API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Aug 4th, 2017
212
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 57.01 KB | None | 0 0
raw download clone embed print report
  1. sina@sina-Z170X-UD3:~/code/caffe$ ./build/tools/caffe train --solver=examples/alexnetfinetune/solver_sina.prototxt
  2. I0805 02:17:47.682703 3354 caffe.cpp:218] Using GPUs 0
  3. I0805 02:17:47.721066 3354 caffe.cpp:223] GPU 0: GeForce GTX 1070
  4. I0805 02:17:47.932503 3354 solver.cpp:44] Initializing solver from parameters:
  5. test_iter: 54
  6. test_interval: 610
  7. base_lr: 0.001
  8. display: 20
  9. max_iter: 5000
  10. lr_policy: "step"
  11. gamma: 0.1
  12. momentum: 0.9
  13. weight_decay: 0.0005
  14. stepsize: 100000
  15. snapshot: 4999
  16. snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_sinatrain"
  17. solver_mode: GPU
  18. device_id: 0
  19. net: "examples/alexnetfinetune/alexnetsade-bn.prototxt"
  20. train_state {
  21. level: 0
  22. stage: ""
  23. }
  24. type: "SGD"
  25. I0805 02:17:47.932651 3354 solver.cpp:87] Creating training net from net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  26. I0805 02:17:47.932893 3354 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  27. I0805 02:17:47.932900 3354 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  28. I0805 02:17:47.932983 3354 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
  29. I0805 02:17:47.932994 3354 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
  30. I0805 02:17:47.933147 3354 net.cpp:53] Initializing net from parameters:
  31. name: "AlexNet"
  32. state {
  33. phase: TRAIN
  34. level: 0
  35. stage: ""
  36. }
  37. layer {
  38. name: "data"
  39. type: "Data"
  40. top: "data"
  41. top: "label"
  42. include {
  43. phase: TRAIN
  44. }
  45. data_param {
  46. source: "examples/Mydataset_train_lmdb"
  47. batch_size: 256
  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: 11
  67. stride: 4
  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: 2
  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: 5
  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: "bn1"
  136. type: "BatchNorm"
  137. bottom: "conv2"
  138. top: "conv2"
  139. param {
  140. lr_mult: 0
  141. decay_mult: 0
  142. }
  143. param {
  144. lr_mult: 0
  145. decay_mult: 0
  146. }
  147. param {
  148. lr_mult: 0
  149. decay_mult: 0
  150. }
  151. }
  152. layer {
  153. name: "scale1"
  154. type: "Scale"
  155. bottom: "conv2"
  156. top: "conv2"
  157. scale_param {
  158. bias_term: true
  159. }
  160. }
  161. layer {
  162. name: "relu2"
  163. type: "ReLU"
  164. bottom: "conv2"
  165. top: "conv2"
  166. }
  167. layer {
  168. name: "norm2"
  169. type: "LRN"
  170. bottom: "conv2"
  171. top: "norm2"
  172. lrn_param {
  173. local_size: 5
  174. alpha: 0.0001
  175. beta: 0.75
  176. }
  177. }
  178. layer {
  179. name: "pool2"
  180. type: "Pooling"
  181. bottom: "norm2"
  182. top: "pool2"
  183. pooling_param {
  184. pool: MAX
  185. kernel_size: 3
  186. stride: 2
  187. }
  188. }
  189. layer {
  190. name: "conv3"
  191. type: "Convolution"
  192. bottom: "pool2"
  193. top: "conv3"
  194. param {
  195. lr_mult: 1
  196. decay_mult: 1
  197. }
  198. param {
  199. lr_mult: 2
  200. decay_mult: 0
  201. }
  202. convolution_param {
  203. num_output: 384
  204. pad: 1
  205. kernel_size: 3
  206. weight_filler {
  207. type: "gaussian"
  208. std: 0.01
  209. }
  210. bias_filler {
  211. type: "constant"
  212. value: 0
  213. }
  214. }
  215. }
  216. layer {
  217. name: "relu3"
  218. type: "ReLU"
  219. bottom: "conv3"
  220. top: "conv3"
  221. }
  222. layer {
  223. name: "conv4"
  224. type: "Convolution"
  225. bottom: "conv3"
  226. top: "conv4"
  227. param {
  228. lr_mult: 1
  229. decay_mult: 1
  230. }
  231. param {
  232. lr_mult: 2
  233. decay_mult: 0
  234. }
  235. convolution_param {
  236. num_output: 384
  237. pad: 1
  238. kernel_size: 3
  239. group: 2
  240. weight_filler {
  241. type: "gaussian"
  242. std: 0.01
  243. }
  244. bias_filler {
  245. type: "constant"
  246. value: 0.1
  247. }
  248. }
  249. }
  250. layer {
  251. name: "relu4"
  252. type: "ReLU"
  253. bottom: "conv4"
  254. top: "conv4"
  255. }
  256. layer {
  257. name: "conv5"
  258. type: "Convolution"
  259. bottom: "conv4"
  260. top: "conv5"
  261. param {
  262. lr_mult: 1
  263. decay_mult: 1
  264. }
  265. param {
  266. lr_mult: 2
  267. decay_mult: 0
  268. }
  269. convolution_param {
  270. num_output: 256
  271. pad: 1
  272. kernel_size: 3
  273. group: 2
  274. weight_filler {
  275. type: "xavier"
  276. std: 0.01
  277. }
  278. bias_filler {
  279. type: "constant"
  280. value: 0.1
