API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Dec 31st, 2016
139
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 67.05 KB | None | 0 0
raw download clone embed print report
  1. ninja@ninja:~/caffe$ ./build/tools/caffe train --solver=models/bvlc_alexnet/solvermnist.prototxt
  2. I1231 09:58:24.341311 3750 caffe.cpp:217] Using GPUs 0
  3. I1231 09:58:24.407222 3750 caffe.cpp:222] GPU 0: GeForce GTX 1070
  4. I1231 09:58:24.956543 3750 solver.cpp:48] Initializing solver from parameters:
  5. test_iter: 1000
  6. test_interval: 1000
  7. base_lr: 0.01
  8. display: 20
  9. max_iter: 2000
  10. lr_policy: "step"
  11. gamma: 0.1
  12. momentum: 0.9
  13. weight_decay: 0.0005
  14. stepsize: 1000
  15. snapshot: 1000
  16. snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_train"
  17. solver_mode: GPU
  18. device_id: 0
  19. net: "models/bvlc_alexnet/train_valmnist.prototxt"
  20. train_state {
  21. level: 0
  22. stage: ""
  23. }
  24. I1231 09:58:24.956770 3750 solver.cpp:91] Creating training net from net file: models/bvlc_alexnet/train_valmnist.prototxt
  25. I1231 09:58:24.957945 3750 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
  26. I1231 09:58:24.957989 3750 net.cpp:322] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
  27. I1231 09:58:24.958338 3750 net.cpp:58] Initializing net from parameters:
  28. name: "AlexNet"
  29. state {
  30. phase: TRAIN
  31. level: 0
  32. stage: ""
  33. }
  34. layer {
  35. name: "data"
  36. type: "Data"
  37. top: "data"
  38. top: "label"
  39. include {
  40. phase: TRAIN
  41. }
  42. transform_param {
  43. scale: 0.00390625
  44. }
  45. data_param {
  46. source: "examples/mnist/mnist_train_lmdb"
  47. batch_size: 64
  48. backend: LMDB
  49. }
  50. }
  51. layer {
  52. name: "conv1"
  53. type: "Convolution"
  54. bottom: "data"
  55. top: "conv1"
  56. param {
  57. lr_mult: 1
  58. decay_mult: 1
  59. }
  60. param {
  61. lr_mult: 2
  62. decay_mult: 0
  63. }
  64. convolution_param {
  65. num_output: 96
  66. kernel_size: 3
  67. stride: 1
  68. weight_filler {
  69. type: "gaussian"
  70. std: 0.01
  71. }
  72. bias_filler {
  73. type: "constant"
  74. value: 0
  75. }
  76. }
  77. }
  78. layer {
  79. name: "relu1"
  80. type: "ReLU"
  81. bottom: "conv1"
  82. top: "conv1"
  83. }
  84. layer {
  85. name: "norm1"
  86. type: "LRN"
  87. bottom: "conv1"
  88. top: "norm1"
  89. lrn_param {
  90. local_size: 5
  91. alpha: 0.0001
  92. beta: 0.75
  93. }
  94. }
  95. layer {
  96. name: "pool1"
  97. type: "Pooling"
  98. bottom: "norm1"
  99. top: "pool1"
  100. pooling_param {
  101. pool: MAX
  102. kernel_size: 3
  103. stride: 1
  104. }
  105. }
  106. layer {
  107. name: "conv2"
  108. type: "Convolution"
  109. bottom: "pool1"
  110. top: "conv2"
  111. param {
  112. lr_mult: 1
  113. decay_mult: 1
  114. }
  115. param {
  116. lr_mult: 2
  117. decay_mult: 0
  118. }
  119. convolution_param {
  120. num_output: 256
  121. pad: 2
  122. kernel_size: 3
  123. group: 2
  124. weight_filler {
  125. type: "gaussian"
  126. std: 0.01
  127. }
  128. bias_filler {
  129. type: "constant"
  130. value: 0.1
  131. }
  132. }
  133. }
  134. layer {
  135. name: "relu2"
  136. type: "ReLU"
  137. bottom: "conv2"
  138. top: "conv2"
  139. }
  140. layer {
  141. name: "norm2"
  142. type: "LRN"
  143. bottom: "conv2"
  144. top: "norm2"
  145. lrn_param {
  146. local_size: 5
  147. alpha: 0.0001
  148. beta: 0.75
  149. }
  150. }
  151. layer {
  152. name: "pool2"
  153. type: "Pooling"
  154. bottom: "norm2"
  155. top: "pool2"
  156. pooling_param {
  157. pool: MAX
  158. kernel_size: 3
  159. stride: 2
  160. }
  161. }
  162. layer {
  163. name: "conv3"
  164. type: "Convolution"
  165. bottom: "pool2"
  166. top: "conv3"
  167. param {
  168. lr_mult: 1
  169. decay_mult: 1
  170. }
  171. param {
  172. lr_mult: 2
  173. decay_mult: 0
  174. }
  175. convolution_param {
  176. num_output: 384
  177. pad: 1
  178. kernel_size: 3
  179. weight_filler {
  180. type: "gaussian"
  181. std: 0.01
  182. }
  183. bias_filler {
  184. type: "constant"
  185. value: 0
  186. }
  187. }
  188. }
  189. layer {
  190. name: "relu3"
  191. type: "ReLU"
  192. bottom: "conv3"
  193. top: "conv3"
  194. }
  195. layer {
  196. name: "conv4"
  197. type: "Convolution"
  198. bottom: "conv3"
  199. top: "conv4"
  200. param {
  201. lr_mult: 1
  202. decay_mult: 1
  203. }
  204. param {
  205. lr_mult: 2
  206. decay_mult: 0
  207. }
  208. convolution_param {
  209. num_output: 384
  210. pad: 1
  211. kernel_size: 3
  212. group: 2
  213. weight_filler {
  214. type: "gaussian"
  215. std: 0.01
  216. }
  217. bias_filler {
  218. type: "constant"
  219. value: 0.1
  220. }
  221. }
  222. }
  223. layer {
  224. name: "relu4"
  225. type: "ReLU"
  226. bottom: "conv4"
  227. top: "conv4"
  228. }
  229. layer {
  230. name: "conv5"
  231. type: "Convolution"
  232. bottom: "conv4"
  233. top: "conv5"
  234. param {
  235. lr_mult: 1
  236. decay_mult: 1
  237. }
  238. param {
  239. lr_mult: 2
  240. decay_mult: 0
  241. }
  242. convolution_param {
  243. num_output: 256
  244. pad: 1
  245. kernel_size: 3
  246. group: 2
  247. weight_filler {
  248. type: "gaussian"
  249. std: 0.01
  250. }
  251. bias_filler {
  252. type: "constant"
  253. value: 0.1
  254. }
  255. }
  256. }
  257. layer {
  258. name: "relu5"
  259. type: "ReLU"
  260. bottom: "conv5"
  261. top: "conv5"
  262. }
  263. layer {
  264. name: "pool5"
  265. type: "Pooling"
  266. bottom: "conv5"
  267. top: "pool5"
  268. pooling_param {
  269. pool: MAX
  270. kernel_size: 3
  271. stride: 2
  272. }
  273. }
  274. layer {
  275. name: "fc6"
  276. type: "InnerProduct"
  277. bottom: "pool5"
  278. top: "fc6"
  279. param {
  280. lr_mult: 1
  281. decay_mult: 1
  282. }
  283. param {
  284. lr_mult: 2
  285. decay_mult: 0
  286. }
  287. inner_product_param {
  288. num_output: 4096
  289. weight_filler {
  290. type: "gaussian"
  291. std: 0.005
  292. }
  293. bias_filler {
  294. type: "constant"
  295. value: 0.1
  296. }
  297. }
  298. }
  299. layer {
  300. name: "relu6"
  301. type: "ReLU"
  302. bottom: "fc6"
  303. top: "fc6"
  304. }
  305. layer {
  306. name: "drop6"
  307. type: "Dropout"
  308. bottom: "fc6"
  309. top: "fc6"
  310. dropout_param {
  311. dropout_ratio: 0.5
  312. }
  313. }
  314. layer {
  315. name: "fc7"
  316. type: "InnerProduct"
  317. bottom: "fc6"
  318. top: "fc7"
  319. param {
  320. lr_mult: 1
  321. decay_mult: 1
  322. }
  323. param {
  324. lr_mult: 2
  325. decay_mult: 0
  326. }
  327. inner_product_param {
  328. num_output: 4096
  329. weight_filler {
  330. type: "gaussian"
  331. std: 0.005
  332. }
  333. bias_filler {
  334. type: "constant"
  335. value: 0.1
  336. }
  337. }
  338. }
  339. layer {
  340. name: "relu7"
  341. type: "ReLU"
  342. bottom: "fc7"
  343. top: "fc7"
  344. }
  345. layer {
  346. name: "drop7"
  347. type: "Dropout"
  348. bottom: "fc7"
  349. top: "fc7"
  350. dropout_param {
  351. dropout_ratio: 0.5
  352. }
  353. }
  354. layer {
  355. name: "fc8"
  356. type: "InnerProduct"
  357. bottom: "fc7"
  358. top: "fc8"
  359. param {
  360. lr_mult: 1
  361. decay_mult: 1
  362. }
  363. param {
  364. lr_mult: 2
  365. decay_mult: 0
  366. }
  367. inner_product_param {
  368. num_output: 1000
  369. weight_filler {
  370. type: "gaussian"
  371. std: 0.01
  372. }
  373. bias_filler {
  374. type: "constant"
  375. value: 0
  376. }
  377. }
  378. }
  379. layer {
  380. name: "loss"
  381. type: "SoftmaxWithLoss"
  382. bottom: "fc8"
  383. bottom: "label"
  384. top: "loss"
