr = 0.001
  1496. I0116 00:07:39.981012 86948 solver.cpp:341] Iteration 34500, Testing net (#0)
  1497. I0116 00:07:40.466253 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1498. I0116 00:07:40.466292 86948 solver.cpp:409] Test net output #1: loss = 16.4261 (* 1 = 16.4261 loss)
  1499. I0116 00:07:53.536689 86948 solver.cpp:237] Iteration 34500, loss = 0.0408948
  1500. I0116 00:07:53.536746 86948 solver.cpp:253] Train net output #0: loss = 0.040895 (* 1 = 0.040895 loss)
  1501. I0116 00:07:53.899355 86948 sgd_solver.cpp:106] Iteration 34500, lr = 0.001
  1502. I0116 00:36:33.319710 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_35000.caffemodel
  1503. I0116 00:36:35.887605 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_35000.solverstate
  1504. I0116 00:36:35.931835 86948 solver.cpp:341] Iteration 35000, Testing net (#0)
  1505. I0116 00:36:36.414018 86948 solver.cpp:409] Test net output #0: accuracy = 0.13
  1506. I0116 00:36:36.414049 86948 solver.cpp:409] Test net output #1: loss = 15.4727 (* 1 = 15.4727 loss)
  1507. I0116 00:36:56.310194 86948 solver.cpp:237] Iteration 35000, loss = 0.043666
  1508. I0116 00:36:56.310250 86948 solver.cpp:253] Train net output #0: loss = 0.0436661 (* 1 = 0.0436661 loss)
  1509. I0116 00:36:56.666497 86948 sgd_solver.cpp:106] Iteration 35000, lr = 0.001
  1510. I0116 01:03:28.900647 86948 solver.cpp:341] Iteration 35500, Testing net (#0)
  1511. I0116 01:03:29.686399 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1512. I0116 01:03:29.686441 86948 solver.cpp:409] Test net output #1: loss = 15.9959 (* 1 = 15.9959 loss)
  1513. I0116 01:03:59.950546 86948 solver.cpp:237] Iteration 35500, loss = 0.0498535
  1514. I0116 01:03:59.950760 86948 solver.cpp:253] Train net output #0: loss = 0.0498535 (* 1 = 0.0498535 loss)
  1515. I0116 01:04:00.292171 86948 sgd_solver.cpp:106] Iteration 35500, lr = 0.001
  1516. I0116 01:30:57.973039 86948 solver.cpp:341] Iteration 36000, Testing net (#0)
  1517. I0116 01:31:12.973151 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1518. I0116 01:31:12.973201 86948 solver.cpp:409] Test net output #1: loss = 16.1145 (* 1 = 16.1145 loss)
  1519. I0116 01:31:27.247730 86948 solver.cpp:237] Iteration 36000, loss = 0.0438099
  1520. I0116 01:31:27.247781 86948 solver.cpp:253] Train net output #0: loss = 0.04381 (* 1 = 0.04381 loss)
  1521. I0116 01:31:27.622767 86948 sgd_solver.cpp:106] Iteration 36000, lr = 0.001
  1522. I0116 02:00:12.971065 86948 solver.cpp:341] Iteration 36500, Testing net (#0)
  1523. I0116 02:00:13.755668 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1524. I0116 02:00:13.755719 86948 solver.cpp:409] Test net output #1: loss = 15.5805 (* 1 = 15.5805 loss)
  1525. I0116 02:00:36.261240 86948 solver.cpp:237] Iteration 36500, loss = 0.132418
  1526. I0116 02:00:36.261309 86948 solver.cpp:253] Train net output #0: loss = 0.132418 (* 1 = 0.132418 loss)
  1527. I0116 02:00:36.563027 86948 sgd_solver.cpp:106] Iteration 36500, lr = 0.001
  1528. I0116 02:29:10.770051 86948 solver.cpp:341] Iteration 37000, Testing net (#0)
  1529. I0116 02:29:11.554711 86948 solver.cpp:409] Test net output #0: accuracy = 0.1
  1530. I0116 02:29:11.554755 86948 solver.cpp:409] Test net output #1: loss = 17.2159 (* 1 = 17.2159 loss)
  1531. I0116 02:29:33.752339 86948 solver.cpp:237] Iteration 37000, loss = 0.049209
  1532. I0116 02:29:33.752393 86948 solver.cpp:253] Train net output #0: loss = 0.049209 (* 1 = 0.049209 loss)
  1533. I0116 02:29:34.055008 86948 sgd_solver.cpp:106] Iteration 37000, lr = 0.001
  1534. I0116 02:59:24.341718 86948 solver.cpp:341] Iteration 37500, Testing net (#0)
  1535. I0116 02:59:24.835041 86948 solver.cpp:409] Test net output #0: accuracy = 0.108
  1536. I0116 02:59:24.835085 86948 solver.cpp:409] Test net output #1: loss = 16.7629 (* 1 = 16.7629 loss)
  1537. I0116 02:59:43.998987 86948 solver.cpp:237] Iteration 37500, loss = 0.0429082
  1538. I0116 02:59:43.999044 86948 solver.cpp:253] Train net output #0: loss = 0.0429082 (* 1 = 0.0429082 loss)
  1539. I0116 02:59:44.358476 86948 sgd_solver.cpp:106] Iteration 37500, lr = 0.001
  1540. I0116 03:25:54.589052 86948 solver.cpp:341] Iteration 38000, Testing net (#0)
  1541. I0116 03:26:07.712508 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1542. I0116 03:26:07.712555 86948 solver.cpp:409] Test net output #1: loss = 16.8652 (* 1 = 16.8652 loss)
  1543. I0116 03:26:25.959569 86948 solver.cpp:237] Iteration 38000, loss = 0.0664696
  1544. I0116 03:26:25.959764 86948 solver.cpp:253] Train net output #0: loss = 0.0664696 (* 1 = 0.0664696 loss)
  1545. I0116 03:26:25.959852 86948 sgd_solver.cpp:106] Iteration 38000, lr = 0.001
  1546. I0116 03:52:32.271384 86948 solver.cpp:341] Iteration 38500, Testing net (#0)
  1547. I0116 03:52:32.757511 86948 solver.cpp:409] Test net output #0: accuracy = 0.102
  1548. I0116 03:52:32.757550 86948 solver.cpp:409] Test net output #1: loss = 16.6012 (* 1 = 16.6012 loss)
  1549. I0116 03:52:45.400449 86948 solver.cpp:237] Iteration 38500, loss = 0.0561813
  1550. I0116 03:52:45.400507 86948 solver.cpp:253] Train net output #0: loss = 0.0561813 (* 1 = 0.0561813 loss)
  1551. I0116 03:52:45.703629 86948 sgd_solver.cpp:106] Iteration 38500, lr = 0.001
  1552. I0116 04:19:38.358794 86948 solver.cpp:341] Iteration 39000, Testing net (#0)
  1553. I0116 04:19:39.143568 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1554. I0116 04:19:39.143609 86948 solver.cpp:409] Test net output #1: loss = 17.4246 (* 1 = 17.4246 loss)
  1555. I0116 04:19:51.897547 86948 solver.cpp:237] Iteration 39000, loss = 0.0750932
  1556. I0116 04:19:51.897608 86948 solver.cpp:253] Train net output #0: loss = 0.0750932 (* 1 = 0.0750932 loss)
  1557. I0116 04:19:52.250996 86948 sgd_solver.cpp:106] Iteration 39000, lr = 0.001
  1558. I0116 04:48:40.934551 86948 solver.cpp:341] Iteration 39500, Testing net (#0)
  1559. I0116 04:48:41.424099 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1560. I0116 04:48:41.424131 86948 solver.cpp:409] Test net output #1: loss = 16.8196 (* 1 = 16.8196 loss)
  1561. I0116 04:49:06.289463 86948 solver.cpp:237] Iteration 39500, loss = 0.0240466
  1562. I0116 04:49:06.289499 86948 solver.cpp:253] Train net output #0: loss = 0.0240466 (* 1 = 0.0240466 loss)
  1563. I0116 04:49:06.289518 86948 sgd_solver.cpp:106] Iteration 39500, lr = 0.001
  1564. I0116 05:18:16.326318 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_40000.caffemodel