  281. }
  282. }
  283. }
  284. layer {
  285. name: "relu5"
  286. type: "ReLU"
  287. bottom: "conv5"
  288. top: "conv5"
  289. }
  290. layer {
  291. name: "pool5"
  292. type: "Pooling"
  293. bottom: "conv5"
  294. top: "pool5"
  295. pooling_param {
  296. pool: MAX
  297. kernel_size: 3
  298. stride: 2
  299. }
  300. }
  301. layer {
  302. name: "fc6"
  303. type: "InnerProduct"
  304. bottom: "pool5"
  305. top: "fc6"
  306. param {
  307. lr_mult: 1
  308. decay_mult: 1
  309. }
  310. param {
  311. lr_mult: 2
  312. decay_mult: 0
  313. }
  314. inner_product_param {
  315. num_output: 4096
  316. weight_filler {
  317. type: "xavier"
  318. std: 0.005
  319. }
  320. bias_filler {
  321. type: "constant"
  322. value: 0.1
  323. }
  324. }
  325. }
  326. layer {
  327. name: "relu6"
  328. type: "ReLU"
  329. bottom: "fc6"
  330. top: "fc6"
  331. }
  332. layer {
  333. name: "drop6"
  334. type: "Dropout"
  335. bottom: "fc6"
  336. top: "fc6"
  337. dropout_param {
  338. dropout_ratio: 0.5
  339. }
  340. }
  341. layer {
  342. name: "fc7"
  343. type: "InnerProduct"
  344. bottom: "fc6"
  345. top: "fc7"
  346. param {
  347. lr_mult: 1
  348. decay_mult: 1
  349. }
  350. param {
  351. lr_mult: 2
  352. decay_mult: 0
  353. }
  354. inner_product_param {
  355. num_output: 4096
  356. weight_filler {
  357. type: "xavier"
  358. std: 0.005
  359. }
  360. bias_filler {
  361. type: "constant"
  362. value: 0.1
  363. }
  364. }
  365. }
  366. layer {
  367. name: "relu7"
  368. type: "ReLU"
  369. bottom: "fc7"
  370. top: "fc7"
  371. }
  372. layer {
  373. name: "drop7"
  374. type: "Dropout"
  375. bottom: "fc7"
  376. top: "fc7"
  377. dropout_param {
  378. dropout_ratio: 0.5
  379. }
  380. }
  381. layer {
  382. name: "fc8"
  383. type: "InnerProduct"
  384. bottom: "fc7"
  385. top: "fc8"
  386. param {
  387. lr_mult: 1
  388. decay_mult: 1
  389. }
  390. param {
  391. lr_mult: 2
  392. decay_mult: 0
  393. }
  394. inner_product_param {
  395. num_output: 4
  396. weight_filler {
  397. type: "xavier"
  398. std: 0.01
  399. }
  400. bias_filler {
  401. type: "constant"
  402. value: 0
  403. }
  404. }
  405. }
  406. layer {
  407. name: "loss"
  408. type: "SoftmaxWithLoss"
  409. bottom: "fc8"
  410. bottom: "label"
  411. top: "loss"
  412. }
  413. I0805 02:17:47.933511 3354 layer_factory.hpp:77] Creating layer data
  414. I0805 02:17:47.933637 3354 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_train_lmdb
  415. I0805 02:17:47.933686 3354 net.cpp:86] Creating Layer data
  416. I0805 02:17:47.933693 3354 net.cpp:382] data -> data
  417. I0805 02:17:47.933733 3354 net.cpp:382] data -> label
  418. I0805 02:17:47.934499 3354 data_layer.cpp:45] output data size: 256,3,227,227
  419. I0805 02:17:48.112761 3354 net.cpp:124] Setting up data
  420. I0805 02:17:48.112792 3354 net.cpp:131] Top shape: 256 3 227 227 (39574272)
  421. I0805 02:17:48.112810 3354 net.cpp:131] Top shape: 256 (256)
  422. I0805 02:17:48.112812 3354 net.cpp:139] Memory required for data: 158298112
  423. I0805 02:17:48.112834 3354 layer_factory.hpp:77] Creating layer conv1
  424. I0805 02:17:48.112849 3354 net.cpp:86] Creating Layer conv1
  425. I0805 02:17:48.112855 3354 net.cpp:408] conv1 <- data
  426. I0805 02:17:48.112864 3354 net.cpp:382] conv1 -> conv1
  427. I0805 02:17:48.310364 3354 net.cpp:124] Setting up conv1
  428. I0805 02:17:48.310398 3354 net.cpp:131] Top shape: 256 96 55 55 (74342400)
  429. I0805 02:17:48.310401 3354 net.cpp:139] Memory required for data: 455667712
  430. I0805 02:17:48.310437 3354 layer_factory.hpp:77] Creating layer relu1
  431. I0805 02:17:48.310448 3354 net.cpp:86] Creating Layer relu1
  432. I0805 02:17:48.310451 3354 net.cpp:408] relu1 <- conv1
  433. I0805 02:17:48.310456 3354 net.cpp:369] relu1 -> conv1 (in-place)
  434. I0805 02:17:48.310626 3354 net.cpp:124] Setting up relu1
  435. I0805 02:17:48.310631 3354 net.cpp:131] Top shape: 256 96 55 55 (74342400)
  436. I0805 02:17:48.310648 3354 net.cpp:139] Memory required for data: 753037312
  437. I0805 02:17:48.310650 3354 layer_factory.hpp:77] Creating layer norm1
  438. I0805 02:17:48.310673 3354 net.cpp:86] Creating Layer norm1
  439. I0805 02:17:48.310676 3354 net.cpp:408] norm1 <- conv1
  440. I0805 02:17:48.310680 3354 net.cpp:382] norm1 -> norm1
  441. I0805 02:17:48.311266 3354 net.cpp:124] Setting up norm1
  442. I0805 02:17:48.311275 3354 net.cpp:131] Top shape: 256 96 55 55 (74342400)
  443. I0805 02:17:48.311290 3354 net.cpp:139] Memory required for data: 1050406912
  444. I0805 02:17:48.311293 3354 layer_factory.hpp:77] Creating layer pool1
  445. I0805 02:17:48.311314 3354 net.cpp:86] Creating Layer pool1