  385. }
  386. I1231 09:58:24.958807 3750 layer_factory.hpp:77] Creating layer data
  387. I1231 09:58:24.960191 3750 net.cpp:100] Creating Layer data
  388. I1231 09:58:24.960222 3750 net.cpp:408] data -> data
  389. I1231 09:58:24.960258 3750 net.cpp:408] data -> label
  390. I1231 09:58:24.962350 3759 db_lmdb.cpp:35] Opened lmdb examples/mnist/mnist_train_lmdb
  391. I1231 09:58:25.051169 3750 data_layer.cpp:41] output data size: 64,1,28,28
  392. I1231 09:58:25.053405 3750 net.cpp:150] Setting up data
  393. I1231 09:58:25.053428 3750 net.cpp:157] Top shape: 64 1 28 28 (50176)
  394. I1231 09:58:25.053438 3750 net.cpp:157] Top shape: 64 (64)
  395. I1231 09:58:25.053444 3750 net.cpp:165] Memory required for data: 200960
  396. I1231 09:58:25.053458 3750 layer_factory.hpp:77] Creating layer conv1
  397. I1231 09:58:25.053489 3750 net.cpp:100] Creating Layer conv1
  398. I1231 09:58:25.053498 3750 net.cpp:434] conv1 <- data
  399. I1231 09:58:25.053514 3750 net.cpp:408] conv1 -> conv1
  400. I1231 09:58:25.055207 3750 net.cpp:150] Setting up conv1
  401. I1231 09:58:25.055236 3750 net.cpp:157] Top shape: 64 96 26 26 (4153344)
  402. I1231 09:58:25.055241 3750 net.cpp:165] Memory required for data: 16814336
  403. I1231 09:58:25.055268 3750 layer_factory.hpp:77] Creating layer relu1
  404. I1231 09:58:25.055281 3750 net.cpp:100] Creating Layer relu1
  405. I1231 09:58:25.055289 3750 net.cpp:434] relu1 <- conv1
  406. I1231 09:58:25.055299 3750 net.cpp:395] relu1 -> conv1 (in-place)
  407. I1231 09:58:25.055310 3750 net.cpp:150] Setting up relu1
  408. I1231 09:58:25.055318 3750 net.cpp:157] Top shape: 64 96 26 26 (4153344)
  409. I1231 09:58:25.055326 3750 net.cpp:165] Memory required for data: 33427712
  410. I1231 09:58:25.055335 3750 layer_factory.hpp:77] Creating layer norm1
  411. I1231 09:58:25.055346 3750 net.cpp:100] Creating Layer norm1
  412. I1231 09:58:25.055352 3750 net.cpp:434] norm1 <- conv1
  413. I1231 09:58:25.055362 3750 net.cpp:408] norm1 -> norm1
  414. I1231 09:58:25.055415 3750 net.cpp:150] Setting up norm1
  415. I1231 09:58:25.055424 3750 net.cpp:157] Top shape: 64 96 26 26 (4153344)
  416. I1231 09:58:25.055430 3750 net.cpp:165] Memory required for data: 50041088
  417. I1231 09:58:25.055438 3750 layer_factory.hpp:77] Creating layer pool1
  418. I1231 09:58:25.055449 3750 net.cpp:100] Creating Layer pool1
  419. I1231 09:58:25.055454 3750 net.cpp:434] pool1 <- norm1
  420. I1231 09:58:25.055462 3750 net.cpp:408] pool1 -> pool1
  421. I1231 09:58:25.055522 3750 net.cpp:150] Setting up pool1
  422. I1231 09:58:25.055533 3750 net.cpp:157] Top shape: 64 96 24 24 (3538944)
  423. I1231 09:58:25.055563 3750 net.cpp:165] Memory required for data: 64196864
  424. I1231 09:58:25.055572 3750 layer_factory.hpp:77] Creating layer conv2
  425. I1231 09:58:25.055588 3750 net.cpp:100] Creating Layer conv2
  426. I1231 09:58:25.055593 3750 net.cpp:434] conv2 <- pool1
  427. I1231 09:58:25.055603 3750 net.cpp:408] conv2 -> conv2
  428. I1231 09:58:25.059763 3750 net.cpp:150] Setting up conv2
  429. I1231 09:58:25.059779 3750 net.cpp:157] Top shape: 64 256 26 26 (11075584)
  430. I1231 09:58:25.059784 3750 net.cpp:165] Memory required for data: 108499200
  431. I1231 09:58:25.059795 3750 layer_factory.hpp:77] Creating layer relu2
  432. I1231 09:58:25.059803 3750 net.cpp:100] Creating Layer relu2
  433. I1231 09:58:25.059809 3750 net.cpp:434] relu2 <- conv2
  434. I1231 09:58:25.059816 3750 net.cpp:395] relu2 -> conv2 (in-place)
  435. I1231 09:58:25.059824 3750 net.cpp:150] Setting up relu2
  436. I1231 09:58:25.059830 3750 net.cpp:157] Top shape: 64 256 26 26 (11075584)
  437. I1231 09:58:25.059835 3750 net.cpp:165] Memory required for data: 152801536
  438. I1231 09:58:25.059840 3750 layer_factory.hpp:77] Creating layer norm2
  439. I1231 09:58:25.059847 3750 net.cpp:100] Creating Layer norm2
  440. I1231 09:58:25.059851 3750 net.cpp:434] norm2 <- conv2
  441. I1231 09:58:25.059859 3750 net.cpp:408] norm2 -> norm2
  442. I1231 09:58:25.059898 3750 net.cpp:150] Setting up norm2
  443. I1231 09:58:25.059906 3750 net.cpp:157] Top shape: 64 256 26 26 (11075584)
  444. I1231 09:58:25.059908 3750 net.cpp:165] Memory required for data: 197103872
  445. I1231 09:58:25.059913 3750 layer_factory.hpp:77] Creating layer pool2
  446. I1231 09:58:25.059919 3750 net.cpp:100] Creating Layer pool2
  447. I1231 09:58:25.059923 3750 net.cpp:434] pool2 <- norm2
  448. I1231 09:58:25.059928 3750 net.cpp:408] pool2 -> pool2
  449. I1231 09:58:25.059962 3750 net.cpp:150] Setting up pool2
  450. I1231 09:58:25.059968 3750 net.cpp:157] Top shape: 64 256 13 13 (2768896)
  451. I1231 09:58:25.059972 3750 net.cpp:165] Memory required for data: 208179456
  452. I1231 09:58:25.059975 3750 layer_factory.hpp:77] Creating layer conv3
  453. I1231 09:58:25.059984 3750 net.cpp:100] Creating Layer conv3
  454. I1231 09:58:25.059988 3750 net.cpp:434] conv3 <- pool2
  455. I1231 09:58:25.059993 3750 net.cpp:408] conv3 -> conv3
  456. I1231 09:58:25.070734 3750 net.cpp:150] Setting up conv3
  457. I1231 09:58:25.070754 3750 net.cpp:157] Top shape: 64 384 13 13 (4153344)
  458. I1231 09:58:25.070755 3750 net.cpp:165] Memory required for data: 224792832
  459. I1231 09:58:25.070765 3750 layer_factory.hpp:77] Creating layer relu3
  460. I1231 09:58:25.070771 3750 net.cpp:100] Creating Layer relu3
  461. I1231 09:58:25.070775 3750 net.cpp:434] relu3 <- conv3
  462. I1231 09:58:25.070780 3750 net.cpp:395] relu3 -> conv3 (in-place)
  463. I1231 09:58:25.070787 3750 net.cpp:150] Setting up relu3
  464. I1231 09:58:25.070792 3750 net.cpp:157] Top shape: 64 384 13 13 (4153344)
  465. I1231 09:58:25.070796 3750 net.cpp:165] Memory required for data: 241406208
  466. I1231 09:58:25.070798 3750 layer_factory.hpp:77] Creating layer conv4
  467. I1231 09:58:25.070806 3750 net.cpp:100] Creating Layer conv4
  468. I1231 09:58:25.070809 3750 net.cpp:434] conv4 <- conv3
  469. I1231 09:58:25.070814 3750 net.cpp:408] conv4 -> conv4
  470. I1231 09:58:25.077191 3750 net.cpp:150] Setting up conv4
  471. I1231 09:58:25.077200 3750 net.cpp:157] Top shape: 64 384 13 13 (4153344)
  472. I1231 09:58:25.077203 3750 net.cpp:165] Memory required for data: 258019584
  473. I1231 09:58:25.077208 3750 layer_factory.hpp:77] Creating layer relu4
  474. I1231 09:58:25.077214 3750 net.cpp:100] Creating Layer relu4
  475. I1231 09:58:25.077215 3750 net.cpp:434] relu4 <- conv4
  476. I1231 09:58:25.077219 3750 net.cpp:395] relu4 -> conv4 (in-place)
  477. I1231 09:58:25.077224 3750 net.cpp:150] Setting up relu4
  478. I1231 09:58:25.077226 3750 net.cpp:157] Top shape: 64 384 13 13 (4153344)
  479. I1231 09:58:25.077229 3750 net.cpp:165] Memory required for data: 274632960
  480. I1231 09:58:25.077232 3750 layer_factory.hpp:77] Creating layer conv5
  481. I1231 09:58:25.077239 3750 net.cpp:100] Creating Layer conv5
  482. I1231 09:58:25.077242 3750 net.cpp:434] conv5 <- conv4
  483. I1231 09:58:25.077247 3750 net.cpp:408] conv5 -> conv5
  484. I1231 09:58:25.081475 3750 net.cpp:150] Setting up conv5
  485. I1231 09:58:25.081492 3750 net.cpp:157] Top shape: 64 256 13 13 (2768896)
  486. I1231 09:58:25.081511 3750 net.cpp:165] Memory required for data: 285708544
  487. I1231 09:58:25.081521 3750 layer_factory.hpp:77] Creating layer relu5
  488. I1231 09:58:25.081526 3750 net.cpp:100] Creating Layer relu5
  489. I1231 09:58:25.081531 3750 net.cpp:434] relu5 <- conv5
  490. I1231 09:58:25.081535 3750 net.cpp:395] relu5 -> conv5 (in-place)
  491. I1231 09:58:25.081542 3750 net.cpp:150] Setting up relu5
  492. I1231 09:58:25.081547 3750 net.cpp:157] Top shape: 64 256 13 13 (2768896)
  493. I1231 09:58:25.081552 3750 net.cpp:165] Memory required for data: 296784128
  494. I1231 09:58:25.081555 3750 layer_factory.hpp:77] Creating layer pool5
  495. I1231 09:58:25.081560 3750 net.cpp:100] Creating Layer pool5
  496. I1231 09:58:25.081564 3750 net.cpp:434] pool5 <- conv5
  497. I1231 09:58:25.081570 3750 net.cpp:408] pool5 -> pool5
  498. I1231 09:58:25.081604 3750 net.cpp:150] Setting up pool5
  499. I1231 09:58:25.081610 3750 net.cpp:157] Top shape: 64 256 6 6 (589824)
  500. I1231 09:58:25.081612 3750 net.cpp:165] Memory required for data: 299143424
  501. I1231 09:58:25.081616 3750 layer_factory.hpp:77] Creating layer fc6
  502. I1231 09:58:25.081622 3750 net.cpp:100] Creating Layer fc6
  503. I1231 09:58:25.081625 3750 net.cpp:434] fc6 <- pool5
  504. I1231 09:58:25.081629 3750 net.cpp:408] fc6 -> fc6
  505. I1231 09:58:25.361205 3750 net.cpp:150] Setting up fc6
  506. I1231 09:58:25.361223 3750 net.cpp:157] Top shape: 64 4096 (262144)
  507. I1231 09:58:25.361225 3750 net.cpp:165] Memory required for data: 300192000
  508. I1231 09:58:25.361232 3750 layer_factory.hpp:77] Creating layer relu6
  509. I1231 09:58:25.361239 3750 net.cpp:100] Creating Layer relu6
  510. I1231 09:58:25.361241 3750 net.cpp:434] relu6 <- fc6
  511. I1231 09:58:25.361245 3750 net.cpp:395] relu6 -> fc6 (in-place)
  512. I1231 09:58:25.361253 3750 net.cpp:150] Setting up relu6
  513. I1231 09:58:25.361256 3750 net.cpp:157] Top shape: 64 4096 (262144)
  514. I1231 09:58:25.361258 3750 net.cpp:165] Memory required for data: 301240576
  515. I1231 09:58:25.361261 3750 layer_factory.hpp:77] Creating layer drop6
  516. I1231 09:58:25.361265 3750 net.cpp:100] Creating Layer drop6
  517. I1231 09:58:25.361268 3750 net.cpp:434] drop6 <- fc6
  518. I1231 09:58:25.361270 3750 net.cpp:395] drop6 -> fc6 (in-place)
  519. I1231 09:58:25.361285 3750 net.cpp:150] Setting up drop6
  520. I1231 09:58:25.361289 3750 net.cpp:157] Top shape: 64 4096 (262144)
  521. I1231 09:58:25.361291 3750 net.cpp:165] Memory required for data: 302289152
  522. I1231 09:58:25.361292 3750 layer_factory.hpp:77] Creating layer fc7
  523. I1231 09:58:25.361296 3750 net.cpp:100] Creating Layer fc7
  524. I1231 09:58:25.361299 3750 net.cpp:434] fc7 <- fc6
  525. I1231 09:58:25.361302 3750 net.cpp:408] fc7 -> fc7
  526. I1231 09:58:25.485281 3750 net.cpp:150] Setting up fc7
  527. I1231 09:58:25.485301 3750 net.cpp:157] Top shape: 64 4096 (262144)
  528. I1231 09:58:25.485302 3750 net.cpp:165] Memory required for data: 303337728
  529. I1231 09:58:25.485311 3750 layer_factory.hpp:77] Creating layer relu7
  530. I1231 09:58:25.485316 3750 net.cpp:100] Creating Layer relu7
  531. I1231 09:58:25.485318 3750 net.cpp:434] relu7 <- fc7
  532. I1231 09:58:25.485323 3750 net.cpp:395] relu7 -> fc7 (in-place)
  533. I1231 09:58:25.485330 3750 net.cpp:150] Setting up relu7
  534. I1231 09:58:25.485333 3750 net.cpp:157] Top shape: 64 4096 (262144)
  535. I1231 09:58:25.485334 3750 net.cpp:165] Memory required for data: 304386304
  536. I1231 09:58:25.485337 3750 layer_factory.hpp:77] Creating layer drop7
  537. I1231 09:58:25.485340 3750 net.cpp:100] Creating Layer drop7
  538. I1231 09:58:25.485342 3750 net.cpp:434] drop7 <- fc7
  539. I1231 09:58:25.485344 3750 net.cpp:395] drop7 -> fc7 (in-place)