  1565. I0116 05:18:19.084363 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_40000.solverstate
  1566. I0116 05:18:19.128511 86948 solver.cpp:341] Iteration 40000, Testing net (#0)
  1567. I0116 05:18:41.247922 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1568. I0116 05:18:41.247972 86948 solver.cpp:409] Test net output #1: loss = 16.2713 (* 1 = 16.2713 loss)
  1569. I0116 05:18:53.945335 86948 solver.cpp:237] Iteration 40000, loss = 0.0314111
  1570. I0116 05:18:53.945523 86948 solver.cpp:253] Train net output #0: loss = 0.0314112 (* 1 = 0.0314112 loss)
  1571. I0116 05:18:54.285737 86948 sgd_solver.cpp:106] Iteration 40000, lr = 0.001
  1572. I0116 05:48:00.650080 86948 solver.cpp:341] Iteration 40500, Testing net (#0)
  1573. I0116 05:48:01.433145 86948 solver.cpp:409] Test net output #0: accuracy = 0.13
  1574. I0116 05:48:01.433185 86948 solver.cpp:409] Test net output #1: loss = 16.1484 (* 1 = 16.1484 loss)
  1575. I0116 05:48:16.489727 86948 solver.cpp:237] Iteration 40500, loss = 0.0346045
  1576. I0116 05:48:16.489781 86948 solver.cpp:253] Train net output #0: loss = 0.0346046 (* 1 = 0.0346046 loss)
  1577. I0116 05:48:16.792301 86948 sgd_solver.cpp:106] Iteration 40500, lr = 0.001
  1578. I0116 06:16:05.988806 86948 solver.cpp:341] Iteration 41000, Testing net (#0)
  1579. I0116 06:16:06.773151 86948 solver.cpp:409] Test net output #0: accuracy = 0.146
  1580. I0116 06:16:06.773193 86948 solver.cpp:409] Test net output #1: loss = 16.7342 (* 1 = 16.7342 loss)
  1581. I0116 06:16:34.219686 86948 solver.cpp:237] Iteration 41000, loss = 0.0419025
  1582. I0116 06:16:34.219740 86948 solver.cpp:253] Train net output #0: loss = 0.0419026 (* 1 = 0.0419026 loss)
  1583. I0116 06:16:34.522406 86948 sgd_solver.cpp:106] Iteration 41000, lr = 0.001
  1584. I0116 06:42:35.883358 86948 solver.cpp:341] Iteration 41500, Testing net (#0)
  1585. I0116 06:42:36.669574 86948 solver.cpp:409] Test net output #0: accuracy = 0.102
  1586. I0116 06:42:36.669616 86948 solver.cpp:409] Test net output #1: loss = 17.7167 (* 1 = 17.7167 loss)
  1587. I0116 06:42:49.773461 86948 solver.cpp:237] Iteration 41500, loss = 0.185499
  1588. I0116 06:42:49.773519 86948 solver.cpp:253] Train net output #0: loss = 0.185499 (* 1 = 0.185499 loss)
  1589. I0116 06:42:50.104356 86948 sgd_solver.cpp:106] Iteration 41500, lr = 0.001
  1590. I0116 07:11:21.097192 86948 solver.cpp:341] Iteration 42000, Testing net (#0)
  1591. I0116 07:11:46.179417 86948 solver.cpp:409] Test net output #0: accuracy = 0.092
  1592. I0116 07:11:46.179472 86948 solver.cpp:409] Test net output #1: loss = 17.2835 (* 1 = 17.2835 loss)
  1593. I0116 07:11:58.591433 86948 solver.cpp:237] Iteration 42000, loss = 0.0348851
  1594. I0116 07:11:58.591639 86948 solver.cpp:253] Train net output #0: loss = 0.0348851 (* 1 = 0.0348851 loss)
  1595. I0116 07:11:58.952178 86948 sgd_solver.cpp:106] Iteration 42000, lr = 0.001
  1596. I0116 07:40:17.186990 86948 solver.cpp:341] Iteration 42500, Testing net (#0)
  1597. I0116 07:40:17.672905 86948 solver.cpp:409] Test net output #0: accuracy = 0.106
  1598. I0116 07:40:17.672950 86948 solver.cpp:409] Test net output #1: loss = 15.8583 (* 1 = 15.8583 loss)
  1599. I0116 07:40:41.425642 86948 solver.cpp:237] Iteration 42500, loss = 0.199413
  1600. I0116 07:40:41.425703 86948 solver.cpp:253] Train net output #0: loss = 0.199413 (* 1 = 0.199413 loss)
  1601. I0116 07:40:41.761833 86948 sgd_solver.cpp:106] Iteration 42500, lr = 0.001
  1602. I0116 08:09:26.900053 86948 solver.cpp:341] Iteration 43000, Testing net (#0)
  1603. I0116 08:09:27.686225 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1604. I0116 08:09:27.686270 86948 solver.cpp:409] Test net output #1: loss = 15.5427 (* 1 = 15.5427 loss)
  1605. I0116 08:09:41.389711 86948 solver.cpp:237] Iteration 43000, loss = 0.0642446
  1606. I0116 08:09:41.389744 86948 solver.cpp:253] Train net output #0: loss = 0.0642446 (* 1 = 0.0642446 loss)
  1607. I0116 08:09:41.389763 86948 sgd_solver.cpp:106] Iteration 43000, lr = 0.001
  1608. I0116 08:38:30.601132 86948 solver.cpp:341] Iteration 43500, Testing net (#0)
  1609. I0116 08:38:31.386473 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1610. I0116 08:38:31.386512 86948 solver.cpp:409] Test net output #1: loss = 17.499 (* 1 = 17.499 loss)
  1611. I0116 08:38:50.902880 86948 solver.cpp:237] Iteration 43500, loss = 0.0141587
  1612. I0116 08:38:50.902937 86948 solver.cpp:253] Train net output #0: loss = 0.0141588 (* 1 = 0.0141588 loss)
  1613. I0116 08:38:51.279814 86948 sgd_solver.cpp:106] Iteration 43500, lr = 0.001
  1614. I0116 09:07:12.407667 86948 solver.cpp:341] Iteration 44000, Testing net (#0)
  1615. I0116 09:07:34.280910 86948 solver.cpp:409] Test net output #0: accuracy = 0.102
  1616. I0116 09:07:34.280952 86948 solver.cpp:409] Test net output #1: loss = 16.8864 (* 1 = 16.8864 loss)
  1617. I0116 09:07:47.130345 86948 solver.cpp:237] Iteration 44000, loss = 0.0879925
  1618. I0116 09:07:47.130524 86948 solver.cpp:253] Train net output #0: loss = 0.0879925 (* 1 = 0.0879925 loss)
  1619. I0116 09:07:47.479935 86948 sgd_solver.cpp:106] Iteration 44000, lr = 0.001
  1620. I0116 09:36:28.176921 86948 solver.cpp:341] Iteration 44500, Testing net (#0)
  1621. I0116 09:36:28.962368 86948 solver.cpp:409] Test net output #0: accuracy = 0.096
  1622. I0116 09:36:28.962402 86948 solver.cpp:409] Test net output #1: loss = 17.2819 (* 1 = 17.2819 loss)
  1623. I0116 09:36:58.008808 86948 solver.cpp:237] Iteration 44500, loss = 0.0444473
  1624. I0116 09:36:58.008863 86948 solver.cpp:253] Train net output #0: loss = 0.0444473 (* 1 = 0.0444473 loss)
  1625. I0116 09:36:58.356963 86948 sgd_solver.cpp:106] Iteration 44500, lr = 0.001
  1626. I0116 10:01:30.690524 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_45000.caffemodel
  1627. I0116 10:01:32.237660 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_45000.solverstate
  1628. I0116 10:01:32.281229 86948 solver.cpp:341] Iteration 45000, Testing net (#0)
  1629. I0116 10:01:32.763458 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1630. I0116 10:01:32.763500 86948 solver.cpp:409] Test net output #1: loss = 17.0017 (* 1 = 17.0017 loss)
  1631. I0116 10:01:57.153758 86948 solver.cpp:237] Iteration 45000, loss = 0.0694832
  1632. I0116 10:01:57.153815 86948 solver.cpp:253] Train net output #0: loss = 0.0694833 (* 1 = 0.0694833 loss)
  1633. I0116 10:01:57.495278 86948 sgd_solver.cpp:106] Iteration 45000, lr = 0.001
  1634. I0116 10:29:44.696324 86948 solver.cpp:341] Iteration 45500, Testing net (#0)