  446. I0805 02:17:48.311317 3354 net.cpp:408] pool1 <- norm1
  447. I0805 02:17:48.311322 3354 net.cpp:382] pool1 -> pool1
  448. I0805 02:17:48.311357 3354 net.cpp:124] Setting up pool1
  449. I0805 02:17:48.311375 3354 net.cpp:131] Top shape: 256 96 27 27 (17915904)
  450. I0805 02:17:48.311378 3354 net.cpp:139] Memory required for data: 1122070528
  451. I0805 02:17:48.311394 3354 layer_factory.hpp:77] Creating layer conv2
  452. I0805 02:17:48.311403 3354 net.cpp:86] Creating Layer conv2
  453. I0805 02:17:48.311406 3354 net.cpp:408] conv2 <- pool1
  454. I0805 02:17:48.311410 3354 net.cpp:382] conv2 -> conv2
  455. I0805 02:17:48.316013 3354 net.cpp:124] Setting up conv2
  456. I0805 02:17:48.316030 3354 net.cpp:131] Top shape: 256 256 27 27 (47775744)
  457. I0805 02:17:48.316047 3354 net.cpp:139] Memory required for data: 1313173504
  458. I0805 02:17:48.316058 3354 layer_factory.hpp:77] Creating layer bn1
  459. I0805 02:17:48.316066 3354 net.cpp:86] Creating Layer bn1
  460. I0805 02:17:48.316071 3354 net.cpp:408] bn1 <- conv2
  461. I0805 02:17:48.316076 3354 net.cpp:369] bn1 -> conv2 (in-place)
  462. I0805 02:17:48.316227 3354 net.cpp:124] Setting up bn1
  463. I0805 02:17:48.316232 3354 net.cpp:131] Top shape: 256 256 27 27 (47775744)
  464. I0805 02:17:48.316249 3354 net.cpp:139] Memory required for data: 1504276480
  465. I0805 02:17:48.316256 3354 layer_factory.hpp:77] Creating layer scale1
  466. I0805 02:17:48.316262 3354 net.cpp:86] Creating Layer scale1
  467. I0805 02:17:48.316265 3354 net.cpp:408] scale1 <- conv2
  468. I0805 02:17:48.316270 3354 net.cpp:369] scale1 -> conv2 (in-place)
  469. I0805 02:17:48.316314 3354 layer_factory.hpp:77] Creating layer scale1
  470. I0805 02:17:48.316419 3354 net.cpp:124] Setting up scale1
  471. I0805 02:17:48.316424 3354 net.cpp:131] Top shape: 256 256 27 27 (47775744)
  472. I0805 02:17:48.316442 3354 net.cpp:139] Memory required for data: 1695379456
  473. I0805 02:17:48.316448 3354 layer_factory.hpp:77] Creating layer relu2
  474. I0805 02:17:48.316454 3354 net.cpp:86] Creating Layer relu2
  475. I0805 02:17:48.316457 3354 net.cpp:408] relu2 <- conv2
  476. I0805 02:17:48.316462 3354 net.cpp:369] relu2 -> conv2 (in-place)
  477. I0805 02:17:48.318275 3354 net.cpp:124] Setting up relu2
  478. I0805 02:17:48.318284 3354 net.cpp:131] Top shape: 256 256 27 27 (47775744)
  479. I0805 02:17:48.318301 3354 net.cpp:139] Memory required for data: 1886482432
  480. I0805 02:17:48.318305 3354 layer_factory.hpp:77] Creating layer norm2
  481. I0805 02:17:48.318310 3354 net.cpp:86] Creating Layer norm2
  482. I0805 02:17:48.318313 3354 net.cpp:408] norm2 <- conv2
  483. I0805 02:17:48.318318 3354 net.cpp:382] norm2 -> norm2
  484. I0805 02:17:48.318470 3354 net.cpp:124] Setting up norm2
  485. I0805 02:17:48.318476 3354 net.cpp:131] Top shape: 256 256 27 27 (47775744)
  486. I0805 02:17:48.318493 3354 net.cpp:139] Memory required for data: 2077585408
  487. I0805 02:17:48.318496 3354 layer_factory.hpp:77] Creating layer pool2
  488. I0805 02:17:48.318501 3354 net.cpp:86] Creating Layer pool2
  489. I0805 02:17:48.318506 3354 net.cpp:408] pool2 <- norm2
  490. I0805 02:17:48.318509 3354 net.cpp:382] pool2 -> pool2
  491. I0805 02:17:48.318549 3354 net.cpp:124] Setting up pool2
  492. I0805 02:17:48.318553 3354 net.cpp:131] Top shape: 256 256 13 13 (11075584)
  493. I0805 02:17:48.318572 3354 net.cpp:139] Memory required for data: 2121887744
  494. I0805 02:17:48.318573 3354 layer_factory.hpp:77] Creating layer conv3
  495. I0805 02:17:48.318581 3354 net.cpp:86] Creating Layer conv3
  496. I0805 02:17:48.318584 3354 net.cpp:408] conv3 <- pool2
  497. I0805 02:17:48.318590 3354 net.cpp:382] conv3 -> conv3
  498. I0805 02:17:48.342455 3354 net.cpp:124] Setting up conv3
  499. I0805 02:17:48.342473 3354 net.cpp:131] Top shape: 256 384 13 13 (16613376)
  500. I0805 02:17:48.342475 3354 net.cpp:139] Memory required for data: 2188341248
  501. I0805 02:17:48.342497 3354 layer_factory.hpp:77] Creating layer relu3
  502. I0805 02:17:48.342519 3354 net.cpp:86] Creating Layer relu3
  503. I0805 02:17:48.342550 3354 net.cpp:408] relu3 <- conv3
  504. I0805 02:17:48.342571 3354 net.cpp:369] relu3 -> conv3 (in-place)
  505. I0805 02:17:48.342731 3354 net.cpp:124] Setting up relu3
  506. I0805 02:17:48.342737 3354 net.cpp:131] Top shape: 256 384 13 13 (16613376)
  507. I0805 02:17:48.342739 3354 net.cpp:139] Memory required for data: 2254794752
  508. I0805 02:17:48.342741 3354 layer_factory.hpp:77] Creating layer conv4
  509. I0805 02:17:48.342764 3354 net.cpp:86] Creating Layer conv4
  510. I0805 02:17:48.342767 3354 net.cpp:408] conv4 <- conv3
  511. I0805 02:17:48.342787 3354 net.cpp:382] conv4 -> conv4