  540. I1231 09:58:25.485357 3750 net.cpp:150] Setting up drop7
  541. I1231 09:58:25.485361 3750 net.cpp:157] Top shape: 64 4096 (262144)
  542. I1231 09:58:25.485363 3750 net.cpp:165] Memory required for data: 305434880
  543. I1231 09:58:25.485364 3750 layer_factory.hpp:77] Creating layer fc8
  544. I1231 09:58:25.485368 3750 net.cpp:100] Creating Layer fc8
  545. I1231 09:58:25.485370 3750 net.cpp:434] fc8 <- fc7
  546. I1231 09:58:25.485374 3750 net.cpp:408] fc8 -> fc8
  547. I1231 09:58:25.516559 3750 net.cpp:150] Setting up fc8
  548. I1231 09:58:25.516577 3750 net.cpp:157] Top shape: 64 1000 (64000)
  549. I1231 09:58:25.516598 3750 net.cpp:165] Memory required for data: 305690880
  550. I1231 09:58:25.516607 3750 layer_factory.hpp:77] Creating layer loss
  551. I1231 09:58:25.516614 3750 net.cpp:100] Creating Layer loss
  552. I1231 09:58:25.516618 3750 net.cpp:434] loss <- fc8
  553. I1231 09:58:25.516623 3750 net.cpp:434] loss <- label
  554. I1231 09:58:25.516628 3750 net.cpp:408] loss -> loss
  555. I1231 09:58:25.516643 3750 layer_factory.hpp:77] Creating layer loss
  556. I1231 09:58:25.516732 3750 net.cpp:150] Setting up loss
  557. I1231 09:58:25.516736 3750 net.cpp:157] Top shape: (1)
  558. I1231 09:58:25.516738 3750 net.cpp:160] with loss weight 1
  559. I1231 09:58:25.516757 3750 net.cpp:165] Memory required for data: 305690884
  560. I1231 09:58:25.516759 3750 net.cpp:226] loss needs backward computation.
  561. I1231 09:58:25.516765 3750 net.cpp:226] fc8 needs backward computation.
  562. I1231 09:58:25.516768 3750 net.cpp:226] drop7 needs backward computation.
  563. I1231 09:58:25.516770 3750 net.cpp:226] relu7 needs backward computation.
  564. I1231 09:58:25.516772 3750 net.cpp:226] fc7 needs backward computation.
  565. I1231 09:58:25.516775 3750 net.cpp:226] drop6 needs backward computation.
  566. I1231 09:58:25.516778 3750 net.cpp:226] relu6 needs backward computation.
  567. I1231 09:58:25.516782 3750 net.cpp:226] fc6 needs backward computation.
  568. I1231 09:58:25.516784 3750 net.cpp:226] pool5 needs backward computation.
  569. I1231 09:58:25.516788 3750 net.cpp:226] relu5 needs backward computation.
  570. I1231 09:58:25.516790 3750 net.cpp:226] conv5 needs backward computation.
  571. I1231 09:58:25.516793 3750 net.cpp:226] relu4 needs backward computation.
  572. I1231 09:58:25.516795 3750 net.cpp:226] conv4 needs backward computation.
  573. I1231 09:58:25.516798 3750 net.cpp:226] relu3 needs backward computation.
  574. I1231 09:58:25.516801 3750 net.cpp:226] conv3 needs backward computation.
  575. I1231 09:58:25.516804 3750 net.cpp:226] pool2 needs backward computation.
  576. I1231 09:58:25.516806 3750 net.cpp:226] norm2 needs backward computation.
  577. I1231 09:58:25.516809 3750 net.cpp:226] relu2 needs backward computation.
  578. I1231 09:58:25.516813 3750 net.cpp:226] conv2 needs backward computation.
  579. I1231 09:58:25.516814 3750 net.cpp:226] pool1 needs backward computation.
  580. I1231 09:58:25.516816 3750 net.cpp:226] norm1 needs backward computation.
  581. I1231 09:58:25.516819 3750 net.cpp:226] relu1 needs backward computation.
  582. I1231 09:58:25.516822 3750 net.cpp:226] conv1 needs backward computation.
  583. I1231 09:58:25.516825 3750 net.cpp:228] data does not need backward computation.
  584. I1231 09:58:25.516827 3750 net.cpp:270] This network produces output loss
  585. I1231 09:58:25.516837 3750 net.cpp:283] Network initialization done.
  586. I1231 09:58:25.517024 3750 solver.cpp:181] Creating test net (#0) specified by net file: models/bvlc_alexnet/train_valmnist.prototxt
  587. I1231 09:58:25.517045 3750 net.cpp:322] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
  588. I1231 09:58:25.517154 3750 net.cpp:58] Initializing net from parameters:
  589. name: "AlexNet"
  590. state {
  591. phase: TEST
  592. }
  593. layer {
  594. name: "data"
  595. type: "Data"
  596. top: "data"
  597. top: "label"
  598. include {
  599. phase: TEST
  600. }
  601. transform_param {
  602. scale: 0.00390625
  603. }
  604. data_param {
  605. source: "examples/mnist/mnist_test_lmdb"
  606. batch_size: 100
  607. backend: LMDB
  608. }
  609. }
  610. layer {
  611. name: "conv1"
  612. type: "Convolution"
  613. bottom: "data"
  614. top: "conv1"
  615. param {
  616. lr_mult: 1
  617. decay_mult: 1
  618. }
  619. param {
  620. lr_mult: 2
  621. decay_mult: 0
  622. }
  623. convolution_param {
  624. num_output: 96
  625. kernel_size: 3
  626. stride: 1
  627. weight_filler {
  628. type: "gaussian"
  629. std: 0.01
  630. }
  631. bias_filler {
  632. type: "constant"
  633. value: 0
  634. }
  635. }
  636. }
  637. layer {
  638. name: "relu1"
  639. type: "ReLU"
  640. bottom: "conv1"
  641. top: "conv1"
  642. }
  643. layer {
  644. name: "norm1"
  645. type: "LRN"
  646. bottom: "conv1"
  647. top: "norm1"
  648. lrn_param {
  649. local_size: 5
  650. alpha: 0.0001
  651. beta: 0.75
  652. }
  653. }
  654. layer {
  655. name: "pool1"
  656. type: "Pooling"
  657. bottom: "norm1"
  658. top: "pool1"
  659. pooling_param {
  660. pool: MAX
  661. kernel_size: 3
  662. stride: 1
  663. }
  664. }
  665. layer {
  666. name: "conv2"
  667. type: "Convolution"
  668. bottom: "pool1"
  669. top: "conv2"
  670. param {
  671. lr_mult: 1
  672. decay_mult: 1
  673. }
  674. param {
  675. lr_mult: 2
  676. decay_mult: 0
  677. }
  678. convolution_param {
  679. num_output: 256
  680. pad: 2
  681. kernel_size: 3
  682. group: 2
  683. weight_filler {
  684. type: "gaussian"
  685. std: 0.01
  686. }
  687. bias_filler {
  688. type: "constant"
  689. value: 0.1
  690. }
  691. }
  692. }
  693. layer {
  694. name: "relu2"
  695. type: "ReLU"
  696. bottom: "conv2"
  697. top: "conv2"
  698. }
  699. layer {
  700. name: "norm2"
  701. type: "LRN"
  702. bottom: "conv2"
  703. top: "norm2"
  704. lrn_param {
  705. local_size: 5
  706. alpha: 0.0001
  707. beta: 0.75
  708. }
  709. }
  710. layer {
  711. name: "pool2"
  712. type: "Pooling"
  713. bottom: "norm2"
  714. top: "pool2"
  715. pooling_param {
  716. pool: MAX
  717. kernel_size: 3
  718. stride: 2
  719. }
  720. }
  721. layer {
  722. name: "conv3"
  723. type: "Convolution"
  724. bottom: "pool2"
  725. top: "conv3"
  726. param {
  727. lr_mult: 1
  728. decay_mult: 1
  729. }
  730. param {
  731. lr_mult: 2
  732. decay_mult: 0
  733. }
  734. convolution_param {
  735. num_output: 384
  736. pad: 1
  737. kernel_size: 3
  738. weight_filler {
  739. type: "gaussian"
  740. std: 0.01
  741. }
  742. bias_filler {
  743. type: "constant"
  744. value: 0
  745. }
  746. }
  747. }
  748. layer {
  749. name: "relu3"
  750. type: "ReLU"
  751. bottom: "conv3"
  752. top: "conv3"
  753. }
  754. layer {
  755. name: "conv4"
  756. type: "Convolution"
  757. bottom: "conv3"
  758. top: "conv4"
  759. param {
  760. lr_mult: 1
  761. decay_mult: 1
  762. }
  763. param {
  764. lr_mult: 2
  765. decay_mult: 0
  766. }
  767. convolution_param {
  768. num_output: 384
  769. pad: 1
  770. kernel_size: 3
  771. group: 2
  772. weight_filler {
  773. type: "gaussian"
  774. std: 0.01
  775. }
  776. bias_filler {
  777. type: "constant"
  778. value: 0.1
  779. }
  780. }
  781. }
  782. layer {
  783. name: "relu4"
  784. type: "ReLU"
  785. bottom: "conv4"
  786. top: "conv4"
  787. }
  788. layer {
  789. name: "conv5"
  790. type: "Convolution"
  791. bottom: "conv4"
  792. top: "conv5"
  793. param {
  794. lr_mult: 1
  795. decay_mult: 1
  796. }
  797. param {
  798. lr_mult: 2
  799. decay_mult: 0
  800. }
  801. convolution_param {
  802. num_output: 256
  803. pad: 1
  804. kernel_size: 3
  805. group: 2
  806. weight_filler {
  807. type: "gaussian"
  808. std: 0.01
  809. }
  810. bias_filler {
  811. type: "constant"
  812. value: 0.1
  813. }
  814. }
  815. }
  816. layer {
  817. name: "relu5"
  818. type: "ReLU"
  819. bottom: "conv5"
  820. top: "conv5"
  821. }
  822. layer {
  823. name: "pool5"
  824. type: "Pooling"
  825. bottom: "conv5"
  826. top: "pool5"
  827. pooling_param {
  828. pool: MAX
  829. kernel_size: 3
  830. stride: 2
  831. }
  832. }
  833. layer {
  834. name: "fc6"
  835. type: "InnerProduct"
  836. bottom: "pool5"
  837. top: "fc6"
  838. param {
  839. lr_mult: 1
  840. decay_mult: 1
  841. }
  842. param {
  843. lr_mult: 2
  844. decay_mult: 0
  845. }
  846. inner_product_param {
  847. num_output: 4096
  848. weight_filler {
  849. type: "gaussian"
  850. std: 0.005
  851. }
  852. bias_filler {
  853. type: "constant"
  854. value: 0.1
  855. }
  856. }
  857. }
  858. layer {
  859. name: "relu6"
  860. type: "ReLU"
  861. bottom: "fc6"
  862. top: "fc6"
  863. }
  864. layer {
  865. name: "drop6"
  866. type: "Dropout"
  867. bottom: "fc6"
  868. top: "fc6"
  869. dropout_param {
  870. dropout_ratio: 0.5
  871. }
  872. }
  873. layer {
  874. name: "fc7"
  875. type: "InnerProduct"
  876. bottom: "fc6"
  877. top: "fc7"
  878. param {
  879. lr_mult: 1
  880. decay_mult: 1
  881. }
  882. param {
  883. lr_mult: 2
  884. decay_mult: 0
  885. }
  886. inner_product_param {
  887. num_output: 4096
  888. weight_filler {
  889. type: "gaussian"
  890. std: 0.005
  891. }
  892. bias_filler {
  893. type: "constant"
  894. value: 0.1
  895. }
  896. }
  897. }
  898. layer {
  899. name: "relu7"
  900. type: "ReLU"
  901. bottom: "fc7"
  902. top: "fc7"
  903. }
  904. layer {
  905. name: "drop7"
  906. type: "Dropout"
  907. bottom: "fc7"
  908. top: "fc7"
  909. dropout_param {
  910. dropout_ratio: 0.5
  911. }
  912. }
  913. layer {
  914. name: "fc8"
  915. type: "InnerProduct"
  916. bottom: "fc7"
  917. top: "fc8"
  918. param {
  919. lr_mult: 1
  920. decay_mult: 1
  921. }
  922. param {
  923. lr_mult: 2
  924. decay_mult: 0
  925. }
  926. inner_product_param {
  927. num_output: 1000
  928. weight_filler {
  929. type: "gaussian"
  930. std: 0.01
  931. }
  932. bias_filler {
  933. type: "constant"
  934. value: 0
  935. }
  936. }
  937. }
  938. layer {
  939. name: "accuracy"
  940. type: "Accuracy"
  941. bottom: "fc8"
  942. bottom: "label"
  943. top: "accuracy"
  944. include {
  945. phase: TEST
  946. }
  947. }
  948. layer {
  949. name: "loss"
  950. type: "SoftmaxWithLoss"
  951. bottom: "fc8"
  952. bottom: "label"
  953. top: "loss"
  954. }
  955. I1231 09:58:25.517351 3750 layer_factory.hpp:77] Creating layer data
  956. I1231 09:58:25.517536 3750 net.cpp:100] Creating Layer data
  957. I1231 09:58:25.517542 3750 net.cpp:408] data -> data
  958. I1231 09:58:25.517549 3750 net.cpp:408] data -> label
  959. I1231 09:58:25.529608 3762 db_lmdb.cpp:35] Opened lmdb examples/mnist/mnist_test_lmdb
  960. I1231 09:58:25.530697 3750 data_layer.cpp:41] output data size: 100,1,28,28
  961. I1231 09:58:25.531313 3750 net.cpp:150] Setting up data
  962. I1231 09:58:25.531321 3750 net.cpp:157] Top shape: 100 1 28 28 (78400)