  1635. I0116 10:29:45.185446 86948 solver.cpp:409] Test net output #0: accuracy = 0.096
  1636. I0116 10:29:45.185494 86948 solver.cpp:409] Test net output #1: loss = 17.6054 (* 1 = 17.6054 loss)
  1637. I0116 10:30:04.526257 86948 solver.cpp:237] Iteration 45500, loss = 0.0797003
  1638. I0116 10:30:04.526335 86948 solver.cpp:253] Train net output #0: loss = 0.0797004 (* 1 = 0.0797004 loss)
  1639. I0116 10:30:04.827440 86948 sgd_solver.cpp:106] Iteration 45500, lr = 0.001
  1640. I0116 10:59:38.546458 86948 solver.cpp:341] Iteration 46000, Testing net (#0)
  1641. I0116 10:59:51.304450 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1642. I0116 10:59:51.304498 86948 solver.cpp:409] Test net output #1: loss = 16.352 (* 1 = 16.352 loss)
  1643. I0116 11:00:03.451951 86948 solver.cpp:237] Iteration 46000, loss = 0.032807
  1644. I0116 11:00:03.451997 86948 solver.cpp:253] Train net output #0: loss = 0.0328071 (* 1 = 0.0328071 loss)
  1645. I0116 11:00:03.806876 86948 sgd_solver.cpp:106] Iteration 46000, lr = 0.001
  1646. I0116 11:28:33.824901 86948 solver.cpp:341] Iteration 46500, Testing net (#0)
  1647. I0116 11:28:34.310847 86948 solver.cpp:409] Test net output #0: accuracy = 0.102
  1648. I0116 11:28:34.310897 86948 solver.cpp:409] Test net output #1: loss = 18.0733 (* 1 = 18.0733 loss)
  1649. I0116 11:29:01.760802 86948 solver.cpp:237] Iteration 46500, loss = 0.0403756
  1650. I0116 11:29:01.760859 86948 solver.cpp:253] Train net output #0: loss = 0.0403757 (* 1 = 0.0403757 loss)
  1651. I0116 11:29:02.063665 86948 sgd_solver.cpp:106] Iteration 46500, lr = 0.001
  1652. I0116 11:58:13.606930 86948 solver.cpp:341] Iteration 47000, Testing net (#0)
  1653. I0116 11:58:14.391067 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1654. I0116 11:58:14.391113 86948 solver.cpp:409] Test net output #1: loss = 17.0273 (* 1 = 17.0273 loss)
  1655. I0116 11:58:26.920436 86948 solver.cpp:237] Iteration 47000, loss = 0.010671
  1656. I0116 11:58:26.920490 86948 solver.cpp:253] Train net output #0: loss = 0.0106711 (* 1 = 0.0106711 loss)
  1657. I0116 11:58:27.222714 86948 sgd_solver.cpp:106] Iteration 47000, lr = 0.001
  1658. I0116 12:25:42.220270 86948 solver.cpp:341] Iteration 47500, Testing net (#0)
  1659. I0116 12:25:42.707748 86948 solver.cpp:409] Test net output #0: accuracy = 0.136
  1660. I0116 12:25:42.707782 86948 solver.cpp:409] Test net output #1: loss = 17.2746 (* 1 = 17.2746 loss)
  1661. I0116 12:26:04.379840 86948 solver.cpp:237] Iteration 47500, loss = 0.00723714
  1662. I0116 12:26:04.379889 86948 solver.cpp:253] Train net output #0: loss = 0.00723719 (* 1 = 0.00723719 loss)
  1663. I0116 12:26:04.719759 86948 sgd_solver.cpp:106] Iteration 47500, lr = 0.001
  1664. I0116 12:50:40.256570 86948 solver.cpp:341] Iteration 48000, Testing net (#0)
  1665. I0116 12:51:04.559180 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1666. I0116 12:51:04.559231 86948 solver.cpp:409] Test net output #1: loss = 17.5079 (* 1 = 17.5079 loss)
  1667. I0116 12:51:30.824625 86948 solver.cpp:237] Iteration 48000, loss = 0.011206
  1668. I0116 12:51:30.824817 86948 solver.cpp:253] Train net output #0: loss = 0.011206 (* 1 = 0.011206 loss)
  1669. I0116 12:51:31.175508 86948 sgd_solver.cpp:106] Iteration 48000, lr = 0.001
  1670. I0116 13:18:06.765019 86948 solver.cpp:341] Iteration 48500, Testing net (#0)
  1671. I0116 13:18:07.413496 86948 solver.cpp:409] Test net output #0: accuracy = 0.136
  1672. I0116 13:18:07.413537 86948 solver.cpp:409] Test net output #1: loss = 16.5992 (* 1 = 16.5992 loss)
  1673. I0116 13:18:22.153182 86948 solver.cpp:237] Iteration 48500, loss = 0.0693408
  1674. I0116 13:18:22.153237 86948 solver.cpp:253] Train net output #0: loss = 0.0693408 (* 1 = 0.0693408 loss)
  1675. I0116 13:18:22.454764 86948 sgd_solver.cpp:106] Iteration 48500, lr = 0.001
  1676. I0116 13:46:26.325832 86948 solver.cpp:341] Iteration 49000, Testing net (#0)
  1677. I0116 13:46:26.812815 86948 solver.cpp:409] Test net output #0: accuracy = 0.082
  1678. I0116 13:46:26.812847 86948 solver.cpp:409] Test net output #1: loss = 17.991 (* 1 = 17.991 loss)
  1679. I0116 13:46:47.049461 86948 solver.cpp:237] Iteration 49000, loss = 0.0393182
  1680. I0116 13:46:47.049496 86948 solver.cpp:253] Train net output #0: loss = 0.0393183 (* 1 = 0.0393183 loss)
  1681. I0116 13:46:47.049516 86948 sgd_solver.cpp:106] Iteration 49000, lr = 0.001
  1682. I0116 14:15:12.316920 86948 solver.cpp:341] Iteration 49500, Testing net (#0)
  1683. I0116 14:15:12.803696 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1684. I0116 14:15:12.803738 86948 solver.cpp:409] Test net output #1: loss = 18.0324 (* 1 = 18.0324 loss)
  1685. I0116 14:15:32.959098 86948 solver.cpp:237] Iteration 49500, loss = 0.00838045
  1686. I0116 14:15:32.959175 86948 solver.cpp:253] Train net output #0: loss = 0.0083806 (* 1 = 0.0083806 loss)
  1687. I0116 14:15:32.959216 86948 sgd_solver.cpp:106] Iteration 49500, lr = 0.001
  1688. I0116 14:45:00.736219 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_50000.caffemodel
  1689. I0116 14:45:04.874878 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_50000.solverstate
  1690. I0116 14:45:04.918927 86948 solver.cpp:341] Iteration 50000, Testing net (#0)
  1691. I0116 14:45:24.349766 86948 solver.cpp:409] Test net output #0: accuracy = 0.156
  1692. I0116 14:45:24.349812 86948 solver.cpp:409] Test net output #1: loss = 17.3452 (* 1 = 17.3452 loss)
  1693. I0116 14:45:46.836374 86948 solver.cpp:237] Iteration 50000, loss = 0.0652684
  1694. I0116 14:45:46.836606 86948 solver.cpp:253] Train net output #0: loss = 0.0652686 (* 1 = 0.0652686 loss)
  1695. I0116 14:45:47.176255 86948 sgd_solver.cpp:106] Iteration 50000, lr = 0.001
  1696. I0116 15:14:51.746951 86948 solver.cpp:341] Iteration 50500, Testing net (#0)
  1697. I0116 15:14:52.531561 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1698. I0116 15:14:52.531596 86948 solver.cpp:409] Test net output #1: loss = 18.264 (* 1 = 18.264 loss)
  1699. I0116 15:15:14.273367 86948 solver.cpp:237] Iteration 50500, loss = 0.00904272
  1700. I0116 15:15:14.273427 86948 solver.cpp:253] Train net output #0: loss = 0.00904288 (* 1 = 0.00904288 loss)
  1701. I0116 15:15:14.617744 86948 sgd_solver.cpp:106] Iteration 50500, lr = 0.001
  1702. I0116 15:40:51.166714 86948 solver.cpp:341] Iteration 51000, Testing net (#0)
  1703. I0116 15:40:51.652114 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1704. I0116 15:40:51.652145 86948 solver.cpp:409] Test net output #1: loss = 18.104 (* 1 = 18.104 loss)