  512. I0805 02:17:48.350215 3354 net.cpp:124] Setting up conv4
  513. I0805 02:17:48.350246 3354 net.cpp:131] Top shape: 256 384 13 13 (16613376)
  514. I0805 02:17:48.350248 3354 net.cpp:139] Memory required for data: 2321248256
  515. I0805 02:17:48.350270 3354 layer_factory.hpp:77] Creating layer relu4
  516. I0805 02:17:48.350281 3354 net.cpp:86] Creating Layer relu4
  517. I0805 02:17:48.350286 3354 net.cpp:408] relu4 <- conv4
  518. I0805 02:17:48.350291 3354 net.cpp:369] relu4 -> conv4 (in-place)
  519. I0805 02:17:48.350456 3354 net.cpp:124] Setting up relu4
  520. I0805 02:17:48.350462 3354 net.cpp:131] Top shape: 256 384 13 13 (16613376)
  521. I0805 02:17:48.350466 3354 net.cpp:139] Memory required for data: 2387701760
  522. I0805 02:17:48.350481 3354 layer_factory.hpp:77] Creating layer conv5
  523. I0805 02:17:48.350488 3354 net.cpp:86] Creating Layer conv5
  524. I0805 02:17:48.350507 3354 net.cpp:408] conv5 <- conv4
  525. I0805 02:17:48.350512 3354 net.cpp:382] conv5 -> conv5
  526. I0805 02:17:48.355535 3354 net.cpp:124] Setting up conv5
  527. I0805 02:17:48.355564 3354 net.cpp:131] Top shape: 256 256 13 13 (11075584)
  528. I0805 02:17:48.355567 3354 net.cpp:139] Memory required for data: 2432004096
  529. I0805 02:17:48.355590 3354 layer_factory.hpp:77] Creating layer relu5
  530. I0805 02:17:48.355604 3354 net.cpp:86] Creating Layer relu5
  531. I0805 02:17:48.355607 3354 net.cpp:408] relu5 <- conv5
  532. I0805 02:17:48.355624 3354 net.cpp:369] relu5 -> conv5 (in-place)
  533. I0805 02:17:48.355823 3354 net.cpp:124] Setting up relu5
  534. I0805 02:17:48.355829 3354 net.cpp:131] Top shape: 256 256 13 13 (11075584)
  535. I0805 02:17:48.355844 3354 net.cpp:139] Memory required for data: 2476306432
  536. I0805 02:17:48.355846 3354 layer_factory.hpp:77] Creating layer pool5
  537. I0805 02:17:48.355867 3354 net.cpp:86] Creating Layer pool5
  538. I0805 02:17:48.355870 3354 net.cpp:408] pool5 <- conv5
  539. I0805 02:17:48.355876 3354 net.cpp:382] pool5 -> pool5
  540. I0805 02:17:48.355913 3354 net.cpp:124] Setting up pool5
  541. I0805 02:17:48.355918 3354 net.cpp:131] Top shape: 256 256 6 6 (2359296)
  542. I0805 02:17:48.355921 3354 net.cpp:139] Memory required for data: 2485743616
  543. I0805 02:17:48.355922 3354 layer_factory.hpp:77] Creating layer fc6
  544. I0805 02:17:48.355927 3354 net.cpp:86] Creating Layer fc6
  545. I0805 02:17:48.355929 3354 net.cpp:408] fc6 <- pool5
  546. I0805 02:17:48.355934 3354 net.cpp:382] fc6 -> fc6
  547. I0805 02:17:48.538027 3354 net.cpp:124] Setting up fc6
  548. I0805 02:17:48.538043 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  549. I0805 02:17:48.538045 3354 net.cpp:139] Memory required for data: 2489937920
  550. I0805 02:17:48.538069 3354 layer_factory.hpp:77] Creating layer relu6
  551. I0805 02:17:48.538091 3354 net.cpp:86] Creating Layer relu6
  552. I0805 02:17:48.538095 3354 net.cpp:408] relu6 <- fc6
  553. I0805 02:17:48.538101 3354 net.cpp:369] relu6 -> fc6 (in-place)
  554. I0805 02:17:48.538276 3354 net.cpp:124] Setting up relu6
  555. I0805 02:17:48.538282 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  556. I0805 02:17:48.538285 3354 net.cpp:139] Memory required for data: 2494132224
  557. I0805 02:17:48.538287 3354 layer_factory.hpp:77] Creating layer drop6
  558. I0805 02:17:48.538292 3354 net.cpp:86] Creating Layer drop6
  559. I0805 02:17:48.538308 3354 net.cpp:408] drop6 <- fc6
  560. I0805 02:17:48.538311 3354 net.cpp:369] drop6 -> fc6 (in-place)
  561. I0805 02:17:48.538362 3354 net.cpp:124] Setting up drop6
  562. I0805 02:17:48.538378 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  563. I0805 02:17:48.538380 3354 net.cpp:139] Memory required for data: 2498326528
  564. I0805 02:17:48.538383 3354 layer_factory.hpp:77] Creating layer fc7
  565. I0805 02:17:48.538429 3354 net.cpp:86] Creating Layer fc7
  566. I0805 02:17:48.538444 3354 net.cpp:408] fc7 <- fc6
  567. I0805 02:17:48.538450 3354 net.cpp:382] fc7 -> fc7
  568. I0805 02:17:48.617386 3354 net.cpp:124] Setting up fc7
  569. I0805 02:17:48.617403 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  570. I0805 02:17:48.617406 3354 net.cpp:139] Memory required for data: 2502520832
  571. I0805 02:17:48.617426 3354 layer_factory.hpp:77] Creating layer relu7
  572. I0805 02:17:48.617432 3354 net.cpp:86] Creating Layer relu7
  573. I0805 02:17:48.617450 3354 net.cpp:408] relu7 <- fc7
  574. I0805 02:17:48.617456 3354 net.cpp:369] relu7 -> fc7 (in-place)
  575. I0805 02:17:48.618033 3354 net.cpp:124] Setting up relu7
  576. I0805 02:17:48.618041 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  577. I0805 02:17:48.618043 3354 net.cpp:139] Memory required for data: 2506715136