  963. I1231 09:58:25.531325 3750 net.cpp:157] Top shape: 100 (100)
  964. I1231 09:58:25.531328 3750 net.cpp:165] Memory required for data: 314000
  965. I1231 09:58:25.531332 3750 layer_factory.hpp:77] Creating layer label_data_1_split
  966. I1231 09:58:25.531339 3750 net.cpp:100] Creating Layer label_data_1_split
  967. I1231 09:58:25.531343 3750 net.cpp:434] label_data_1_split <- label
  968. I1231 09:58:25.531348 3750 net.cpp:408] label_data_1_split -> label_data_1_split_0
  969. I1231 09:58:25.531354 3750 net.cpp:408] label_data_1_split -> label_data_1_split_1
  970. I1231 09:58:25.531381 3750 net.cpp:150] Setting up label_data_1_split
  971. I1231 09:58:25.531385 3750 net.cpp:157] Top shape: 100 (100)
  972. I1231 09:58:25.531389 3750 net.cpp:157] Top shape: 100 (100)
  973. I1231 09:58:25.531393 3750 net.cpp:165] Memory required for data: 314800
  974. I1231 09:58:25.531395 3750 layer_factory.hpp:77] Creating layer conv1
  975. I1231 09:58:25.531404 3750 net.cpp:100] Creating Layer conv1
  976. I1231 09:58:25.531406 3750 net.cpp:434] conv1 <- data
  977. I1231 09:58:25.531411 3750 net.cpp:408] conv1 -> conv1
  978. I1231 09:58:25.531586 3750 net.cpp:150] Setting up conv1
  979. I1231 09:58:25.531592 3750 net.cpp:157] Top shape: 100 96 26 26 (6489600)
  980. I1231 09:58:25.531596 3750 net.cpp:165] Memory required for data: 26273200
  981. I1231 09:58:25.531604 3750 layer_factory.hpp:77] Creating layer relu1
  982. I1231 09:58:25.531607 3750 net.cpp:100] Creating Layer relu1
  983. I1231 09:58:25.531611 3750 net.cpp:434] relu1 <- conv1
  984. I1231 09:58:25.531615 3750 net.cpp:395] relu1 -> conv1 (in-place)
  985. I1231 09:58:25.531618 3750 net.cpp:150] Setting up relu1
  986. I1231 09:58:25.531622 3750 net.cpp:157] Top shape: 100 96 26 26 (6489600)
  987. I1231 09:58:25.531625 3750 net.cpp:165] Memory required for data: 52231600
  988. I1231 09:58:25.531628 3750 layer_factory.hpp:77] Creating layer norm1
  989. I1231 09:58:25.531635 3750 net.cpp:100] Creating Layer norm1
  990. I1231 09:58:25.531636 3750 net.cpp:434] norm1 <- conv1
  991. I1231 09:58:25.531641 3750 net.cpp:408] norm1 -> norm1
  992. I1231 09:58:25.531661 3750 net.cpp:150] Setting up norm1
  993. I1231 09:58:25.531666 3750 net.cpp:157] Top shape: 100 96 26 26 (6489600)
  994. I1231 09:58:25.531668 3750 net.cpp:165] Memory required for data: 78190000
  995. I1231 09:58:25.531672 3750 layer_factory.hpp:77] Creating layer pool1
  996. I1231 09:58:25.531677 3750 net.cpp:100] Creating Layer pool1
  997. I1231 09:58:25.531679 3750 net.cpp:434] pool1 <- norm1
  998. I1231 09:58:25.531683 3750 net.cpp:408] pool1 -> pool1
  999. I1231 09:58:25.531702 3750 net.cpp:150] Setting up pool1
  1000. I1231 09:58:25.531707 3750 net.cpp:157] Top shape: 100 96 24 24 (5529600)
  1001. I1231 09:58:25.531709 3750 net.cpp:165] Memory required for data: 100308400
  1002. I1231 09:58:25.531713 3750 layer_factory.hpp:77] Creating layer conv2
  1003. I1231 09:58:25.531719 3750 net.cpp:100] Creating Layer conv2
  1004. I1231 09:58:25.531723 3750 net.cpp:434] conv2 <- pool1
  1005. I1231 09:58:25.531728 3750 net.cpp:408] conv2 -> conv2
  1006. I1231 09:58:25.532605 3750 net.cpp:150] Setting up conv2
  1007. I1231 09:58:25.532611 3750 net.cpp:157] Top shape: 100 256 26 26 (17305600)
  1008. I1231 09:58:25.532614 3750 net.cpp:165] Memory required for data: 169530800
  1009. I1231 09:58:25.532621 3750 layer_factory.hpp:77] Creating layer relu2
  1010. I1231 09:58:25.532626 3750 net.cpp:100] Creating Layer relu2
  1011. I1231 09:58:25.532630 3750 net.cpp:434] relu2 <- conv2
  1012. I1231 09:58:25.532635 3750 net.cpp:395] relu2 -> conv2 (in-place)
  1013. I1231 09:58:25.532639 3750 net.cpp:150] Setting up relu2
  1014. I1231 09:58:25.532645 3750 net.cpp:157] Top shape: 100 256 26 26 (17305600)
  1015. I1231 09:58:25.532649 3750 net.cpp:165] Memory required for data: 238753200
  1016. I1231 09:58:25.532652 3750 layer_factory.hpp:77] Creating layer norm2
  1017. I1231 09:58:25.532658 3750 net.cpp:100] Creating Layer norm2
  1018. I1231 09:58:25.532661 3750 net.cpp:434] norm2 <- conv2
  1019. I1231 09:58:25.532667 3750 net.cpp:408] norm2 -> norm2
  1020. I1231 09:58:25.532724 3750 net.cpp:150] Setting up norm2
  1021. I1231 09:58:25.532730 3750 net.cpp:157] Top shape: 100 256 26 26 (17305600)
  1022. I1231 09:58:25.532733 3750 net.cpp:165] Memory required for data: 307975600
  1023. I1231 09:58:25.532737 3750 layer_factory.hpp:77] Creating layer pool2
  1024. I1231 09:58:25.532742 3750 net.cpp:100] Creating Layer pool2
  1025. I1231 09:58:25.532747 3750 net.cpp:434] pool2 <- norm2
  1026. I1231 09:58:25.532752 3750 net.cpp:408] pool2 -> pool2
  1027. I1231 09:58:25.532773 3750 net.cpp:150] Setting up pool2
  1028. I1231 09:58:25.532778 3750 net.cpp:157] Top shape: 100 256 13 13 (4326400)
  1029. I1231 09:58:25.532781 3750 net.cpp:165] Memory required for data: 325281200
  1030. I1231 09:58:25.532785 3750 layer_factory.hpp:77] Creating layer conv3
  1031. I1231 09:58:25.532793 3750 net.cpp:100] Creating Layer conv3
  1032. I1231 09:58:25.532795 3750 net.cpp:434] conv3 <- pool2
  1033. I1231 09:58:25.532801 3750 net.cpp:408] conv3 -> conv3
  1034. I1231 09:58:25.539577 3750 net.cpp:150] Setting up conv3
  1035. I1231 09:58:25.539587 3750 net.cpp:157] Top shape: 100 384 13 13 (6489600)
  1036. I1231 09:58:25.539592 3750 net.cpp:165] Memory required for data: 351239600
  1037. I1231 09:58:25.539599 3750 layer_factory.hpp:77] Creating layer relu3
  1038. I1231 09:58:25.539604 3750 net.cpp:100] Creating Layer relu3
  1039. I1231 09:58:25.539608 3750 net.cpp:434] relu3 <- conv3
  1040. I1231 09:58:25.539613 3750 net.cpp:395] relu3 -> conv3 (in-place)
  1041. I1231 09:58:25.539621 3750 net.cpp:150] Setting up relu3
  1042. I1231 09:58:25.539625 3750 net.cpp:157] Top shape: 100 384 13 13 (6489600)
  1043. I1231 09:58:25.539629 3750 net.cpp:165] Memory required for data: 377198000
  1044. I1231 09:58:25.539638 3750 layer_factory.hpp:77] Creating layer conv4
  1045. I1231 09:58:25.539649 3750 net.cpp:100] Creating Layer conv4
  1046. I1231 09:58:25.539652 3750 net.cpp:434] conv4 <- conv3
  1047. I1231 09:58:25.539659 3750 net.cpp:408] conv4 -> conv4
  1048. I1231 09:58:25.544891 3750 net.cpp:150] Setting up conv4
  1049. I1231 09:58:25.544909 3750 net.cpp:157] Top shape: 100 384 13 13 (6489600)
  1050. I1231 09:58:25.544912 3750 net.cpp:165] Memory required for data: 403156400
  1051. I1231 09:58:25.544921 3750 layer_factory.hpp:77] Creating layer relu4
  1052. I1231 09:58:25.544929 3750 net.cpp:100] Creating Layer relu4
  1053. I1231 09:58:25.544931 3750 net.cpp:434] relu4 <- conv4
  1054. I1231 09:58:25.544937 3750 net.cpp:395] relu4 -> conv4 (in-place)
  1055. I1231 09:58:25.544944 3750 net.cpp:150] Setting up relu4
  1056. I1231 09:58:25.544947 3750 net.cpp:157] Top shape: 100 384 13 13 (6489600)
  1057. I1231 09:58:25.544950 3750 net.cpp:165] Memory required for data: 429114800
  1058. I1231 09:58:25.544955 3750 layer_factory.hpp:77] Creating layer conv5
  1059. I1231 09:58:25.544962 3750 net.cpp:100] Creating Layer conv5
  1060. I1231 09:58:25.544965 3750 net.cpp:434] conv5 <- conv4
  1061. I1231 09:58:25.544970 3750 net.cpp:408] conv5 -> conv5
  1062. I1231 09:58:25.548626 3750 net.cpp:150] Setting up conv5
  1063. I1231 09:58:25.548636 3750 net.cpp:157] Top shape: 100 256 13 13 (4326400)
  1064. I1231 09:58:25.548640 3750 net.cpp:165] Memory required for data: 446420400
  1065. I1231 09:58:25.548648 3750 layer_factory.hpp:77] Creating layer relu5
  1066. I1231 09:58:25.548653 3750 net.cpp:100] Creating Layer relu5
  1067. I1231 09:58:25.548656 3750 net.cpp:434] relu5 <- conv5
  1068. I1231 09:58:25.548661 3750 net.cpp:395] relu5 -> conv5 (in-place)
  1069. I1231 09:58:25.548666 3750 net.cpp:150] Setting up relu5
  1070. I1231 09:58:25.548669 3750 net.cpp:157] Top shape: 100 256 13 13 (4326400)
  1071. I1231 09:58:25.548673 3750 net.cpp:165] Memory required for data: 463726000
  1072. I1231 09:58:25.548676 3750 layer_factory.hpp:77] Creating layer pool5
  1073. I1231 09:58:25.548683 3750 net.cpp:100] Creating Layer pool5
  1074. I1231 09:58:25.548686 3750 net.cpp:434] pool5 <- conv5
  1075. I1231 09:58:25.548691 3750 net.cpp:408] pool5 -> pool5
  1076. I1231 09:58:25.548715 3750 net.cpp:150] Setting up pool5
  1077. I1231 09:58:25.548720 3750 net.cpp:157] Top shape: 100 256 6 6 (921600)
  1078. I1231 09:58:25.548722 3750 net.cpp:165] Memory required for data: 467412400
  1079. I1231 09:58:25.548727 3750 layer_factory.hpp:77] Creating layer fc6
  1080. I1231 09:58:25.548732 3750 net.cpp:100] Creating Layer fc6
  1081. I1231 09:58:25.548735 3750 net.cpp:434] fc6 <- pool5
  1082. I1231 09:58:25.548739 3750 net.cpp:408] fc6 -> fc6
  1083. I1231 09:58:25.825659 3750 net.cpp:150] Setting up fc6
  1084. I1231 09:58:25.825680 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1085. I1231 09:58:25.825685 3750 net.cpp:165] Memory required for data: 469050800
  1086. I1231 09:58:25.825693 3750 layer_factory.hpp:77] Creating layer relu6
  1087. I1231 09:58:25.825701 3750 net.cpp:100] Creating Layer relu6
  1088. I1231 09:58:25.825706 3750 net.cpp:434] relu6 <- fc6
  1089. I1231 09:58:25.825712 3750 net.cpp:395] relu6 -> fc6 (in-place)
  1090. I1231 09:58:25.825719 3750 net.cpp:150] Setting up relu6
  1091. I1231 09:58:25.825724 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1092. I1231 09:58:25.825727 3750 net.cpp:165] Memory required for data: 470689200
  1093. I1231 09:58:25.825731 3750 layer_factory.hpp:77] Creating layer drop6
  1094. I1231 09:58:25.825736 3750 net.cpp:100] Creating Layer drop6
  1095. I1231 09:58:25.825739 3750 net.cpp:434] drop6 <- fc6
  1096. I1231 09:58:25.825744 3750 net.cpp:395] drop6 -> fc6 (in-place)
  1097. I1231 09:58:25.825763 3750 net.cpp:150] Setting up drop6
  1098. I1231 09:58:25.825767 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1099. I1231 09:58:25.825769 3750 net.cpp:165] Memory required for data: 472327600
  1100. I1231 09:58:25.825772 3750 layer_factory.hpp:77] Creating layer fc7
  1101. I1231 09:58:25.825776 3750 net.cpp:100] Creating Layer fc7
  1102. I1231 09:58:25.825778 3750 net.cpp:434] fc7 <- fc6
  1103. I1231 09:58:25.825781 3750 net.cpp:408] fc7 -> fc7
  1104. I1231 09:58:25.950098 3750 net.cpp:150] Setting up fc7
  1105. I1231 09:58:25.950117 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1106. I1231 09:58:25.950121 3750 net.cpp:165] Memory required for data: 473966000
  1107. I1231 09:58:25.950130 3750 layer_factory.hpp:77] Creating layer relu7
  1108. I1231 09:58:25.950139 3750 net.cpp:100] Creating Layer relu7
  1109. I1231 09:58:25.950142 3750 net.cpp:434] relu7 <- fc7
  1110. I1231 09:58:25.950148 3750 net.cpp:395] relu7 -> fc7 (in-place)
  1111. I1231 09:58:25.950155 3750 net.cpp:150] Setting up relu7
  1112. I1231 09:58:25.950160 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1113. I1231 09:58:25.950162 3750 net.cpp:165] Memory required for data: 475604400