  1705. I0116 15:41:06.815246 86948 solver.cpp:237] Iteration 51000, loss = 0.0139255
  1706. I0116 15:41:06.815307 86948 solver.cpp:253] Train net output #0: loss = 0.0139257 (* 1 = 0.0139257 loss)
  1707. I0116 15:41:07.148931 86948 sgd_solver.cpp:106] Iteration 51000, lr = 0.001
  1708. I0116 16:10:14.715281 86948 solver.cpp:341] Iteration 51500, Testing net (#0)
  1709. I0116 16:10:15.497902 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1710. I0116 16:10:15.497944 86948 solver.cpp:409] Test net output #1: loss = 16.8469 (* 1 = 16.8469 loss)
  1711. I0116 16:10:30.471583 86948 solver.cpp:237] Iteration 51500, loss = 0.00585824
  1712. I0116 16:10:30.471635 86948 solver.cpp:253] Train net output #0: loss = 0.0058584 (* 1 = 0.0058584 loss)
  1713. I0116 16:10:30.773869 86948 sgd_solver.cpp:106] Iteration 51500, lr = 0.001
  1714. I0116 16:40:01.208590 86948 solver.cpp:341] Iteration 52000, Testing net (#0)
  1715. I0116 16:40:14.902446 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1716. I0116 16:40:14.902493 86948 solver.cpp:409] Test net output #1: loss = 17.7235 (* 1 = 17.7235 loss)
  1717. I0116 16:40:35.560999 86948 solver.cpp:237] Iteration 52000, loss = 0.00612289
  1718. I0116 16:40:35.561203 86948 solver.cpp:253] Train net output #0: loss = 0.00612306 (* 1 = 0.00612306 loss)
  1719. I0116 16:40:35.912436 86948 sgd_solver.cpp:106] Iteration 52000, lr = 0.001
  1720. I0116 17:08:55.274165 86948 solver.cpp:341] Iteration 52500, Testing net (#0)
  1721. I0116 17:08:55.761176 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1722. I0116 17:08:55.761209 86948 solver.cpp:409] Test net output #1: loss = 17.5617 (* 1 = 17.5617 loss)
  1723. I0116 17:09:21.139521 86948 solver.cpp:237] Iteration 52500, loss = 0.014499
  1724. I0116 17:09:21.139580 86948 solver.cpp:253] Train net output #0: loss = 0.0144992 (* 1 = 0.0144992 loss)
  1725. I0116 17:09:21.471717 86948 sgd_solver.cpp:106] Iteration 52500, lr = 0.001
  1726. I0116 17:39:18.728796 86948 solver.cpp:341] Iteration 53000, Testing net (#0)
  1727. I0116 17:39:19.215486 86948 solver.cpp:409] Test net output #0: accuracy = 0.14
  1728. I0116 17:39:19.215517 86948 solver.cpp:409] Test net output #1: loss = 17.877 (* 1 = 17.877 loss)
  1729. I0116 17:39:31.311669 86948 solver.cpp:237] Iteration 53000, loss = 0.00112286
  1730. I0116 17:39:31.311717 86948 solver.cpp:253] Train net output #0: loss = 0.00112301 (* 1 = 0.00112301 loss)
  1731. I0116 17:39:31.657688 86948 sgd_solver.cpp:106] Iteration 53000, lr = 0.001
  1732. I0116 18:08:52.516023 86948 solver.cpp:341] Iteration 53500, Testing net (#0)
  1733. I0116 18:08:53.300496 86948 solver.cpp:409] Test net output #0: accuracy = 0.146
  1734. I0116 18:08:53.300536 86948 solver.cpp:409] Test net output #1: loss = 17.146 (* 1 = 17.146 loss)
  1735. I0116 18:09:07.123847 86948 solver.cpp:237] Iteration 53500, loss = 0.0122715
  1736. I0116 18:09:07.123903 86948 solver.cpp:253] Train net output #0: loss = 0.0122717 (* 1 = 0.0122717 loss)
  1737. I0116 18:09:07.487051 86948 sgd_solver.cpp:106] Iteration 53500, lr = 0.001
  1738. I0116 18:38:23.766234 86948 solver.cpp:341] Iteration 54000, Testing net (#0)
  1739. I0116 18:38:42.114236 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  1740. I0116 18:38:42.114279 86948 solver.cpp:409] Test net output #1: loss = 17.6793 (* 1 = 17.6793 loss)
  1741. I0116 18:38:55.071338 86948 solver.cpp:237] Iteration 54000, loss = 0.00152475
  1742. I0116 18:38:55.071554 86948 solver.cpp:253] Train net output #0: loss = 0.00152491 (* 1 = 0.00152491 loss)
  1743. I0116 18:38:55.444756 86948 sgd_solver.cpp:106] Iteration 54000, lr = 0.001
  1744. I0116 19:04:11.690522 86948 solver.cpp:341] Iteration 54500, Testing net (#0)
  1745. I0116 19:04:12.474172 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1746. I0116 19:04:12.474216 86948 solver.cpp:409] Test net output #1: loss = 18.4609 (* 1 = 18.4609 loss)
  1747. I0116 19:04:26.059206 86948 solver.cpp:237] Iteration 54500, loss = 0.00278085
  1748. I0116 19:04:26.059262 86948 solver.cpp:253] Train net output #0: loss = 0.00278101 (* 1 = 0.00278101 loss)
  1749. I0116 19:04:26.362669 86948 sgd_solver.cpp:106] Iteration 54500, lr = 0.001
  1750. I0116 19:32:57.336834 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_55000.caffemodel
  1751. I0116 19:33:02.048396 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_55000.solverstate
  1752. I0116 19:33:02.093430 86948 solver.cpp:341] Iteration 55000, Testing net (#0)
  1753. I0116 19:33:02.575043 86948 solver.cpp:409] Test net output #0: accuracy = 0.1
  1754. I0116 19:33:02.575074 86948 solver.cpp:409] Test net output #1: loss = 17.5898 (* 1 = 17.5898 loss)
  1755. I0116 19:33:14.721052 86948 solver.cpp:237] Iteration 55000, loss = 0.00244798
  1756. I0116 19:33:14.721104 86948 solver.cpp:253] Train net output #0: loss = 0.00244815 (* 1 = 0.00244815 loss)
  1757. I0116 19:33:15.023622 86948 sgd_solver.cpp:106] Iteration 55000, lr = 0.001
  1758. I0116 20:00:52.280625 86948 solver.cpp:341] Iteration 55500, Testing net (#0)
  1759. I0116 20:00:53.064491 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1760. I0116 20:00:53.064527 86948 solver.cpp:409] Test net output #1: loss = 18.1016 (* 1 = 18.1016 loss)
  1761. I0116 20:01:10.699457 86948 solver.cpp:237] Iteration 55500, loss = 0.00185457
  1762. I0116 20:01:10.699515 86948 solver.cpp:253] Train net output #0: loss = 0.00185475 (* 1 = 0.00185475 loss)
  1763. I0116 20:01:11.001893 86948 sgd_solver.cpp:106] Iteration 55500, lr = 0.001
  1764. I0116 20:30:32.961246 86948 solver.cpp:341] Iteration 56000, Testing net (#0)
  1765. I0116 20:30:46.017097 86948 solver.cpp:409] Test net output #0: accuracy = 0.138
  1766. I0116 20:30:46.017145 86948 solver.cpp:409] Test net output #1: loss = 17.466 (* 1 = 17.466 loss)
  1767. I0116 20:31:08.532201 86948 solver.cpp:237] Iteration 56000, loss = 0.00167043
  1768. I0116 20:31:08.532402 86948 solver.cpp:253] Train net output #0: loss = 0.00167062 (* 1 = 0.00167062 loss)
  1769. I0116 20:31:08.892717 86948 sgd_solver.cpp:106] Iteration 56000, lr = 0.001
  1770. I0116 21:00:52.884006 86948 solver.cpp:341] Iteration 56500, Testing net (#0)
  1771. I0116 21:00:53.370746 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1772. I0116 21:00:53.370777 86948 solver.cpp:409] Test net output #1: loss = 18.5152 (* 1 = 18.5152 loss)
  1773. I0116 21:01:29.835618 86948 solver.cpp:237] Iteration 56500, loss = 0.00127041