  578. I0805 02:17:48.618046 3354 layer_factory.hpp:77] Creating layer drop7
  579. I0805 02:17:48.618050 3354 net.cpp:86] Creating Layer drop7
  580. I0805 02:17:48.618052 3354 net.cpp:408] drop7 <- fc7
  581. I0805 02:17:48.618072 3354 net.cpp:369] drop7 -> fc7 (in-place)
  582. I0805 02:17:48.618116 3354 net.cpp:124] Setting up drop7
  583. I0805 02:17:48.618121 3354 net.cpp:131] Top shape: 256 4096 (1048576)
  584. I0805 02:17:48.618137 3354 net.cpp:139] Memory required for data: 2510909440
  585. I0805 02:17:48.618139 3354 layer_factory.hpp:77] Creating layer fc8
  586. I0805 02:17:48.618160 3354 net.cpp:86] Creating Layer fc8
  587. I0805 02:17:48.618162 3354 net.cpp:408] fc8 <- fc7
  588. I0805 02:17:48.618166 3354 net.cpp:382] fc8 -> fc8
  589. I0805 02:17:48.618891 3354 net.cpp:124] Setting up fc8
  590. I0805 02:17:48.618897 3354 net.cpp:131] Top shape: 256 4 (1024)
  591. I0805 02:17:48.618899 3354 net.cpp:139] Memory required for data: 2510913536
  592. I0805 02:17:48.618904 3354 layer_factory.hpp:77] Creating layer loss
  593. I0805 02:17:48.618923 3354 net.cpp:86] Creating Layer loss
  594. I0805 02:17:48.618926 3354 net.cpp:408] loss <- fc8
  595. I0805 02:17:48.618928 3354 net.cpp:408] loss <- label
  596. I0805 02:17:48.618949 3354 net.cpp:382] loss -> loss
  597. I0805 02:17:48.618969 3354 layer_factory.hpp:77] Creating layer loss
  598. I0805 02:17:48.619225 3354 net.cpp:124] Setting up loss
  599. I0805 02:17:48.619230 3354 net.cpp:131] Top shape: (1)
  600. I0805 02:17:48.619232 3354 net.cpp:134] with loss weight 1
  601. I0805 02:17:48.619266 3354 net.cpp:139] Memory required for data: 2510913540
  602. I0805 02:17:48.619283 3354 net.cpp:200] loss needs backward computation.
  603. I0805 02:17:48.619289 3354 net.cpp:200] fc8 needs backward computation.
  604. I0805 02:17:48.619307 3354 net.cpp:200] drop7 needs backward computation.
  605. I0805 02:17:48.619309 3354 net.cpp:200] relu7 needs backward computation.
  606. I0805 02:17:48.619312 3354 net.cpp:200] fc7 needs backward computation.
  607. I0805 02:17:48.619314 3354 net.cpp:200] drop6 needs backward computation.
  608. I0805 02:17:48.619318 3354 net.cpp:200] relu6 needs backward computation.
  609. I0805 02:17:48.619320 3354 net.cpp:200] fc6 needs backward computation.
  610. I0805 02:17:48.619323 3354 net.cpp:200] pool5 needs backward computation.
  611. I0805 02:17:48.619328 3354 net.cpp:200] relu5 needs backward computation.
  612. I0805 02:17:48.619329 3354 net.cpp:200] conv5 needs backward computation.
  613. I0805 02:17:48.619333 3354 net.cpp:200] relu4 needs backward computation.
  614. I0805 02:17:48.619336 3354 net.cpp:200] conv4 needs backward computation.
  615. I0805 02:17:48.619338 3354 net.cpp:200] relu3 needs backward computation.
  616. I0805 02:17:48.619354 3354 net.cpp:200] conv3 needs backward computation.
  617. I0805 02:17:48.619357 3354 net.cpp:200] pool2 needs backward computation.
  618. I0805 02:17:48.619361 3354 net.cpp:200] norm2 needs backward computation.
  619. I0805 02:17:48.619379 3354 net.cpp:200] relu2 needs backward computation.
  620. I0805 02:17:48.619381 3354 net.cpp:200] scale1 needs backward computation.
  621. I0805 02:17:48.619385 3354 net.cpp:200] bn1 needs backward computation.
  622. I0805 02:17:48.619387 3354 net.cpp:200] conv2 needs backward computation.
  623. I0805 02:17:48.619390 3354 net.cpp:200] pool1 needs backward computation.
  624. I0805 02:17:48.619393 3354 net.cpp:200] norm1 needs backward computation.
  625. I0805 02:17:48.619397 3354 net.cpp:200] relu1 needs backward computation.
  626. I0805 02:17:48.619441 3354 net.cpp:200] conv1 needs backward computation.
  627. I0805 02:17:48.619444 3354 net.cpp:202] data does not need backward computation.
  628. I0805 02:17:48.619446 3354 net.cpp:244] This network produces output loss
  629. I0805 02:17:48.619460 3354 net.cpp:257] Network initialization done.
  630. I0805 02:17:48.619674 3354 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  631. I0805 02:17:48.619679 3354 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  632. I0805 02:17:48.619684 3354 solver.cpp:173] Creating test net (#0) specified by net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  633. I0805 02:17:48.619707 3354 net.cpp:296] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
  634. I0805 02:17:48.619833 3354 net.cpp:53] Initializing net from parameters:
  635. name: "AlexNet"
  636. state {
  637. phase: TEST
  638. }
  639. layer {
  640. name: "data"
  641. type: "Data"
  642. top: "data"
  643. top: "label"
  644. include {
  645. phase: TEST
  646. }
  647. data_param {
  648. source: "examples/Mydataset_test_lmdb"
  649. batch_size: 50
  650. backend: LMDB
  651. }
  652. }
  653. layer {
  654. name: "conv1"
  655. type: "Convolution"
  656. bottom: "data"
  657. top: "conv1"
  658. param {
  659. lr_mult: 1
  660. decay_mult: 1
  661. }
  662. param {
  663. lr_mult: 2
  664. decay_mult: 0
  665. }
  666. convolution_param {
  667. num_output: 96
  668. kernel_size: 11
  669. stride: 4
  670. weight_filler {
  671. type: "gaussian"
  672. std: 0.01
  673. }
  674. bias_filler {
  675. type: "constant"
  676. value: 0
  677. }
  678. }
  679. }
  680. layer {
  681. name: "relu1"
  682. type: "ReLU"
  683. bottom: "conv1"
  684. top: "conv1"
  685. }
  686. layer {
  687. name: "norm1"
  688. type: "LRN"
  689. bottom: "conv1"
  690. top: "norm1"
  691. lrn_param {
  692. local_size: 5
  693. alpha: 0.0001
  694. beta: 0.75
  695. }
  696. }
  697. layer {
  698. name: "pool1"