  1114. I1231 09:58:25.950166 3750 layer_factory.hpp:77] Creating layer drop7
  1115. I1231 09:58:25.950172 3750 net.cpp:100] Creating Layer drop7
  1116. I1231 09:58:25.950176 3750 net.cpp:434] drop7 <- fc7
  1117. I1231 09:58:25.950179 3750 net.cpp:395] drop7 -> fc7 (in-place)
  1118. I1231 09:58:25.950199 3750 net.cpp:150] Setting up drop7
  1119. I1231 09:58:25.950203 3750 net.cpp:157] Top shape: 100 4096 (409600)
  1120. I1231 09:58:25.950206 3750 net.cpp:165] Memory required for data: 477242800
  1121. I1231 09:58:25.950207 3750 layer_factory.hpp:77] Creating layer fc8
  1122. I1231 09:58:25.950212 3750 net.cpp:100] Creating Layer fc8
  1123. I1231 09:58:25.950213 3750 net.cpp:434] fc8 <- fc7
  1124. I1231 09:58:25.950217 3750 net.cpp:408] fc8 -> fc8
  1125. I1231 09:58:25.980584 3750 net.cpp:150] Setting up fc8
  1126. I1231 09:58:25.980602 3750 net.cpp:157] Top shape: 100 1000 (100000)
  1127. I1231 09:58:25.980604 3750 net.cpp:165] Memory required for data: 477642800
  1128. I1231 09:58:25.980612 3750 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
  1129. I1231 09:58:25.980618 3750 net.cpp:100] Creating Layer fc8_fc8_0_split
  1130. I1231 09:58:25.980620 3750 net.cpp:434] fc8_fc8_0_split <- fc8
  1131. I1231 09:58:25.980625 3750 net.cpp:408] fc8_fc8_0_split -> fc8_fc8_0_split_0
  1132. I1231 09:58:25.980659 3750 net.cpp:408] fc8_fc8_0_split -> fc8_fc8_0_split_1
  1133. I1231 09:58:25.980685 3750 net.cpp:150] Setting up fc8_fc8_0_split
  1134. I1231 09:58:25.980689 3750 net.cpp:157] Top shape: 100 1000 (100000)
  1135. I1231 09:58:25.980691 3750 net.cpp:157] Top shape: 100 1000 (100000)
  1136. I1231 09:58:25.980693 3750 net.cpp:165] Memory required for data: 478442800
  1137. I1231 09:58:25.980695 3750 layer_factory.hpp:77] Creating layer accuracy
  1138. I1231 09:58:25.980700 3750 net.cpp:100] Creating Layer accuracy
  1139. I1231 09:58:25.980701 3750 net.cpp:434] accuracy <- fc8_fc8_0_split_0
  1140. I1231 09:58:25.980705 3750 net.cpp:434] accuracy <- label_data_1_split_0
  1141. I1231 09:58:25.980707 3750 net.cpp:408] accuracy -> accuracy
  1142. I1231 09:58:25.980712 3750 net.cpp:150] Setting up accuracy
  1143. I1231 09:58:25.980715 3750 net.cpp:157] Top shape: (1)
  1144. I1231 09:58:25.980718 3750 net.cpp:165] Memory required for data: 478442804
  1145. I1231 09:58:25.980741 3750 layer_factory.hpp:77] Creating layer loss
  1146. I1231 09:58:25.980746 3750 net.cpp:100] Creating Layer loss
  1147. I1231 09:58:25.980749 3750 net.cpp:434] loss <- fc8_fc8_0_split_1
  1148. I1231 09:58:25.980753 3750 net.cpp:434] loss <- label_data_1_split_1
  1149. I1231 09:58:25.980757 3750 net.cpp:408] loss -> loss
  1150. I1231 09:58:25.980764 3750 layer_factory.hpp:77] Creating layer loss
  1151. I1231 09:58:25.980870 3750 net.cpp:150] Setting up loss
  1152. I1231 09:58:25.980873 3750 net.cpp:157] Top shape: (1)
  1153. I1231 09:58:25.980875 3750 net.cpp:160] with loss weight 1
  1154. I1231 09:58:25.980883 3750 net.cpp:165] Memory required for data: 478442808
  1155. I1231 09:58:25.980885 3750 net.cpp:226] loss needs backward computation.
  1156. I1231 09:58:25.980888 3750 net.cpp:228] accuracy does not need backward computation.
  1157. I1231 09:58:25.980890 3750 net.cpp:226] fc8_fc8_0_split needs backward computation.
  1158. I1231 09:58:25.980892 3750 net.cpp:226] fc8 needs backward computation.
  1159. I1231 09:58:25.980895 3750 net.cpp:226] drop7 needs backward computation.
  1160. I1231 09:58:25.980896 3750 net.cpp:226] relu7 needs backward computation.
  1161. I1231 09:58:25.980898 3750 net.cpp:226] fc7 needs backward computation.
  1162. I1231 09:58:25.980901 3750 net.cpp:226] drop6 needs backward computation.
  1163. I1231 09:58:25.980906 3750 net.cpp:226] relu6 needs backward computation.
  1164. I1231 09:58:25.980908 3750 net.cpp:226] fc6 needs backward computation.
  1165. I1231 09:58:25.980911 3750 net.cpp:226] pool5 needs backward computation.
  1166. I1231 09:58:25.980913 3750 net.cpp:226] relu5 needs backward computation.
  1167. I1231 09:58:25.980917 3750 net.cpp:226] conv5 needs backward computation.
  1168. I1231 09:58:25.980921 3750 net.cpp:226] relu4 needs backward computation.
  1169. I1231 09:58:25.980924 3750 net.cpp:226] conv4 needs backward computation.
  1170. I1231 09:58:25.980927 3750 net.cpp:226] relu3 needs backward computation.
  1171. I1231 09:58:25.980931 3750 net.cpp:226] conv3 needs backward computation.
  1172. I1231 09:58:25.980936 3750 net.cpp:226] pool2 needs backward computation.
  1173. I1231 09:58:25.980939 3750 net.cpp:226] norm2 needs backward computation.
  1174. I1231 09:58:25.980942 3750 net.cpp:226] relu2 needs backward computation.
  1175. I1231 09:58:25.980947 3750 net.cpp:226] conv2 needs backward computation.
  1176. I1231 09:58:25.980949 3750 net.cpp:226] pool1 needs backward computation.
  1177. I1231 09:58:25.980953 3750 net.cpp:226] norm1 needs backward computation.
  1178. I1231 09:58:25.980957 3750 net.cpp:226] relu1 needs backward computation.
  1179. I1231 09:58:25.980960 3750 net.cpp:226] conv1 needs backward computation.
  1180. I1231 09:58:25.980964 3750 net.cpp:228] label_data_1_split does not need backward computation.
  1181. I1231 09:58:25.980968 3750 net.cpp:228] data does not need backward computation.
  1182. I1231 09:58:25.980973 3750 net.cpp:270] This network produces output accuracy
  1183. I1231 09:58:25.980976 3750 net.cpp:270] This network produces output loss
  1184. I1231 09:58:25.980988 3750 net.cpp:283] Network initialization done.
  1185. I1231 09:58:25.981042 3750 solver.cpp:60] Solver scaffolding done.
  1186. I1231 09:58:25.981361 3750 caffe.cpp:251] Starting Optimization
  1187. I1231 09:58:25.981365 3750 solver.cpp:279] Solving AlexNet
  1188. I1231 09:58:25.981369 3750 solver.cpp:280] Learning Rate Policy: step
  1189. I1231 09:58:25.982930 3750 solver.cpp:337] Iteration 0, Testing net (#0)
  1190. I1231 09:59:34.489457 3750 solver.cpp:404] Test net output #0: accuracy = 0
  1191. I1231 09:59:34.489511 3750 solver.cpp:404] Test net output #1: loss = 6.91249 (* 1 = 6.91249 loss)
  1192. I1231 09:59:34.574491 3750 solver.cpp:228] Iteration 0, loss = 6.90698
  1193. I1231 09:59:34.574506 3750 solver.cpp:244] Train net output #0: loss = 6.90698 (* 1 = 6.90698 loss)
  1194. I1231 09:59:34.574520 3750 sgd_solver.cpp:106] Iteration 0, lr = 0.01
  1195. I1231 09:59:37.203480 3750 solver.cpp:228] Iteration 20, loss = 2.87205
  1196. I1231 09:59:37.203502 3750 solver.cpp:244] Train net output #0: loss = 2.87205 (* 1 = 2.87205 loss)
  1197. I1231 09:59:37.203507 3750 sgd_solver.cpp:106] Iteration 20, lr = 0.01
  1198. I1231 09:59:39.827961 3750 solver.cpp:228] Iteration 40, loss = 2.44488
  1199. I1231 09:59:39.827983 3750 solver.cpp:244] Train net output #0: loss = 2.44488 (* 1 = 2.44488 loss)
  1200. I1231 09:59:39.827987 3750 sgd_solver.cpp:106] Iteration 40, lr = 0.01
  1201. I1231 09:59:42.452934 3750 solver.cpp:228] Iteration 60, loss = 2.30728
  1202. I1231 09:59:42.452955 3750 solver.cpp:244] Train net output #0: loss = 2.30728 (* 1 = 2.30728 loss)
  1203. I1231 09:59:42.452960 3750 sgd_solver.cpp:106] Iteration 60, lr = 0.01
  1204. I1231 09:59:45.078403 3750 solver.cpp:228] Iteration 80, loss = 2.37298
  1205. I1231 09:59:45.078423 3750 solver.cpp:244] Train net output #0: loss = 2.37298 (* 1 = 2.37298 loss)
  1206. I1231 09:59:45.078426 3750 sgd_solver.cpp:106] Iteration 80, lr = 0.01
  1207. I1231 09:59:47.703523 3750 solver.cpp:228] Iteration 100, loss = 2.35641
  1208. I1231 09:59:47.703547 3750 solver.cpp:244] Train net output #0: loss = 2.35641 (* 1 = 2.35641 loss)
  1209. I1231 09:59:47.703552 3750 sgd_solver.cpp:106] Iteration 100, lr = 0.01
  1210. I1231 09:59:50.335674 3750 solver.cpp:228] Iteration 120, loss = 2.37395
  1211. I1231 09:59:50.335695 3750 solver.cpp:244] Train net output #0: loss = 2.37395 (* 1 = 2.37395 loss)
  1212. I1231 09:59:50.335700 3750 sgd_solver.cpp:106] Iteration 120, lr = 0.01
  1213. I1231 09:59:52.969204 3750 solver.cpp:228] Iteration 140, loss = 2.40437
  1214. I1231 09:59:52.969228 3750 solver.cpp:244] Train net output #0: loss = 2.40437 (* 1 = 2.40437 loss)
  1215. I1231 09:59:52.969233 3750 sgd_solver.cpp:106] Iteration 140, lr = 0.01
  1216. I1231 09:59:55.607323 3750 solver.cpp:228] Iteration 160, loss = 2.38931
  1217. I1231 09:59:55.607345 3750 solver.cpp:244] Train net output #0: loss = 2.38931 (* 1 = 2.38931 loss)
  1218. I1231 09:59:55.607350 3750 sgd_solver.cpp:106] Iteration 160, lr = 0.01
  1219. I1231 09:59:58.231901 3750 solver.cpp:228] Iteration 180, loss = 2.37691