  1774. I0116 21:01:29.835860 86948 solver.cpp:253] Train net output #0: loss = 0.0012706 (* 1 = 0.0012706 loss)
  1775. I0116 21:01:30.164361 86948 sgd_solver.cpp:106] Iteration 56500, lr = 0.001
  1776. I0116 21:29:50.693728 86948 solver.cpp:341] Iteration 57000, Testing net (#0)
  1777. I0116 21:29:51.477064 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1778. I0116 21:29:51.477094 86948 solver.cpp:409] Test net output #1: loss = 17.2161 (* 1 = 17.2161 loss)
  1779. I0116 21:30:16.758760 86948 solver.cpp:237] Iteration 57000, loss = 0.00125112
  1780. I0116 21:30:16.758816 86948 solver.cpp:253] Train net output #0: loss = 0.00125128 (* 1 = 0.00125128 loss)
  1781. I0116 21:30:17.094545 86948 sgd_solver.cpp:106] Iteration 57000, lr = 0.001
  1782. I0116 21:56:25.575891 86948 solver.cpp:341] Iteration 57500, Testing net (#0)
  1783. I0116 21:56:26.359499 86948 solver.cpp:409] Test net output #0: accuracy = 0.102
  1784. I0116 21:56:26.359541 86948 solver.cpp:409] Test net output #1: loss = 17.1281 (* 1 = 17.1281 loss)
  1785. I0116 21:56:45.616204 86948 solver.cpp:237] Iteration 57500, loss = 0.000995827
  1786. I0116 21:56:45.616246 86948 solver.cpp:253] Train net output #0: loss = 0.000995983 (* 1 = 0.000995983 loss)
  1787. I0116 21:56:45.616267 86948 sgd_solver.cpp:106] Iteration 57500, lr = 0.001
  1788. I0116 22:24:15.822245 86948 solver.cpp:341] Iteration 58000, Testing net (#0)
  1789. I0116 22:24:28.277140 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1790. I0116 22:24:28.277187 86948 solver.cpp:409] Test net output #1: loss = 17.7945 (* 1 = 17.7945 loss)
  1791. I0116 22:24:46.763841 86948 solver.cpp:237] Iteration 58000, loss = 0.00149116
  1792. I0116 22:24:46.764065 86948 solver.cpp:253] Train net output #0: loss = 0.00149132 (* 1 = 0.00149132 loss)
  1793. I0116 22:24:47.134806 86948 sgd_solver.cpp:106] Iteration 58000, lr = 0.001
  1794. I0116 22:53:01.435055 86948 solver.cpp:341] Iteration 58500, Testing net (#0)
  1795. I0116 22:53:02.219271 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1796. I0116 22:53:02.219316 86948 solver.cpp:409] Test net output #1: loss = 17.5235 (* 1 = 17.5235 loss)
  1797. I0116 22:53:21.272271 86948 solver.cpp:237] Iteration 58500, loss = 0.000840927
  1798. I0116 22:53:21.272325 86948 solver.cpp:253] Train net output #0: loss = 0.000841088 (* 1 = 0.000841088 loss)
  1799. I0116 22:53:21.631062 86948 sgd_solver.cpp:106] Iteration 58500, lr = 0.001
  1800. I0116 23:22:36.109424 86948 solver.cpp:341] Iteration 59000, Testing net (#0)
  1801. I0116 23:22:36.595427 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1802. I0116 23:22:36.595466 86948 solver.cpp:409] Test net output #1: loss = 17.8167 (* 1 = 17.8167 loss)
  1803. I0116 23:22:58.007833 86948 solver.cpp:237] Iteration 59000, loss = 0.000619527
  1804. I0116 23:22:58.007884 86948 solver.cpp:253] Train net output #0: loss = 0.000619694 (* 1 = 0.000619694 loss)
  1805. I0116 23:22:58.388481 86948 sgd_solver.cpp:106] Iteration 59000, lr = 0.001
  1806. I0116 23:51:47.412817 86948 solver.cpp:341] Iteration 59500, Testing net (#0)
  1807. I0116 23:51:47.905879 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1808. I0116 23:51:47.905912 86948 solver.cpp:409] Test net output #1: loss = 17.7125 (* 1 = 17.7125 loss)
  1809. I0116 23:52:21.172386 86948 solver.cpp:237] Iteration 59500, loss = 0.000833786
  1810. I0116 23:52:21.172557 86948 solver.cpp:253] Train net output #0: loss = 0.000833953 (* 1 = 0.000833953 loss)
  1811. I0116 23:52:21.172621 86948 sgd_solver.cpp:106] Iteration 59500, lr = 0.001
  1812. I0117 00:19:51.501459 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_60000.caffemodel
  1813. I0117 00:19:53.112973 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_60000.solverstate
  1814. I0117 00:19:53.159576 86948 solver.cpp:341] Iteration 60000, Testing net (#0)
  1815. I0117 00:20:13.622352 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1816. I0117 00:20:13.622414 86948 solver.cpp:409] Test net output #1: loss = 16.4636 (* 1 = 16.4636 loss)
  1817. I0117 00:20:38.179111 86948 solver.cpp:237] Iteration 60000, loss = 0.000698102
  1818. I0117 00:20:38.179344 86948 solver.cpp:253] Train net output #0: loss = 0.000698267 (* 1 = 0.000698267 loss)
  1819. I0117 00:20:38.518753 86948 sgd_solver.cpp:106] Iteration 60000, lr = 0.001
  1820. I0117 00:48:01.094512 86948 solver.cpp:341] Iteration 60500, Testing net (#0)
  1821. I0117 00:48:01.583814 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  1822. I0117 00:48:01.583847 86948 solver.cpp:409] Test net output #1: loss = 16.9208 (* 1 = 16.9208 loss)
  1823. I0117 00:48:02.826798 86948 solver.cpp:237] Iteration 60500, loss = 0.00134723
  1824. I0117 00:48:02.826843 86948 solver.cpp:253] Train net output #0: loss = 0.00134739 (* 1 = 0.00134739 loss)
  1825. I0117 00:48:03.129700 86948 sgd_solver.cpp:106] Iteration 60500, lr = 0.001
  1826. I0117 01:15:35.234369 86948 solver.cpp:341] Iteration 61000, Testing net (#0)
  1827. I0117 01:15:36.017329 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1828. I0117 01:15:36.017371 86948 solver.cpp:409] Test net output #1: loss = 16.7038 (* 1 = 16.7038 loss)
  1829. I0117 01:15:48.144557 86948 solver.cpp:237] Iteration 61000, loss = 0.00100287
  1830. I0117 01:15:48.144616 86948 solver.cpp:253] Train net output #0: loss = 0.00100303 (* 1 = 0.00100303 loss)
  1831. I0117 01:15:48.447521 86948 sgd_solver.cpp:106] Iteration 61000, lr = 0.001
  1832. I0117 01:43:27.736707 86948 solver.cpp:341] Iteration 61500, Testing net (#0)
  1833. I0117 01:43:28.520886 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1834. I0117 01:43:28.520928 86948 solver.cpp:409] Test net output #1: loss = 16.7259 (* 1 = 16.7259 loss)
  1835. I0117 01:43:43.177824 86948 solver.cpp:237] Iteration 61500, loss = 0.00111827
  1836. I0117 01:43:43.177875 86948 solver.cpp:253] Train net output #0: loss = 0.00111844 (* 1 = 0.00111844 loss)
  1837. I0117 01:43:43.177896 86948 sgd_solver.cpp:106] Iteration 61500, lr = 0.001
  1838. I0117 02:12:18.621260 86948 solver.cpp:341] Iteration 62000, Testing net (#0)
  1839. I0117 02:12:39.971473 86948 solver.cpp:409] Test net output #0: accuracy = 0.142
  1840. I0117 02:12:39.971523 86948 solver.cpp:409] Test net output #1: loss = 15.8327 (* 1 = 15.8327 loss)
  1841. I0117 02:12:57.808661 86948 solver.cpp:237] Iteration 62000, loss = 0.000379236
  1842. I0117 02:12:57.808817 86948 solver.cpp:253] Train net output #0: loss = 0.000379402 (* 1 = 0.000379402 loss)