  699. type: "Pooling"
  700. bottom: "norm1"
  701. top: "pool1"
  702. pooling_param {
  703. pool: MAX
  704. kernel_size: 3
  705. stride: 2
  706. }
  707. }
  708. layer {
  709. name: "conv2"
  710. type: "Convolution"
  711. bottom: "pool1"
  712. top: "conv2"
  713. param {
  714. lr_mult: 1
  715. decay_mult: 1
  716. }
  717. param {
  718. lr_mult: 2
  719. decay_mult: 0
  720. }
  721. convolution_param {
  722. num_output: 256
  723. pad: 2
  724. kernel_size: 5
  725. group: 2
  726. weight_filler {
  727. type: "gaussian"
  728. std: 0.01
  729. }
  730. bias_filler {
  731. type: "constant"
  732. value: 0.1
  733. }
  734. }
  735. }
  736. layer {
  737. name: "bn1"
  738. type: "BatchNorm"
  739. bottom: "conv2"
  740. top: "conv2"
  741. param {
  742. lr_mult: 0
  743. decay_mult: 0
  744. }
  745. param {
  746. lr_mult: 0
  747. decay_mult: 0
  748. }
  749. param {
  750. lr_mult: 0
  751. decay_mult: 0
  752. }
  753. }
  754. layer {
  755. name: "scale1"
  756. type: "Scale"
  757. bottom: "conv2"
  758. top: "conv2"
  759. scale_param {
  760. bias_term: true
  761. }
  762. }
  763. layer {
  764. name: "relu2"
  765. type: "ReLU"
  766. bottom: "conv2"
  767. top: "conv2"
  768. }
  769. layer {
  770. name: "norm2"
  771. type: "LRN"
  772. bottom: "conv2"
  773. top: "norm2"
  774. lrn_param {
  775. local_size: 5
  776. alpha: 0.0001
  777. beta: 0.75
  778. }
  779. }
  780. layer {
  781. name: "pool2"
  782. type: "Pooling"
  783. bottom: "norm2"
  784. top: "pool2"
  785. pooling_param {
  786. pool: MAX
  787. kernel_size: 3
  788. stride: 2
  789. }
  790. }
  791. layer {
  792. name: "conv3"
  793. type: "Convolution"
  794. bottom: "pool2"
  795. top: "conv3"
  796. param {
  797. lr_mult: 1
  798. decay_mult: 1
  799. }
  800. param {
  801. lr_mult: 2
  802. decay_mult: 0
  803. }
  804. convolution_param {
  805. num_output: 384
  806. pad: 1
  807. kernel_size: 3
  808. weight_filler {
  809. type: "gaussian"
  810. std: 0.01
  811. }
  812. bias_filler {
  813. type: "constant"
  814. value: 0
  815. }
  816. }
  817. }
  818. layer {
  819. name: "relu3"
  820. type: "ReLU"
  821. bottom: "conv3"
  822. top: "conv3"
  823. }
  824. layer {
  825. name: "conv4"
  826. type: "Convolution"
  827. bottom: "conv3"
  828. top: "conv4"
  829. param {
  830. lr_mult: 1
  831. decay_mult: 1
  832. }
  833. param {
  834. lr_mult: 2
  835. decay_mult: 0
  836. }
  837. convolution_param {
  838. num_output: 384
  839. pad: 1
  840. kernel_size: 3
  841. group: 2
  842. weight_filler {
  843. type: "gaussian"
  844. std: 0.01
  845. }
  846. bias_filler {
  847. type: "constant"
  848. value: 0.1
  849. }
  850. }
  851. }
  852. layer {
  853. name: "relu4"
  854. type: "ReLU"
  855. bottom: "conv4"
  856. top: "conv4"
  857. }
  858. layer {
  859. name: "conv5"
  860. type: "Convolution"
  861. bottom: "conv4"
  862. top: "conv5"
  863. param {
  864. lr_mult: 1
  865. decay_mult: 1
  866. }
  867. param {
  868. lr_mult: 2
  869. decay_mult: 0
  870. }
  871. convolution_param {
  872. num_output: 256
  873. pad: 1
  874. kernel_size: 3
  875. group: 2
  876. weight_filler {
  877. type: "xavier"
  878. std: 0.01
  879. }
  880. bias_filler {
  881. type: "constant"
  882. value: 0.1
  883. }
  884. }
  885. }
  886. layer {
  887. name: "relu5"
  888. type: "ReLU"
  889. bottom: "conv5"
  890. top: "conv5"
  891. }
  892. layer {
  893. name: "pool5"
  894. type: "Pooling"
  895. bottom: "conv5"
  896. top: "pool5"
  897. pooling_param {
  898. pool: MAX
  899. kernel_size: 3
  900. stride: 2
  901. }
  902. }
  903. layer {
  904. name: "fc6"
  905. type: "InnerProduct"
  906. bottom: "pool5"
  907. top: "fc6"
  908. param {
  909. lr_mult: 1
  910. decay_mult: 1
  911. }
  912. param {
  913. lr_mult: 2
  914. decay_mult: 0
  915. }
  916. inner_product_param {
  917. num_output: 4096
  918. weight_filler {
  919. type: "xavier"
  920. std: 0.005
  921. }
  922. bias_filler {
  923. type: "constant"
  924. value: 0.1
  925. }
  926. }
  927. }
  928. layer {
  929. name: "relu6"
  930. type: "ReLU"
  931. bottom: "fc6"
  932. top: "fc6"
  933. }
  934. layer {
  935. name: "drop6"
  936. type: "Dropout"
  937. bottom: "fc6"
  938. top: "fc6"
  939. dropout_param {
  940. dropout_ratio: 0.5
  941. }
  942. }
  943. layer {
  944. name: "fc7"
  945. type: "InnerProduct"
  946. bottom: "fc6"
  947. top: "fc7"
  948. param {
  949. lr_mult: 1
  950. decay_mult: 1
  951. }
  952. param {
  953. lr_mult: 2
  954. decay_mult: 0
  955. }
  956. inner_product_param {
  957. num_output: 4096
  958. weight_filler {
  959. type: "xavier"
  960. std: 0.005
  961. }
  962. bias_filler {
  963. type: "constant"
  964. value: 0.1
  965. }
  966. }
  967. }
  968. layer {
  969. name: "relu7"
  970. type: "ReLU"
  971. bottom: "fc7"
  972. top: "fc7"
  973. }
  974. layer {
  975. name: "drop7"
  976. type: "Dropout"
  977. bottom: "fc7"
  978. top: "fc7"
  979. dropout_param {
  980. dropout_ratio: 0.5
  981. }
  982. }
  983. layer {
  984. name: "fc8"
  985. type: "InnerProduct"
  986. bottom: "fc7"
  987. top: "fc8"
  988. param {
  989. lr_mult: 1
  990. decay_mult: 1
  991. }
  992. param {
  993. lr_mult: 2
  994. decay_mult: 0
  995. }
  996. inner_product_param {
  997. num_output: 4
  998. weight_filler {
  999. type: "xavier"
  1000. std: 0.01
  1001. }
  1002. bias_filler {
  1003. type: "constant"
  1004. value: 0
  1005. }
  1006. }
  1007. }
  1008. layer {
  1009. name: "accuracy"
  1010. type: "Accuracy"