  1220. I1231 09:59:58.231922 3750 solver.cpp:244] Train net output #0: loss = 2.37691 (* 1 = 2.37691 loss)
  1221. I1231 09:59:58.231925 3750 sgd_solver.cpp:106] Iteration 180, lr = 0.01
  1222. I1231 10:00:00.858958 3750 solver.cpp:228] Iteration 200, loss = 2.30506
  1223. I1231 10:00:00.858978 3750 solver.cpp:244] Train net output #0: loss = 2.30506 (* 1 = 2.30506 loss)
  1224. I1231 10:00:00.858984 3750 sgd_solver.cpp:106] Iteration 200, lr = 0.01
  1225. I1231 10:00:03.483427 3750 solver.cpp:228] Iteration 220, loss = 2.31115
  1226. I1231 10:00:03.483449 3750 solver.cpp:244] Train net output #0: loss = 2.31115 (* 1 = 2.31115 loss)
  1227. I1231 10:00:03.483453 3750 sgd_solver.cpp:106] Iteration 220, lr = 0.01
  1228. I1231 10:00:06.110460 3750 solver.cpp:228] Iteration 240, loss = 2.30202
  1229. I1231 10:00:06.110561 3750 solver.cpp:244] Train net output #0: loss = 2.30202 (* 1 = 2.30202 loss)
  1230. I1231 10:00:06.110566 3750 sgd_solver.cpp:106] Iteration 240, lr = 0.01
  1231. I1231 10:00:08.747812 3750 solver.cpp:228] Iteration 260, loss = 2.31446
  1232. I1231 10:00:08.747835 3750 solver.cpp:244] Train net output #0: loss = 2.31446 (* 1 = 2.31446 loss)
  1233. I1231 10:00:08.747841 3750 sgd_solver.cpp:106] Iteration 260, lr = 0.01
  1234. I1231 10:00:11.374794 3750 solver.cpp:228] Iteration 280, loss = 2.31768
  1235. I1231 10:00:11.374815 3750 solver.cpp:244] Train net output #0: loss = 2.31768 (* 1 = 2.31768 loss)
  1236. I1231 10:00:11.374820 3750 sgd_solver.cpp:106] Iteration 280, lr = 0.01
  1237. I1231 10:00:14.000993 3750 solver.cpp:228] Iteration 300, loss = 2.35765
  1238. I1231 10:00:14.001019 3750 solver.cpp:244] Train net output #0: loss = 2.35765 (* 1 = 2.35765 loss)
  1239. I1231 10:00:14.001024 3750 sgd_solver.cpp:106] Iteration 300, lr = 0.01
  1240. I1231 10:00:16.629060 3750 solver.cpp:228] Iteration 320, loss = 2.40547
  1241. I1231 10:00:16.629081 3750 solver.cpp:244] Train net output #0: loss = 2.40547 (* 1 = 2.40547 loss)
  1242. I1231 10:00:16.629086 3750 sgd_solver.cpp:106] Iteration 320, lr = 0.01
  1243. I1231 10:00:19.260450 3750 solver.cpp:228] Iteration 340, loss = 2.35476
  1244. I1231 10:00:19.260473 3750 solver.cpp:244] Train net output #0: loss = 2.35476 (* 1 = 2.35476 loss)
  1245. I1231 10:00:19.260478 3750 sgd_solver.cpp:106] Iteration 340, lr = 0.01
  1246. I1231 10:00:21.892527 3750 solver.cpp:228] Iteration 360, loss = 2.38624
  1247. I1231 10:00:21.892549 3750 solver.cpp:244] Train net output #0: loss = 2.38624 (* 1 = 2.38624 loss)
  1248. I1231 10:00:21.892554 3750 sgd_solver.cpp:106] Iteration 360, lr = 0.01
  1249. I1231 10:00:24.523481 3750 solver.cpp:228] Iteration 380, loss = 2.28768
  1250. I1231 10:00:24.523502 3750 solver.cpp:244] Train net output #0: loss = 2.28768 (* 1 = 2.28768 loss)
  1251. I1231 10:00:24.523507 3750 sgd_solver.cpp:106] Iteration 380, lr = 0.01
  1252. I1231 10:00:27.156440 3750 solver.cpp:228] Iteration 400, loss = 2.28866
  1253. I1231 10:00:27.156461 3750 solver.cpp:244] Train net output #0: loss = 2.28866 (* 1 = 2.28866 loss)
  1254. I1231 10:00:27.156466 3750 sgd_solver.cpp:106] Iteration 400, lr = 0.01
  1255. I1231 10:00:29.784386 3750 solver.cpp:228] Iteration 420, loss = 2.34239
  1256. I1231 10:00:29.784409 3750 solver.cpp:244] Train net output #0: loss = 2.34239 (* 1 = 2.34239 loss)
  1257. I1231 10:00:29.784415 3750 sgd_solver.cpp:106] Iteration 420, lr = 0.01
  1258. I1231 10:00:32.433159 3750 solver.cpp:228] Iteration 440, loss = 2.31338
  1259. I1231 10:00:32.433181 3750 solver.cpp:244] Train net output #0: loss = 2.31338 (* 1 = 2.31338 loss)
  1260. I1231 10:00:32.433185 3750 sgd_solver.cpp:106] Iteration 440, lr = 0.01
  1261. I1231 10:00:35.096091 3750 solver.cpp:228] Iteration 460, loss = 2.33569
  1262. I1231 10:00:35.096112 3750 solver.cpp:244] Train net output #0: loss = 2.33569 (* 1 = 2.33569 loss)
  1263. I1231 10:00:35.096115 3750 sgd_solver.cpp:106] Iteration 460, lr = 0.01
  1264. I1231 10:00:37.747421 3750 solver.cpp:228] Iteration 480, loss = 2.31552
  1265. I1231 10:00:37.747535 3750 solver.cpp:244] Train net output #0: loss = 2.31552 (* 1 = 2.31552 loss)
  1266. I1231 10:00:37.747541 3750 sgd_solver.cpp:106] Iteration 480, lr = 0.01
  1267. I1231 10:00:40.380364 3750 solver.cpp:228] Iteration 500, loss = 2.34981
  1268. I1231 10:00:40.380388 3750 solver.cpp:244] Train net output #0: loss = 2.34981 (* 1 = 2.34981 loss)
  1269. I1231 10:00:40.380393 3750 sgd_solver.cpp:106] Iteration 500, lr = 0.01
  1270. I1231 10:00:43.006512 3750 solver.cpp:228] Iteration 520, loss = 2.34311
  1271. I1231 10:00:43.006536 3750 solver.cpp:244] Train net output #0: loss = 2.34311 (* 1 = 2.34311 loss)
  1272. I1231 10:00:43.006541 3750 sgd_solver.cpp:106] Iteration 520, lr = 0.01
  1273. I1231 10:00:45.633585 3750 solver.cpp:228] Iteration 540, loss = 2.35487
  1274. I1231 10:00:45.633607 3750 solver.cpp:244] Train net output #0: loss = 2.35487 (* 1 = 2.35487 loss)
  1275. I1231 10:00:45.633612 3750 sgd_solver.cpp:106] Iteration 540, lr = 0.01
  1276. I1231 10:00:48.272758 3750 solver.cpp:228] Iteration 560, loss = 2.32368
  1277. I1231 10:00:48.272781 3750 solver.cpp:244] Train net output #0: loss = 2.32368 (* 1 = 2.32368 loss)
  1278. I1231 10:00:48.272785 3750 sgd_solver.cpp:106] Iteration 560, lr = 0.01
  1279. I1231 10:00:50.903023 3750 solver.cpp:228] Iteration 580, loss = 2.35437
  1280. I1231 10:00:50.903046 3750 solver.cpp:244] Train net output #0: loss = 2.35437 (* 1 = 2.35437 loss)
  1281. I1231 10:00:50.903051 3750 sgd_solver.cpp:106] Iteration 580, lr = 0.01
  1282. I1231 10:00:53.531841 3750 solver.cpp:228] Iteration 600, loss = 2.3758
  1283. I1231 10:00:53.531873 3750 solver.cpp:244] Train net output #0: loss = 2.3758 (* 1 = 2.3758 loss)
  1284. I1231 10:00:53.531880 3750 sgd_solver.cpp:106] Iteration 600, lr = 0.01
  1285. I1231 10:00:56.163364 3750 solver.cpp:228] Iteration 620, loss = 2.33631
  1286. I1231 10:00:56.163386 3750 solver.cpp:244] Train net output #0: loss = 2.33631 (* 1 = 2.33631 loss)
  1287. I1231 10:00:56.163391 3750 sgd_solver.cpp:106] Iteration 620, lr = 0.01
  1288. I1231 10:00:58.801231 3750 solver.cpp:228] Iteration 640, loss = 2.35923
  1289. I1231 10:00:58.801256 3750 solver.cpp:244] Train net output #0: loss = 2.35923 (* 1 = 2.35923 loss)
  1290. I1231 10:00:58.801261 3750 sgd_solver.cpp:106] Iteration 640, lr = 0.01
  1291. I1231 10:01:01.437595 3750 solver.cpp:228] Iteration 660, loss = 2.35392
  1292. I1231 10:01:01.437618 3750 solver.cpp:244] Train net output #0: loss = 2.35392 (* 1 = 2.35392 loss)
  1293. I1231 10:01:01.437621 3750 sgd_solver.cpp:106] Iteration 660, lr = 0.01
  1294. I1231 10:01:04.066489 3750 solver.cpp:228] Iteration 680, loss = 2.35177
  1295. I1231 10:01:04.066514 3750 solver.cpp:244] Train net output #0: loss = 2.35177 (* 1 = 2.35177 loss)
  1296. I1231 10:01:04.066519 3750 sgd_solver.cpp:106] Iteration 680, lr = 0.01
  1297. I1231 10:01:06.700254 3750 solver.cpp:228] Iteration 700, loss = 2.38331
  1298. I1231 10:01:06.700275 3750 solver.cpp:244] Train net output #0: loss = 2.38331 (* 1 = 2.38331 loss)
  1299. I1231 10:01:06.700281 3750 sgd_solver.cpp:106] Iteration 700, lr = 0.01
  1300. I1231 10:01:09.336349 3750 solver.cpp:228] Iteration 720, loss = 2.27814
  1301. I1231 10:01:09.336441 3750 solver.cpp:244] Train net output #0: loss = 2.27814 (* 1 = 2.27814 loss)
  1302. I1231 10:01:09.336447 3750 sgd_solver.cpp:106] Iteration 720, lr = 0.01
  1303. I1231 10:01:11.968747 3750 solver.cpp:228] Iteration 740, loss = 2.33237
  1304. I1231 10:01:11.968767 3750 solver.cpp:244] Train net output #0: loss = 2.33237 (* 1 = 2.33237 loss)
  1305. I1231 10:01:11.968770 3750 sgd_solver.cpp:106] Iteration 740, lr = 0.01
  1306. I1231 10:01:14.601928 3750 solver.cpp:228] Iteration 760, loss = 2.32255
  1307. I1231 10:01:14.601950 3750 solver.cpp:244] Train net output #0: loss = 2.32255 (* 1 = 2.32255 loss)
  1308. I1231 10:01:14.601955 3750 sgd_solver.cpp:106] Iteration 760, lr = 0.01
  1309. I1231 10:01:17.236897 3750 solver.cpp:228] Iteration 780, loss = 2.39546
  1310. I1231 10:01:17.236919 3750 solver.cpp:244] Train net output #0: loss = 2.39546 (* 1 = 2.39546 loss)
  1311. I1231 10:01:17.236924 3750 sgd_solver.cpp:106] Iteration 780, lr = 0.01
  1312. I1231 10:01:19.873028 3750 solver.cpp:228] Iteration 800, loss = 2.34442
  1313. I1231 10:01:19.873051 3750 solver.cpp:244] Train net output #0: loss = 2.34442 (* 1 = 2.34442 loss)
  1314. I1231 10:01:19.873056 3750 sgd_solver.cpp:106] Iteration 800, lr = 0.01
  1315. I1231 10:01:22.513250 3750 solver.cpp:228] Iteration 820, loss = 2.35183
  1316. I1231 10:01:22.513273 3750 solver.cpp:244] Train net output #0: loss = 2.35183 (* 1 = 2.35183 loss)