  1843. I0117 02:12:58.166721 86948 sgd_solver.cpp:106] Iteration 62000, lr = 0.001
  1844. I0117 02:41:14.882915 86948 solver.cpp:341] Iteration 62500, Testing net (#0)
  1845. I0117 02:41:15.368656 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1846. I0117 02:41:15.368685 86948 solver.cpp:409] Test net output #1: loss = 16.6882 (* 1 = 16.6882 loss)
  1847. I0117 02:41:34.613448 86948 solver.cpp:237] Iteration 62500, loss = 0.00201296
  1848. I0117 02:41:34.613498 86948 solver.cpp:253] Train net output #0: loss = 0.00201312 (* 1 = 0.00201312 loss)
  1849. I0117 02:41:34.967422 86948 sgd_solver.cpp:106] Iteration 62500, lr = 0.001
  1850. I0117 03:09:53.952474 86948 solver.cpp:341] Iteration 63000, Testing net (#0)
  1851. I0117 03:09:54.738646 86948 solver.cpp:409] Test net output #0: accuracy = 0.118
  1852. I0117 03:09:54.738690 86948 solver.cpp:409] Test net output #1: loss = 15.5435 (* 1 = 15.5435 loss)
  1853. I0117 03:10:19.191061 86948 solver.cpp:237] Iteration 63000, loss = 0.00052981
  1854. I0117 03:10:19.191112 86948 solver.cpp:253] Train net output #0: loss = 0.000529975 (* 1 = 0.000529975 loss)
  1855. I0117 03:10:19.493536 86948 sgd_solver.cpp:106] Iteration 63000, lr = 0.001
  1856. I0117 03:38:18.031046 86948 solver.cpp:341] Iteration 63500, Testing net (#0)
  1857. I0117 03:38:18.815105 86948 solver.cpp:409] Test net output #0: accuracy = 0.154
  1858. I0117 03:38:18.815142 86948 solver.cpp:409] Test net output #1: loss = 16.497 (* 1 = 16.497 loss)
  1859. I0117 03:38:34.141821 86948 solver.cpp:237] Iteration 63500, loss = 0.000767982
  1860. I0117 03:38:34.141876 86948 solver.cpp:253] Train net output #0: loss = 0.000768148 (* 1 = 0.000768148 loss)
  1861. I0117 03:38:34.443259 86948 sgd_solver.cpp:106] Iteration 63500, lr = 0.001
  1862. I0117 04:03:25.591889 86948 solver.cpp:341] Iteration 64000, Testing net (#0)
  1863. I0117 04:03:48.695626 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1864. I0117 04:03:48.695669 86948 solver.cpp:409] Test net output #1: loss = 16.1517 (* 1 = 16.1517 loss)
  1865. I0117 04:04:01.971127 86948 solver.cpp:237] Iteration 64000, loss = 0.000629346
  1866. I0117 04:04:01.971350 86948 solver.cpp:253] Train net output #0: loss = 0.000629513 (* 1 = 0.000629513 loss)
  1867. I0117 04:04:02.324965 86948 sgd_solver.cpp:106] Iteration 64000, lr = 0.001
  1868. I0117 04:32:27.745319 86948 solver.cpp:341] Iteration 64500, Testing net (#0)
  1869. I0117 04:32:28.231277 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1870. I0117 04:32:28.231320 86948 solver.cpp:409] Test net output #1: loss = 16.2351 (* 1 = 16.2351 loss)
  1871. I0117 04:32:54.835930 86948 solver.cpp:237] Iteration 64500, loss = 0.00131145
  1872. I0117 04:32:54.835990 86948 solver.cpp:253] Train net output #0: loss = 0.00131162 (* 1 = 0.00131162 loss)
  1873. I0117 04:32:55.187325 86948 sgd_solver.cpp:106] Iteration 64500, lr = 0.001
  1874. I0117 05:00:55.586280 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_65000.caffemodel
  1875. I0117 05:01:00.939674 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_65000.solverstate
  1876. I0117 05:01:02.193336 86948 solver.cpp:341] Iteration 65000, Testing net (#0)
  1877. I0117 05:01:02.676085 86948 solver.cpp:409] Test net output #0: accuracy = 0.134
  1878. I0117 05:01:02.676128 86948 solver.cpp:409] Test net output #1: loss = 15.8553 (* 1 = 15.8553 loss)
  1879. I0117 05:01:26.228950 86948 solver.cpp:237] Iteration 65000, loss = 0.000773424
  1880. I0117 05:01:26.229154 86948 solver.cpp:253] Train net output #0: loss = 0.000773591 (* 1 = 0.000773591 loss)
  1881. I0117 05:01:26.229217 86948 sgd_solver.cpp:106] Iteration 65000, lr = 0.001
  1882. I0117 05:29:58.098109 86948 solver.cpp:341] Iteration 65500, Testing net (#0)
  1883. I0117 05:29:58.882112 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1884. I0117 05:29:58.882163 86948 solver.cpp:409] Test net output #1: loss = 16.5149 (* 1 = 16.5149 loss)
  1885. I0117 05:30:31.572700 86948 solver.cpp:237] Iteration 65500, loss = 0.000899338
  1886. I0117 05:30:31.572927 86948 solver.cpp:253] Train net output #0: loss = 0.000899504 (* 1 = 0.000899504 loss)
  1887. I0117 05:30:31.903906 86948 sgd_solver.cpp:106] Iteration 65500, lr = 0.001
  1888. I0117 06:00:43.981590 86948 solver.cpp:341] Iteration 66000, Testing net (#0)
  1889. I0117 06:01:08.399309 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  1890. I0117 06:01:08.399371 86948 solver.cpp:409] Test net output #1: loss = 15.3885 (* 1 = 15.3885 loss)
  1891. I0117 06:01:44.210752 86948 solver.cpp:237] Iteration 66000, loss = 0.00092309
  1892. I0117 06:01:44.211006 86948 solver.cpp:253] Train net output #0: loss = 0.000923256 (* 1 = 0.000923256 loss)
  1893. I0117 06:01:44.571835 86948 sgd_solver.cpp:106] Iteration 66000, lr = 0.001
  1894. I0117 06:30:41.497572 86948 solver.cpp:341] Iteration 66500, Testing net (#0)
  1895. I0117 06:30:42.280725 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  1896. I0117 06:30:42.280772 86948 solver.cpp:409] Test net output #1: loss = 14.6409 (* 1 = 14.6409 loss)
  1897. I0117 06:30:58.502143 86948 solver.cpp:237] Iteration 66500, loss = 0.000562749
  1898. I0117 06:30:58.502197 86948 solver.cpp:253] Train net output #0: loss = 0.000562914 (* 1 = 0.000562914 loss)
  1899. I0117 06:30:58.870307 86948 sgd_solver.cpp:106] Iteration 66500, lr = 0.001
  1900. I0117 06:56:45.541666 86948 solver.cpp:341] Iteration 67000, Testing net (#0)
  1901. I0117 06:56:46.032340 86948 solver.cpp:409] Test net output #0: accuracy = 0.11
  1902. I0117 06:56:46.032380 86948 solver.cpp:409] Test net output #1: loss = 16.1808 (* 1 = 16.1808 loss)
  1903. I0117 06:57:01.526554 86948 solver.cpp:237] Iteration 67000, loss = 0.00102099
  1904. I0117 06:57:01.526605 86948 solver.cpp:253] Train net output #0: loss = 0.00102115 (* 1 = 0.00102115 loss)
  1905. I0117 06:57:01.865414 86948 sgd_solver.cpp:106] Iteration 67000, lr = 0.001
  1906. I0117 07:24:53.334848 86948 solver.cpp:341] Iteration 67500, Testing net (#0)
  1907. I0117 07:24:54.120802 86948 solver.cpp:409] Test net output #0: accuracy = 0.112
  1908. I0117 07:24:54.120841 86948 solver.cpp:409] Test net output #1: loss = 16.0582 (* 1 = 16.0582 loss)
  1909. I0117 07:25:07.953392 86948 solver.cpp:237] Iteration 67500, loss = 0.00112115
  1910. I0117 07:25:07.953441 86948 solver.cpp:253] Train net output #0: loss = 0.00112132 (* 1 = 0.00112132 loss)