  1011. bottom: "fc8"
  1012. bottom: "label"
  1013. top: "accuracy"
  1014. include {
  1015. phase: TEST
  1016. }
  1017. }
  1018. layer {
  1019. name: "loss"
  1020. type: "SoftmaxWithLoss"
  1021. bottom: "fc8"
  1022. bottom: "label"
  1023. top: "loss"
  1024. }
  1025. I0805 02:17:48.620080 3354 layer_factory.hpp:77] Creating layer data
  1026. I0805 02:17:48.620121 3354 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_test_lmdb
  1027. I0805 02:17:48.620132 3354 net.cpp:86] Creating Layer data
  1028. I0805 02:17:48.620138 3354 net.cpp:382] data -> data
  1029. I0805 02:17:48.620144 3354 net.cpp:382] data -> label
  1030. I0805 02:17:48.620304 3354 data_layer.cpp:45] output data size: 50,3,227,227
  1031. I0805 02:17:48.658275 3354 net.cpp:124] Setting up data
  1032. I0805 02:17:48.658320 3354 net.cpp:131] Top shape: 50 3 227 227 (7729350)
  1033. I0805 02:17:48.658324 3354 net.cpp:131] Top shape: 50 (50)
  1034. I0805 02:17:48.658327 3354 net.cpp:139] Memory required for data: 30917600
  1035. I0805 02:17:48.658354 3354 layer_factory.hpp:77] Creating layer label_data_1_split
  1036. I0805 02:17:48.658362 3354 net.cpp:86] Creating Layer label_data_1_split
  1037. I0805 02:17:48.658366 3354 net.cpp:408] label_data_1_split <- label
  1038. I0805 02:17:48.658372 3354 net.cpp:382] label_data_1_split -> label_data_1_split_0
  1039. I0805 02:17:48.658380 3354 net.cpp:382] label_data_1_split -> label_data_1_split_1
  1040. I0805 02:17:48.658511 3354 net.cpp:124] Setting up label_data_1_split
  1041. I0805 02:17:48.658536 3354 net.cpp:131] Top shape: 50 (50)
  1042. I0805 02:17:48.658540 3354 net.cpp:131] Top shape: 50 (50)
  1043. I0805 02:17:48.658543 3354 net.cpp:139] Memory required for data: 30918000
  1044. I0805 02:17:48.658560 3354 layer_factory.hpp:77] Creating layer conv1
  1045. I0805 02:17:48.658571 3354 net.cpp:86] Creating Layer conv1
  1046. I0805 02:17:48.658573 3354 net.cpp:408] conv1 <- data
  1047. I0805 02:17:48.658578 3354 net.cpp:382] conv1 -> conv1
  1048. I0805 02:17:48.662369 3354 net.cpp:124] Setting up conv1
  1049. I0805 02:17:48.662395 3354 net.cpp:131] Top shape: 50 96 55 55 (14520000)
  1050. I0805 02:17:48.662396 3354 net.cpp:139] Memory required for data: 88998000
  1051. I0805 02:17:48.662421 3354 layer_factory.hpp:77] Creating layer relu1
  1052. I0805 02:17:48.662427 3354 net.cpp:86] Creating Layer relu1
  1053. I0805 02:17:48.662431 3354 net.cpp:408] relu1 <- conv1
  1054. I0805 02:17:48.662434 3354 net.cpp:369] relu1 -> conv1 (in-place)
  1055. I0805 02:17:48.662585 3354 net.cpp:124] Setting up relu1
  1056. I0805 02:17:48.662591 3354 net.cpp:131] Top shape: 50 96 55 55 (14520000)
  1057. I0805 02:17:48.662612 3354 net.cpp:139] Memory required for data: 147078000
  1058. I0805 02:17:48.662616 3354 layer_factory.hpp:77] Creating layer norm1
  1059. I0805 02:17:48.662621 3354 net.cpp:86] Creating Layer norm1
  1060. I0805 02:17:48.662624 3354 net.cpp:408] norm1 <- conv1
  1061. I0805 02:17:48.662628 3354 net.cpp:382] norm1 -> norm1
  1062. I0805 02:17:48.662780 3354 net.cpp:124] Setting up norm1
  1063. I0805 02:17:48.662786 3354 net.cpp:131] Top shape: 50 96 55 55 (14520000)
  1064. I0805 02:17:48.662802 3354 net.cpp:139] Memory required for data: 205158000
  1065. I0805 02:17:48.662806 3354 layer_factory.hpp:77] Creating layer pool1
  1066. I0805 02:17:48.662811 3354 net.cpp:86] Creating Layer pool1
  1067. I0805 02:17:48.662814 3354 net.cpp:408] pool1 <- norm1
  1068. I0805 02:17:48.662818 3354 net.cpp:382] pool1 -> pool1
  1069. I0805 02:17:48.662848 3354 net.cpp:124] Setting up pool1
  1070. I0805 02:17:48.662853 3354 net.cpp:131] Top shape: 50 96 27 27 (3499200)
  1071. I0805 02:17:48.662856 3354 net.cpp:139] Memory required for data: 219154800
  1072. I0805 02:17:48.662859 3354 layer_factory.hpp:77] Creating layer conv2
  1073. I0805 02:17:48.662865 3354 net.cpp:86] Creating Layer conv2
  1074. I0805 02:17:48.662868 3354 net.cpp:408] conv2 <- pool1
  1075. I0805 02:17:48.662874 3354 net.cpp:382] conv2 -> conv2
  1076. I0805 02:17:48.667425 3354 net.cpp:124] Setting up conv2
  1077. I0805 02:17:48.667439 3354 net.cpp:131] Top shape: 50 256 27 27 (9331200)
  1078. I0805 02:17:48.667457 3354 net.cpp:139] Memory required for data: 256479600
  1079. I0805 02:17:48.667465 3354 layer_factory.hpp:77] Creating layer bn1
  1080. I0805 02:17:48.667471 3354 net.cpp:86] Creating Layer bn1
  1081. I0805 02:17:48.667476 3354 net.cpp:408] bn1 <- conv2
  1082. I0805 02:17:48.667481 3354 net.cpp:369] bn1 -> conv2 (in-place)
  1083. I0805 02:17:48.667645 3354 net.cpp:124] Setting up bn1
  1084. I0805 02:17:48.667650 3354 net.cpp:131] Top shape: 50 256 27 27 (9331200)
  1085. I0805 02:17:48.667667 3354 net.cpp:139] Memory required for data: 293804400
  1086. I0805 02:17:48.667675 3354 layer_factory.hpp:77] Creating layer scale1
  1087. I0805 02:17:48.667682 3354 net.cpp:86] Creating Layer scale1
  1088. I0805 02:17:48.667685 3354 net.cpp:408] scale1 <- conv2
  1089. I0805 02:17:48.667690 3354 net.cpp:369] scale1 -> conv2 (in-place)
  1090. I0805 02:17:48.667734 3354 layer_factory.hpp:77] Creating layer scale1
  1091. I0805 02:17:48.667842 3354 net.cpp:124] Setting up scale1
  1092. I0805 02:17:48.667847 3354 net.cpp:131] Top shape: 50 256 27 27 (9331200)
  1093. I0805 02:17:48.667863 3354 net.cpp:139] Memory required for data: 331129200
  1094. I0805 02:17:48.667870 3354 layer_factory.hpp:77] Creating layer relu2
  1095. I0805 02:17:48.667876 3354 net.cpp:86] Creating Layer relu2
  1096. I0805 02:17:48.667878 3354 net.cpp:408] relu2 <- conv2