  1317. I1231 10:01:22.513278 3750 sgd_solver.cpp:106] Iteration 820, lr = 0.01
  1318. I1231 10:01:25.159651 3750 solver.cpp:228] Iteration 840, loss = 2.31314
  1319. I1231 10:01:25.159672 3750 solver.cpp:244] Train net output #0: loss = 2.31314 (* 1 = 2.31314 loss)
  1320. I1231 10:01:25.159677 3750 sgd_solver.cpp:106] Iteration 840, lr = 0.01
  1321. I1231 10:01:27.799268 3750 solver.cpp:228] Iteration 860, loss = 2.34888
  1322. I1231 10:01:27.799293 3750 solver.cpp:244] Train net output #0: loss = 2.34888 (* 1 = 2.34888 loss)
  1323. I1231 10:01:27.799298 3750 sgd_solver.cpp:106] Iteration 860, lr = 0.01
  1324. I1231 10:01:30.440199 3750 solver.cpp:228] Iteration 880, loss = 2.35703
  1325. I1231 10:01:30.440222 3750 solver.cpp:244] Train net output #0: loss = 2.35703 (* 1 = 2.35703 loss)
  1326. I1231 10:01:30.440224 3750 sgd_solver.cpp:106] Iteration 880, lr = 0.01
  1327. I1231 10:01:33.077867 3750 solver.cpp:228] Iteration 900, loss = 2.27924
  1328. I1231 10:01:33.077890 3750 solver.cpp:244] Train net output #0: loss = 2.27924 (* 1 = 2.27924 loss)
  1329. I1231 10:01:33.077895 3750 sgd_solver.cpp:106] Iteration 900, lr = 0.01
  1330. I1231 10:01:35.712937 3750 solver.cpp:228] Iteration 920, loss = 2.32514
  1331. I1231 10:01:35.712960 3750 solver.cpp:244] Train net output #0: loss = 2.32514 (* 1 = 2.32514 loss)
  1332. I1231 10:01:35.712965 3750 sgd_solver.cpp:106] Iteration 920, lr = 0.01
  1333. I1231 10:01:38.350399 3750 solver.cpp:228] Iteration 940, loss = 2.29946
  1334. I1231 10:01:38.350424 3750 solver.cpp:244] Train net output #0: loss = 2.29946 (* 1 = 2.29946 loss)
  1335. I1231 10:01:38.350427 3750 sgd_solver.cpp:106] Iteration 940, lr = 0.01
  1336. I1231 10:01:40.988335 3750 solver.cpp:228] Iteration 960, loss = 2.34165
  1337. I1231 10:01:40.988406 3750 solver.cpp:244] Train net output #0: loss = 2.34165 (* 1 = 2.34165 loss)
  1338. I1231 10:01:40.988411 3750 sgd_solver.cpp:106] Iteration 960, lr = 0.01
  1339. I1231 10:01:43.627944 3750 solver.cpp:228] Iteration 980, loss = 2.33231
  1340. I1231 10:01:43.627971 3750 solver.cpp:244] Train net output #0: loss = 2.33231 (* 1 = 2.33231 loss)
  1341. I1231 10:01:43.627979 3750 sgd_solver.cpp:106] Iteration 980, lr = 0.01
  1342. I1231 10:01:46.134196 3750 solver.cpp:454] Snapshotting to binary proto file models/bvlc_alexnet/caffe_alexnet_train_iter_1000.caffemodel
  1343. I1231 10:01:47.144084 3750 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/bvlc_alexnet/caffe_alexnet_train_iter_1000.solverstate
  1344. I1231 10:01:47.574143 3750 solver.cpp:337] Iteration 1000, Testing net (#0)
  1345. I1231 10:02:56.518460 3750 solver.cpp:404] Test net output #0: accuracy = 0.1028
  1346. I1231 10:02:56.518558 3750 solver.cpp:404] Test net output #1: loss = 2.31395 (* 1 = 2.31395 loss)
  1347. I1231 10:02:56.601646 3750 solver.cpp:228] Iteration 1000, loss = 2.34483
  1348. I1231 10:02:56.601668 3750 solver.cpp:244] Train net output #0: loss = 2.34483 (* 1 = 2.34483 loss)
  1349. I1231 10:02:56.601673 3750 sgd_solver.cpp:106] Iteration 1000, lr = 0.001
  1350. I1231 10:02:59.287474 3750 solver.cpp:228] Iteration 1020, loss = 2.37257
  1351. I1231 10:02:59.287495 3750 solver.cpp:244] Train net output #0: loss = 2.37257 (* 1 = 2.37257 loss)
  1352. I1231 10:02:59.287499 3750 sgd_solver.cpp:106] Iteration 1020, lr = 0.001
  1353. I1231 10:03:01.971499 3750 solver.cpp:228] Iteration 1040, loss = 2.3016
  1354. I1231 10:03:01.971521 3750 solver.cpp:244] Train net output #0: loss = 2.3016 (* 1 = 2.3016 loss)
  1355. I1231 10:03:01.971525 3750 sgd_solver.cpp:106] Iteration 1040, lr = 0.001
  1356. I1231 10:03:04.606308 3750 solver.cpp:228] Iteration 1060, loss = 2.354
  1357. I1231 10:03:04.606333 3750 solver.cpp:244] Train net output #0: loss = 2.354 (* 1 = 2.354 loss)
  1358. I1231 10:03:04.606336 3750 sgd_solver.cpp:106] Iteration 1060, lr = 0.001
  1359. I1231 10:03:07.306514 3750 solver.cpp:228] Iteration 1080, loss = 2.34248
  1360. I1231 10:03:07.306541 3750 solver.cpp:244] Train net output #0: loss = 2.34248 (* 1 = 2.34248 loss)
  1361. I1231 10:03:07.306547 3750 sgd_solver.cpp:106] Iteration 1080, lr = 0.001
  1362. I1231 10:03:10.036783 3750 solver.cpp:228] Iteration 1100, loss = 2.30048
  1363. I1231 10:03:10.036805 3750 solver.cpp:244] Train net output #0: loss = 2.30048 (* 1 = 2.30048 loss)
  1364. I1231 10:03:10.036809 3750 sgd_solver.cpp:106] Iteration 1100, lr = 0.001
  1365. I1231 10:03:12.784339 3750 solver.cpp:228] Iteration 1120, loss = 2.34276
  1366. I1231 10:03:12.784364 3750 solver.cpp:244] Train net output #0: loss = 2.34276 (* 1 = 2.34276 loss)
  1367. I1231 10:03:12.784370 3750 sgd_solver.cpp:106] Iteration 1120, lr = 0.001
  1368. I1231 10:03:15.536545 3750 solver.cpp:228] Iteration 1140, loss = 2.31621
  1369. I1231 10:03:15.536567 3750 solver.cpp:244] Train net output #0: loss = 2.31621 (* 1 = 2.31621 loss)
  1370. I1231 10:03:15.536572 3750 sgd_solver.cpp:106] Iteration 1140, lr = 0.001
  1371. I1231 10:03:18.267773 3750 solver.cpp:228] Iteration 1160, loss = 2.31255
  1372. I1231 10:03:18.267798 3750 solver.cpp:244] Train net output #0: loss = 2.31255 (* 1 = 2.31255 loss)
  1373. I1231 10:03:18.267802 3750 sgd_solver.cpp:106] Iteration 1160, lr = 0.001
  1374. I1231 10:03:21.012823 3750 solver.cpp:228] Iteration 1180, loss = 2.30417
  1375. I1231 10:03:21.012845 3750 solver.cpp:244] Train net output #0: loss = 2.30417 (* 1 = 2.30417 loss)
  1376. I1231 10:03:21.012851 3750 sgd_solver.cpp:106] Iteration 1180, lr = 0.001
  1377. I1231 10:03:23.759482 3750 solver.cpp:228] Iteration 1200, loss = 2.36573
  1378. I1231 10:03:23.759505 3750 solver.cpp:244] Train net output #0: loss = 2.36573 (* 1 = 2.36573 loss)
  1379. I1231 10:03:23.759510 3750 sgd_solver.cpp:106] Iteration 1200, lr = 0.001
  1380. I1231 10:03:26.480738 3750 solver.cpp:228] Iteration 1220, loss = 2.33764
  1381. I1231 10:03:26.480762 3750 solver.cpp:244] Train net output #0: loss = 2.33764 (* 1 = 2.33764 loss)
  1382. I1231 10:03:26.480765 3750 sgd_solver.cpp:106] Iteration 1220, lr = 0.001
  1383. I1231 10:03:29.201123 3750 solver.cpp:228] Iteration 1240, loss = 2.30219
  1384. I1231 10:03:29.201222 3750 solver.cpp:244] Train net output #0: loss = 2.30219 (* 1 = 2.30219 loss)
  1385. I1231 10:03:29.201227 3750 sgd_solver.cpp:106] Iteration 1240, lr = 0.001
  1386. I1231 10:03:31.956243 3750 solver.cpp:228] Iteration 1260, loss = 2.28197
  1387. I1231 10:03:31.956265 3750 solver.cpp:244] Train net output #0: loss = 2.28197 (* 1 = 2.28197 loss)
  1388. I1231 10:03:31.956269 3750 sgd_solver.cpp:106] Iteration 1260, lr = 0.001
  1389. I1231 10:03:34.732066 3750 solver.cpp:228] Iteration 1280, loss = 2.28547
  1390. I1231 10:03:34.732089 3750 solver.cpp:244] Train net output #0: loss = 2.28547 (* 1 = 2.28547 loss)
  1391. I1231 10:03:34.732092 3750 sgd_solver.cpp:106] Iteration 1280, lr = 0.001
  1392. I1231 10:03:37.533429 3750 solver.cpp:228] Iteration 1300, loss = 2.30908
  1393. I1231 10:03:37.533452 3750 solver.cpp:244] Train net output #0: loss = 2.30908 (* 1 = 2.30908 loss)
  1394. I1231 10:03:37.533457 3750 sgd_solver.cpp:106] Iteration 1300, lr = 0.001
  1395. I1231 10:03:40.328743 3750 solver.cpp:228] Iteration 1320, loss = 2.3286
  1396. I1231 10:03:40.328765 3750 solver.cpp:244] Train net output #0: loss = 2.3286 (* 1 = 2.3286 loss)
  1397. I1231 10:03:40.328770 3750 sgd_solver.cpp:106] Iteration 1320, lr = 0.001
  1398. I1231 10:03:43.134333 3750 solver.cpp:228] Iteration 1340, loss = 2.30516
  1399. I1231 10:03:43.134354 3750 solver.cpp:244] Train net output #0: loss = 2.30516 (* 1 = 2.30516 loss)
  1400. I1231 10:03:43.134359 3750 sgd_solver.cpp:106] Iteration 1340, lr = 0.001
  1401. I1231 10:03:45.942070 3750 solver.cpp:228] Iteration 1360, loss = 2.3061
  1402. I1231 10:03:45.942091 3750 solver.cpp:244] Train net output #0: loss = 2.3061 (* 1 = 2.3061 loss)
  1403. I1231 10:03:45.942096 3750 sgd_solver.cpp:106] Iteration 1360, lr = 0.001
  1404. I1231 10:03:48.709724 3750 solver.cpp:228] Iteration 1380, loss = 2.28462
  1405. I1231 10:03:48.709746 3750 solver.cpp:244] Train net output #0: loss = 2.28462 (* 1 = 2.28462 loss)
  1406. I1231 10:03:48.709750 3750 sgd_solver.cpp:106] Iteration 1380, lr = 0.001
  1407. I1231 10:03:51.422637 3750 solver.cpp:228] Iteration 1400, loss = 2.35386
  1408. I1231 10:03:51.422660 3750 solver.cpp:244] Train net output #0: loss = 2.35386 (* 1 = 2.35386 loss)
  1409. I1231 10:03:51.422664 3750 sgd_solver.cpp:106] Iteration 1400, lr = 0.001
  1410. I1231 10:03:54.144119 3750 solver.cpp:228] Iteration 1420, loss = 2.31562