  1911. I0117 07:25:08.307104 86948 sgd_solver.cpp:106] Iteration 67500, lr = 0.001
  1912. I0117 07:53:18.890822 86948 solver.cpp:341] Iteration 68000, Testing net (#0)
  1913. I0117 07:53:32.259802 86948 solver.cpp:409] Test net output #0: accuracy = 0.148
  1914. I0117 07:53:32.259845 86948 solver.cpp:409] Test net output #1: loss = 15.5333 (* 1 = 15.5333 loss)
  1915. I0117 07:53:47.024539 86948 solver.cpp:237] Iteration 68000, loss = 0.000853823
  1916. I0117 07:53:47.024593 86948 solver.cpp:253] Train net output #0: loss = 0.000853988 (* 1 = 0.000853988 loss)
  1917. I0117 07:53:47.327491 86948 sgd_solver.cpp:106] Iteration 68000, lr = 0.001
  1918. I0117 08:22:41.051843 86948 solver.cpp:341] Iteration 68500, Testing net (#0)
  1919. I0117 08:22:41.542336 86948 solver.cpp:409] Test net output #0: accuracy = 0.126
  1920. I0117 08:22:41.542385 86948 solver.cpp:409] Test net output #1: loss = 14.4018 (* 1 = 14.4018 loss)
  1921. I0117 08:22:57.310125 86948 solver.cpp:237] Iteration 68500, loss = 0.0009701
  1922. I0117 08:22:57.310174 86948 solver.cpp:253] Train net output #0: loss = 0.000970266 (* 1 = 0.000970266 loss)
  1923. I0117 08:22:57.310201 86948 sgd_solver.cpp:106] Iteration 68500, lr = 0.001
  1924. I0117 08:51:31.132508 86948 solver.cpp:341] Iteration 69000, Testing net (#0)
  1925. I0117 08:51:31.918012 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1926. I0117 08:51:31.918045 86948 solver.cpp:409] Test net output #1: loss = 15.011 (* 1 = 15.011 loss)
  1927. I0117 08:51:55.438107 86948 solver.cpp:237] Iteration 69000, loss = 0.00144422
  1928. I0117 08:51:55.438156 86948 solver.cpp:253] Train net output #0: loss = 0.00144438 (* 1 = 0.00144438 loss)
  1929. I0117 08:51:55.438179 86948 sgd_solver.cpp:106] Iteration 69000, lr = 0.001
  1930. I0117 09:20:06.035789 86948 solver.cpp:341] Iteration 69500, Testing net (#0)
  1931. I0117 09:20:06.530084 86948 solver.cpp:409] Test net output #0: accuracy = 0.144
  1932. I0117 09:20:06.530124 86948 solver.cpp:409] Test net output #1: loss = 15.2529 (* 1 = 15.2529 loss)
  1933. I0117 09:20:24.312736 86948 solver.cpp:237] Iteration 69500, loss = 0.00116137
  1934. I0117 09:20:24.312789 86948 solver.cpp:253] Train net output #0: loss = 0.00116154 (* 1 = 0.00116154 loss)
  1935. I0117 09:20:24.615401 86948 sgd_solver.cpp:106] Iteration 69500, lr = 0.001
  1936. I0117 09:48:31.437814 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_70000.caffemodel
  1937. I0117 09:48:33.344377 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_70000.solverstate
  1938. I0117 09:48:33.388301 86948 solver.cpp:341] Iteration 70000, Testing net (#0)
  1939. I0117 09:48:46.943537 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  1940. I0117 09:48:46.943588 86948 solver.cpp:409] Test net output #1: loss = 14.5244 (* 1 = 14.5244 loss)
  1941. I0117 09:49:03.112295 86948 solver.cpp:237] Iteration 70000, loss = 0.00122999
  1942. I0117 09:49:03.112499 86948 solver.cpp:253] Train net output #0: loss = 0.00123015 (* 1 = 0.00123015 loss)
  1943. I0117 09:49:03.471725 86948 sgd_solver.cpp:106] Iteration 70000, lr = 0.001
  1944. I0117 10:16:26.051054 86948 solver.cpp:341] Iteration 70500, Testing net (#0)
  1945. I0117 10:16:26.835522 86948 solver.cpp:409] Test net output #0: accuracy = 0.13
  1946. I0117 10:16:26.835569 86948 solver.cpp:409] Test net output #1: loss = 15.0903 (* 1 = 15.0903 loss)
  1947. I0117 10:16:43.701015 86948 solver.cpp:237] Iteration 70500, loss = 0.00145832
  1948. I0117 10:16:43.701076 86948 solver.cpp:253] Train net output #0: loss = 0.00145848 (* 1 = 0.00145848 loss)
  1949. I0117 10:16:43.701113 86948 sgd_solver.cpp:106] Iteration 70500, lr = 0.001
  1950. I0117 10:44:12.330981 86948 solver.cpp:341] Iteration 71000, Testing net (#0)
  1951. I0117 10:44:13.115999 86948 solver.cpp:409] Test net output #0: accuracy = 0.148
  1952. I0117 10:44:13.116049 86948 solver.cpp:409] Test net output #1: loss = 14.8792 (* 1 = 14.8792 loss)
  1953. I0117 10:44:41.000583 86948 solver.cpp:237] Iteration 71000, loss = 0.00174855
  1954. I0117 10:44:41.000628 86948 solver.cpp:253] Train net output #0: loss = 0.00174872 (* 1 = 0.00174872 loss)
  1955. I0117 10:44:41.000649 86948 sgd_solver.cpp:106] Iteration 71000, lr = 0.001
  1956. I0117 11:14:19.942044 86948 solver.cpp:341] Iteration 71500, Testing net (#0)
  1957. I0117 11:14:20.427345 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1958. I0117 11:14:20.427377 86948 solver.cpp:409] Test net output #1: loss = 14.2639 (* 1 = 14.2639 loss)
  1959. I0117 11:14:37.532204 86948 solver.cpp:237] Iteration 71500, loss = 0.000676939
  1960. I0117 11:14:37.532251 86948 solver.cpp:253] Train net output #0: loss = 0.000677105 (* 1 = 0.000677105 loss)
  1961. I0117 11:14:37.879487 86948 sgd_solver.cpp:106] Iteration 71500, lr = 0.001
  1962. I0117 11:43:06.430202 86948 solver.cpp:341] Iteration 72000, Testing net (#0)
  1963. I0117 11:43:22.537875 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  1964. I0117 11:43:22.537925 86948 solver.cpp:409] Test net output #1: loss = 15.7033 (* 1 = 15.7033 loss)
  1965. I0117 11:43:34.772270 86948 solver.cpp:237] Iteration 72000, loss = 0.00116305
  1966. I0117 11:43:34.772325 86948 solver.cpp:253] Train net output #0: loss = 0.00116322 (* 1 = 0.00116322 loss)
  1967. I0117 11:43:35.127001 86948 sgd_solver.cpp:106] Iteration 72000, lr = 0.001
  1968. I0117 12:13:05.521544 86948 solver.cpp:341] Iteration 72500, Testing net (#0)
  1969. I0117 12:13:06.007657 86948 solver.cpp:409] Test net output #0: accuracy = 0.114
  1970. I0117 12:13:06.007705 86948 solver.cpp:409] Test net output #1: loss = 14.6482 (* 1 = 14.6482 loss)
  1971. I0117 12:13:19.264045 86948 solver.cpp:237] Iteration 72500, loss = 0.00117693
  1972. I0117 12:13:19.264101 86948 solver.cpp:253] Train net output #0: loss = 0.0011771 (* 1 = 0.0011771 loss)
  1973. I0117 12:13:19.566165 86948 sgd_solver.cpp:106] Iteration 72500, lr = 0.001
  1974. I0117 12:40:20.751565 86948 solver.cpp:341] Iteration 73000, Testing net (#0)
  1975. I0117 12:40:21.238452 86948 solver.cpp:409] Test net output #0: accuracy = 0.14
  1976. I0117 12:40:21.238520 86948 solver.cpp:409] Test net output #1: loss = 13.8224 (* 1 = 13.8224 loss)
  1977. I0117 12:40:44.118749 86948 solver.cpp:237] Iteration 73000, loss = 0.00160221
  1978. I0117 12:40:44.118808 86948 solver.cpp:253] Train net output #0: loss = 0.00160238 (* 1 = 0.00160238 loss)
  1979. I0117 12:40:44.492357 86948 sgd_solver.cpp:106] Iteration 73000, lr = 0.001
  1980. I0117 13:05:49.983078 86948 solver.cpp:341] Iteration 73500, Testing net (#0)
  1981. I0117 13:05:50.769421 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  1982. I0117 13:05:50.769461 86948 solver.cpp:409] Test net output #1: loss = 14.2741 (* 1 = 14.2741 loss)
  1983. I0117 13:06:13.503284 86948 solver.cpp:237] Iteration 73500, loss = 0.00136268
  1984. I0117 13:06:13.503340 86948 solver.cpp:253] Train net output #0: loss = 0.00136284 (* 1 = 0.00136284 loss)
  1985. I0117 13:06:13.805341 86948 sgd_solver.cpp:106] Iteration 73500, lr = 0.001
  1986. I0117 13:34:08.023531 86948 solver.cpp:341] Iteration 74000, Testing net (#0)
  1987. I0117 13:34:27.022164 86948 solver.cpp:409] Test net output #0: accuracy = 0.104
  1988. I0117 13:34:27.022241 86948 solver.cpp:409] Test net output #1: loss = 14.4191 (* 1 = 14.4191 loss)
  1989. I0117 13:34:39.493510 86948 solver.cpp:237] Iteration 74000, loss = 0.00152815
  1990. I0117 13:34:39.493753 86948 solver.cpp:253] Train net output #0: loss = 0.00152831 (* 1 = 0.00152831 loss)
  1991. I0117 13:34:39.854380 86948 sgd_solver.cpp:106] Iteration 74000, lr = 0.001
  1992. I0117 14:02:55.896353 86948 solver.cpp:341] Iteration 74500, Testing net (#0)
  1993. I0117 14:02:56.681279 86948 solver.cpp:409] Test net output #0: accuracy = 0.124
  1994. I0117 14:02:56.681329 86948 solver.cpp:409] Test net output #1: loss = 15.249 (* 1 = 15.249 loss)
  1995. I0117 14:03:17.415796 86948 solver.cpp:237] Iteration 74500, loss = 0.00171971
  1996. I0117 14:03:17.415844 86948 solver.cpp:253] Train net output #0: loss = 0.00171988 (* 1 = 0.00171988 loss)
  1997. I0117 14:03:17.761025 86948 sgd_solver.cpp:106] Iteration 74500, lr = 0.001
  1998. I0117 14:32:28.479389 86948 solver.cpp:459] Snapshotting to binary proto file models/mv16f/mv16f1__iter_75000.caffemodel
  1999. I0117 14:32:32.218113 86948 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/mv16f/mv16f1__iter_75000.solverstate
  2000. I0117 14:32:32.261855 86948 solver.cpp:341] Iteration 75000, Testing net (#0)
  2001. I0117 14:32:32.778828 86948 solver.cpp:409] Test net output #0: accuracy = 0.13
  2002. I0117 14:32:32.778861 86948 solver.cpp:409] Test net output #1: loss = 14.905 (* 1 = 14.905 loss)
  2003. I0117 14:32:50.592114 86948 solver.cpp:237] Iteration 75000, loss = 0.000816114
  2004. I0117 14:32:50.592182 86948 solver.cpp:253] Train net output #0: loss = 0.000816279 (* 1 = 0.000816279 loss)
  2005. I0117 14:32:50.894603 86948 sgd_solver.cpp:106] Iteration 75000, lr = 0.001
  2006. I0117 15:02:30.998363 86948 solver.cpp:341] Iteration 75500, Testing net (#0)
  2007. I0117 15:02:31.485143 86948 solver.cpp:409] Test net output #0: accuracy = 0.12
  2008. I0117 15:02:31.485188 86948 solver.cpp:409] Test net output #1: loss = 14.9655 (* 1 = 14.9655 loss)
  2009. I0117 15:02:52.363260 86948 solver.cpp:237] Iteration 75500, loss = 0.00145175
  2010. I0117 15:02:52.363328 86948 solver.cpp:253] Train net output #0: loss = 0.00145191 (* 1 = 0.00145191 loss)
  2011. I0117 15:02:52.684723 86948 sgd_solver.cpp:106] Iteration 75500, lr = 0.001
  2012. I0117 15:32:25.288990 86948 solver.cpp:341] Iteration 76000, Testing net (#0)
  2013. I0117 15:32:38.957864 86948 solver.cpp:409] Test net output #0: accuracy = 0.116
  2014. I0117 15:32:38.957921 86948 solver.cpp:409] Test net output #1: loss = 14.0129 (* 1 = 14.0129 loss)
  2015. I0117 15:32:51.730339 86948 solver.cpp:237] Iteration 76000, loss = 0.00122516
  2016. I0117 15:32:51.730393 86948 solver.cpp:253] Train net output #0: loss = 0.00122533 (* 1 = 0.00122533 loss)
  2017. I0117 15:32:52.032361 86948 sgd_solver.cpp:106] Iteration 76000, lr = 0.001
  2018. I0117 15:58:04.766777 86948 solver.cpp:341] Iteration 76500, Testing net (#0)
  2019. I0117 15:58:05.253283 86948 solver.cpp:409] Test net output #0: accuracy = 0.128
  2020. I0117 15:58:05.253428 86948 solver.cpp:409] Test net output #1: loss = 14.5003 (* 1 = 14.5003 loss)
  2021. I0117 15:58:36.863298 86948 solver.cpp:237] Iteration 76500, loss = 0.00182002
  2022. I0117 15:58:36.863498 86948 solver.cpp:253] Train net output #0: loss = 0.00182018 (* 1 = 0.00182018 loss)
  2023. I0117 15:58:37.219053 86948 sgd_solver.cpp:106] Iteration 76500, lr = 0.001
  2024. I0117 16:26:28.663875 86948 solver.cpp:341] Iteration 77000, Testing net (#0)
  2025. I0117 16:26:29.149608 86948 solver.cpp:409] Test net output #0: accuracy = 0.158
  2026. I0117 16:26:29.149651 86948 solver.cpp:409] Test net output #1: loss = 12.9062 (* 1 = 12.9062 loss)
  2027. I0117 16:26:49.170310 86948 solver.cpp:237] Iteration 77000, loss = 0.00109042
  2028. I0117 16:26:49.170364 86948 solver.cpp:253] Train net output #0: loss = 0.00109059 (* 1 = 0.00109059 loss)
  2029. I0117 16:26:49.472210 86948 sgd_solver.cpp:106] Iteration 77000, lr = 0.001
  2030. I0117 16:53:34.338349 86948 solver.cpp:341] Iteration 77500, Testing net (#0)
  2031. I0117 16:53:34.824774 86948 solver.cpp:409] Test net output #0: accuracy = 0.122
  2032. I0117 16:53:34.824815 86948 solver.cpp:409] Test net output #1: loss = 14.0104 (* 1 = 14.0104 loss)
  2033. I0117 16:53:49.258014 86948 solver.cpp:237] Iteration 77500, loss = 0.00227474
  2034. I0117 16:53:49.258074 86948 solver.cpp:253] Train net output #0: loss = 0.00227491 (* 1 = 0.00227491 loss)
  2035. I0117 16:53:49.604596 86948 sgd_solver.cpp:106] Iteration 77500, lr = 0.001
  2036. I0117 17:23:28.599503 86948 solver.cpp:341] Iteration 78000, Testing net (#0)
  2037. I0117 17:23:42.401674 86948 solver.cpp:409] Test net output #0: accuracy = 0.134
  2038. I0117 17:23:42.401721 86948 solver.cpp:409] Test net output #1: loss = 13.5035 (* 1 = 13.5035 loss)
  2039. I0117 17:23:59.519179 86948 solver.cpp:237] Iteration 78000, loss = 0.00130983
  2040. I0117 17:23:59.519390 86948 solver.cpp:253] Train net output #0: loss = 0.00131 (* 1 = 0.00131 loss)
  2041. I0117 17:23:59.820915 86948 sgd_solver.cpp:106] Iteration 78000, lr = 0.001
  2042. I0117 17:52:58.458806 86948 solver.cpp:341] Iteration 78500, Testing net (#0)
  2043. I0117 17:52:58.944988 86948 solver.cpp:409] Test net output #0: accuracy = 0.132
  2044. I0117 17:52:58.945024 86948 solver.cpp:409] Test net output #1: loss = 14.1663 (* 1 = 14.1663 loss)
  2045. I0117 17:53:19.896749 86948 solver.cpp:237] Iteration 78500, loss = 0.00147733
  2046. I0117 17:53:19.896806 86948 solver.cpp:253] Train net output #0: loss = 0.0014775 (* 1 = 0.0014775 loss)
  2047. I0117 17:53:19.896831 86948 sgd_solver.cpp:106] Iteration 78500, lr = 0.001
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!