  1411. I1231 10:03:54.144141 3750 solver.cpp:244] Train net output #0: loss = 2.31562 (* 1 = 2.31562 loss)
  1412. I1231 10:03:54.144145 3750 sgd_solver.cpp:106] Iteration 1420, lr = 0.001
  1413. I1231 10:03:56.881331 3750 solver.cpp:228] Iteration 1440, loss = 2.33559
  1414. I1231 10:03:56.881353 3750 solver.cpp:244] Train net output #0: loss = 2.33559 (* 1 = 2.33559 loss)
  1415. I1231 10:03:56.881358 3750 sgd_solver.cpp:106] Iteration 1440, lr = 0.001
  1416. I1231 10:03:59.589541 3750 solver.cpp:228] Iteration 1460, loss = 2.28929
  1417. I1231 10:03:59.589658 3750 solver.cpp:244] Train net output #0: loss = 2.28929 (* 1 = 2.28929 loss)
  1418. I1231 10:03:59.589663 3750 sgd_solver.cpp:106] Iteration 1460, lr = 0.001
  1419. I1231 10:04:02.302563 3750 solver.cpp:228] Iteration 1480, loss = 2.30321
  1420. I1231 10:04:02.302585 3750 solver.cpp:244] Train net output #0: loss = 2.30321 (* 1 = 2.30321 loss)
  1421. I1231 10:04:02.302588 3750 sgd_solver.cpp:106] Iteration 1480, lr = 0.001
  1422. I1231 10:04:04.999634 3750 solver.cpp:228] Iteration 1500, loss = 2.31914
  1423. I1231 10:04:04.999655 3750 solver.cpp:244] Train net output #0: loss = 2.31914 (* 1 = 2.31914 loss)
  1424. I1231 10:04:04.999660 3750 sgd_solver.cpp:106] Iteration 1500, lr = 0.001
  1425. I1231 10:04:07.734257 3750 solver.cpp:228] Iteration 1520, loss = 2.36144
  1426. I1231 10:04:07.734282 3750 solver.cpp:244] Train net output #0: loss = 2.36144 (* 1 = 2.36144 loss)
  1427. I1231 10:04:07.734287 3750 sgd_solver.cpp:106] Iteration 1520, lr = 0.001
  1428. I1231 10:04:10.445163 3750 solver.cpp:228] Iteration 1540, loss = 2.36364
  1429. I1231 10:04:10.445186 3750 solver.cpp:244] Train net output #0: loss = 2.36364 (* 1 = 2.36364 loss)
  1430. I1231 10:04:10.445190 3750 sgd_solver.cpp:106] Iteration 1540, lr = 0.001
  1431. I1231 10:04:13.147294 3750 solver.cpp:228] Iteration 1560, loss = 2.30884
  1432. I1231 10:04:13.147315 3750 solver.cpp:244] Train net output #0: loss = 2.30884 (* 1 = 2.30884 loss)
  1433. I1231 10:04:13.147320 3750 sgd_solver.cpp:106] Iteration 1560, lr = 0.001
  1434. I1231 10:04:15.847362 3750 solver.cpp:228] Iteration 1580, loss = 2.33459
  1435. I1231 10:04:15.847383 3750 solver.cpp:244] Train net output #0: loss = 2.33459 (* 1 = 2.33459 loss)
  1436. I1231 10:04:15.847386 3750 sgd_solver.cpp:106] Iteration 1580, lr = 0.001
  1437. I1231 10:04:18.543267 3750 solver.cpp:228] Iteration 1600, loss = 2.31264
  1438. I1231 10:04:18.543288 3750 solver.cpp:244] Train net output #0: loss = 2.31264 (* 1 = 2.31264 loss)
  1439. I1231 10:04:18.543293 3750 sgd_solver.cpp:106] Iteration 1600, lr = 0.001
  1440. I1231 10:04:21.242331 3750 solver.cpp:228] Iteration 1620, loss = 2.35213
  1441. I1231 10:04:21.242353 3750 solver.cpp:244] Train net output #0: loss = 2.35213 (* 1 = 2.35213 loss)
  1442. I1231 10:04:21.242357 3750 sgd_solver.cpp:106] Iteration 1620, lr = 0.001
  1443. I1231 10:04:23.938932 3750 solver.cpp:228] Iteration 1640, loss = 2.30634
  1444. I1231 10:04:23.938957 3750 solver.cpp:244] Train net output #0: loss = 2.30634 (* 1 = 2.30634 loss)
  1445. I1231 10:04:23.938963 3750 sgd_solver.cpp:106] Iteration 1640, lr = 0.001
  1446. I1231 10:04:26.635504 3750 solver.cpp:228] Iteration 1660, loss = 2.2856
  1447. I1231 10:04:26.635525 3750 solver.cpp:244] Train net output #0: loss = 2.2856 (* 1 = 2.2856 loss)
  1448. I1231 10:04:26.635529 3750 sgd_solver.cpp:106] Iteration 1660, lr = 0.001
  1449. I1231 10:04:29.332518 3750 solver.cpp:228] Iteration 1680, loss = 2.27774
  1450. I1231 10:04:29.332541 3750 solver.cpp:244] Train net output #0: loss = 2.27774 (* 1 = 2.27774 loss)
  1451. I1231 10:04:29.332545 3750 sgd_solver.cpp:106] Iteration 1680, lr = 0.001
  1452. I1231 10:04:32.028904 3750 solver.cpp:228] Iteration 1700, loss = 2.34021
  1453. I1231 10:04:32.029019 3750 solver.cpp:244] Train net output #0: loss = 2.34021 (* 1 = 2.34021 loss)
  1454. I1231 10:04:32.029024 3750 sgd_solver.cpp:106] Iteration 1700, lr = 0.001
  1455. I1231 10:04:34.727354 3750 solver.cpp:228] Iteration 1720, loss = 2.34049
  1456. I1231 10:04:34.727377 3750 solver.cpp:244] Train net output #0: loss = 2.34049 (* 1 = 2.34049 loss)
  1457. I1231 10:04:34.727382 3750 sgd_solver.cpp:106] Iteration 1720, lr = 0.001
  1458. I1231 10:04:37.425602 3750 solver.cpp:228] Iteration 1740, loss = 2.32911
  1459. I1231 10:04:37.425624 3750 solver.cpp:244] Train net output #0: loss = 2.32911 (* 1 = 2.32911 loss)
  1460. I1231 10:04:37.425629 3750 sgd_solver.cpp:106] Iteration 1740, lr = 0.001
  1461. I1231 10:04:40.125151 3750 solver.cpp:228] Iteration 1760, loss = 2.31598
  1462. I1231 10:04:40.125174 3750 solver.cpp:244] Train net output #0: loss = 2.31598 (* 1 = 2.31598 loss)
  1463. I1231 10:04:40.125177 3750 sgd_solver.cpp:106] Iteration 1760, lr = 0.001
  1464. I1231 10:04:42.824857 3750 solver.cpp:228] Iteration 1780, loss = 2.33156
  1465. I1231 10:04:42.824879 3750 solver.cpp:244] Train net output #0: loss = 2.33156 (* 1 = 2.33156 loss)
  1466. I1231 10:04:42.824883 3750 sgd_solver.cpp:106] Iteration 1780, lr = 0.001
  1467. I1231 10:04:45.522495 3750 solver.cpp:228] Iteration 1800, loss = 2.34056
  1468. I1231 10:04:45.522517 3750 solver.cpp:244] Train net output #0: loss = 2.34056 (* 1 = 2.34056 loss)
  1469. I1231 10:04:45.522521 3750 sgd_solver.cpp:106] Iteration 1800, lr = 0.001
  1470. I1231 10:04:48.254500 3750 solver.cpp:228] Iteration 1820, loss = 2.30518
  1471. I1231 10:04:48.254523 3750 solver.cpp:244] Train net output #0: loss = 2.30518 (* 1 = 2.30518 loss)
  1472. I1231 10:04:48.254528 3750 sgd_solver.cpp:106] Iteration 1820, lr = 0.001
  1473. I1231 10:04:50.992200 3750 solver.cpp:228] Iteration 1840, loss = 2.32255
  1474. I1231 10:04:50.992223 3750 solver.cpp:244] Train net output #0: loss = 2.32255 (* 1 = 2.32255 loss)
  1475. I1231 10:04:50.992226 3750 sgd_solver.cpp:106] Iteration 1840, lr = 0.001
  1476. I1231 10:04:53.703897 3750 solver.cpp:228] Iteration 1860, loss = 2.30668
  1477. I1231 10:04:53.703946 3750 solver.cpp:244] Train net output #0: loss = 2.30668 (* 1 = 2.30668 loss)
  1478. I1231 10:04:53.703953 3750 sgd_solver.cpp:106] Iteration 1860, lr = 0.001
  1479. I1231 10:04:56.405320 3750 solver.cpp:228] Iteration 1880, loss = 2.29324
  1480. I1231 10:04:56.405341 3750 solver.cpp:244] Train net output #0: loss = 2.29324 (* 1 = 2.29324 loss)
  1481. I1231 10:04:56.405346 3750 sgd_solver.cpp:106] Iteration 1880, lr = 0.001
  1482. I1231 10:04:59.104897 3750 solver.cpp:228] Iteration 1900, loss = 2.29406
  1483. I1231 10:04:59.104918 3750 solver.cpp:244] Train net output #0: loss = 2.29406 (* 1 = 2.29406 loss)
  1484. I1231 10:04:59.104923 3750 sgd_solver.cpp:106] Iteration 1900, lr = 0.001
  1485. I1231 10:05:01.806949 3750 solver.cpp:228] Iteration 1920, loss = 2.30289
  1486. I1231 10:05:01.806970 3750 solver.cpp:244] Train net output #0: loss = 2.30289 (* 1 = 2.30289 loss)
  1487. I1231 10:05:01.806974 3750 sgd_solver.cpp:106] Iteration 1920, lr = 0.001
  1488. I1231 10:05:04.504257 3750 solver.cpp:228] Iteration 1940, loss = 2.31311
  1489. I1231 10:05:04.504365 3750 solver.cpp:244] Train net output #0: loss = 2.31311 (* 1 = 2.31311 loss)
  1490. I1231 10:05:04.504372 3750 sgd_solver.cpp:106] Iteration 1940, lr = 0.001
  1491. I1231 10:05:07.205976 3750 solver.cpp:228] Iteration 1960, loss = 2.30893
  1492. I1231 10:05:07.205998 3750 solver.cpp:244] Train net output #0: loss = 2.30893 (* 1 = 2.30893 loss)
  1493. I1231 10:05:07.206003 3750 sgd_solver.cpp:106] Iteration 1960, lr = 0.001
  1494. I1231 10:05:09.903568 3750 solver.cpp:228] Iteration 1980, loss = 2.2799
  1495. I1231 10:05:09.903589 3750 solver.cpp:244] Train net output #0: loss = 2.2799 (* 1 = 2.2799 loss)
  1496. I1231 10:05:09.903594 3750 sgd_solver.cpp:106] Iteration 1980, lr = 0.001
  1497. I1231 10:05:12.469777 3750 solver.cpp:454] Snapshotting to binary proto file models/bvlc_alexnet/caffe_alexnet_train_iter_2000.caffemodel
  1498. I1231 10:05:13.320540 3750 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/bvlc_alexnet/caffe_alexnet_train_iter_2000.solverstate
  1499. I1231 10:05:13.880272 3750 solver.cpp:317] Iteration 2000, loss = 2.34475
  1500. I1231 10:05:13.880293 3750 solver.cpp:337] Iteration 2000, Testing net (#0)
  1501. I1231 10:06:23.795788 3750 solver.cpp:404] Test net output #0: accuracy = 0.1135
  1502. I1231 10:06:23.795878 3750 solver.cpp:404] Test net output #1: loss = 2.3047 (* 1 = 2.3047 loss)
  1503. I1231 10:06:23.795884 3750 solver.cpp:322] Optimization Done.
  1504. I1231 10:06:23.795887 3750 caffe.cpp:254] Optimization Done.
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!