  1097. I0805 02:17:48.667881 3354 net.cpp:369] relu2 -> conv2 (in-place)
  1098. I0805 02:17:48.668392 3354 net.cpp:124] Setting up relu2
  1099. I0805 02:17:48.668416 3354 net.cpp:131] Top shape: 50 256 27 27 (9331200)
  1100. I0805 02:17:48.668418 3354 net.cpp:139] Memory required for data: 368454000
  1101. I0805 02:17:48.668421 3354 layer_factory.hpp:77] Creating layer norm2
  1102. I0805 02:17:48.668427 3354 net.cpp:86] Creating Layer norm2
  1103. I0805 02:17:48.668431 3354 net.cpp:408] norm2 <- conv2
  1104. I0805 02:17:48.668448 3354 net.cpp:382] norm2 -> norm2
  1105. I0805 02:17:48.668623 3354 net.cpp:124] Setting up norm2
  1106. I0805 02:17:48.668629 3354 net.cpp:131] Top shape: 50 256 27 27 (9331200)
  1107. I0805 02:17:48.668647 3354 net.cpp:139] Memory required for data: 405778800
  1108. I0805 02:17:48.668649 3354 layer_factory.hpp:77] Creating layer pool2
  1109. I0805 02:17:48.668658 3354 net.cpp:86] Creating Layer pool2
  1110. I0805 02:17:48.668660 3354 net.cpp:408] pool2 <- norm2
  1111. I0805 02:17:48.668664 3354 net.cpp:382] pool2 -> pool2
  1112. I0805 02:17:48.668694 3354 net.cpp:124] Setting up pool2
  1113. I0805 02:17:48.668699 3354 net.cpp:131] Top shape: 50 256 13 13 (2163200)
  1114. I0805 02:17:48.668702 3354 net.cpp:139] Memory required for data: 414431600
  1115. I0805 02:17:48.668706 3354 layer_factory.hpp:77] Creating layer conv3
  1116. I0805 02:17:48.668714 3354 net.cpp:86] Creating Layer conv3
  1117. I0805 02:17:48.668730 3354 net.cpp:408] conv3 <- pool2
  1118. I0805 02:17:48.668736 3354 net.cpp:382] conv3 -> conv3
  1119. I0805 02:17:48.678231 3354 net.cpp:124] Setting up conv3
  1120. I0805 02:17:48.678263 3354 net.cpp:131] Top shape: 50 384 13 13 (3244800)
  1121. I0805 02:17:48.678267 3354 net.cpp:139] Memory required for data: 427410800
  1122. I0805 02:17:48.678275 3354 layer_factory.hpp:77] Creating layer relu3
  1123. I0805 02:17:48.678294 3354 net.cpp:86] Creating Layer relu3
  1124. I0805 02:17:48.678297 3354 net.cpp:408] relu3 <- conv3
  1125. I0805 02:17:48.678302 3354 net.cpp:369] relu3 -> conv3 (in-place)
  1126. I0805 02:17:48.678485 3354 net.cpp:124] Setting up relu3
  1127. I0805 02:17:48.678493 3354 net.cpp:131] Top shape: 50 384 13 13 (3244800)
  1128. I0805 02:17:48.678509 3354 net.cpp:139] Memory required for data: 440390000
  1129. I0805 02:17:48.678511 3354 layer_factory.hpp:77] Creating layer conv4
  1130. I0805 02:17:48.678521 3354 net.cpp:86] Creating Layer conv4
  1131. I0805 02:17:48.678525 3354 net.cpp:408] conv4 <- conv3
  1132. I0805 02:17:48.678531 3354 net.cpp:382] conv4 -> conv4
  1133. I0805 02:17:48.687863 3354 net.cpp:124] Setting up conv4
  1134. I0805 02:17:48.687894 3354 net.cpp:131] Top shape: 50 384 13 13 (3244800)
  1135. I0805 02:17:48.687899 3354 net.cpp:139] Memory required for data: 453369200
  1136. I0805 02:17:48.687921 3354 layer_factory.hpp:77] Creating layer relu4
  1137. I0805 02:17:48.687930 3354 net.cpp:86] Creating Layer relu4
  1138. I0805 02:17:48.687934 3354 net.cpp:408] relu4 <- conv4
  1139. I0805 02:17:48.687940 3354 net.cpp:369] relu4 -> conv4 (in-place)
  1140. I0805 02:17:48.688117 3354 net.cpp:124] Setting up relu4
  1141. I0805 02:17:48.688123 3354 net.cpp:131] Top shape: 50 384 13 13 (3244800)
  1142. I0805 02:17:48.688125 3354 net.cpp:139] Memory required for data: 466348400
  1143. I0805 02:17:48.688128 3354 layer_factory.hpp:77] Creating layer conv5
  1144. I0805 02:17:48.688153 3354 net.cpp:86] Creating Layer conv5
  1145. I0805 02:17:48.688169 3354 net.cpp:408] conv5 <- conv4
  1146. I0805 02:17:48.688174 3354 net.cpp:382] conv5 -> conv5
  1147. I0805 02:17:48.693069 3354 net.cpp:124] Setting up conv5
  1148. I0805 02:17:48.693085 3354 net.cpp:131] Top shape: 50 256 13 13 (2163200)
  1149. I0805 02:17:48.693101 3354 net.cpp:139] Memory required for data: 475001200
  1150. I0805 02:17:48.693107 3354 layer_factory.hpp:77] Creating layer relu5
  1151. I0805 02:17:48.693114 3354 net.cpp:86] Creating Layer relu5
  1152. I0805 02:17:48.693119 3354 net.cpp:408] relu5 <- conv5
  1153. I0805 02:17:48.693122 3354 net.cpp:369] relu5 -> conv5 (in-place)
  1154. I0805 02:17:48.693282 3354 net.cpp:124] Setting up relu5
  1155. I0805 02:17:48.693290 3354 net.cpp:131] Top shape: 50 256 13 13 (2163200)
  1156. I0805 02:17:48.693305 3354 net.cpp:139] Memory required for data: 483654000
  1157. I0805 02:17:48.693306 3354 layer_factory.hpp:77] Creating layer pool5
  1158. I0805 02:17:48.693310 3354 net.cpp:86] Creating Layer pool5
  1159. I0805 02:17:48.693327 3354 net.cpp:408] pool5 <- conv5
  1160. I0805 02:17:48.693331 3354 net.cpp:382] pool5 -> pool5
  1161. I0805 02:17:48.693397 3354 net.cpp:124] Setting up pool5
  1162. I0805 02:17:48.693401 3354 net.cpp:131] Top shape: 50 256 6 6 (460800)
  1163. I0805 02:17:48.693404 3354 net.cpp:139] Memory required for data: 485497200
  1164. I0805 02:17:48.693421 3354 layer_factory.hpp:77] Creating layer fc6
  1165. I0805 02:17:48.693428 3354 net.cpp:86] Creating Layer fc6
  1166. I0805 02:17:48.693431 3354 net.cpp:408] fc6 <- pool5
  1167. I0805 02:17:48.693436 3354 net.cpp:382] fc6 -> fc6
  1168. I0805 02:17:48.872923 3354 net.cpp:124] Setting up fc6
  1169. I0805 02:17:48.872939 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1170. I0805 02:17:48.872941 3354 net.cpp:139] Memory required for data: 486316400
  1171. I0805 02:17:48.872966 3354 layer_factory.hpp:77] Creating layer relu6
  1172. I0805 02:17:48.872972 3354 net.cpp:86] Creating Layer relu6
  1173. I0805 02:17:48.872989 3354 net.cpp:408] relu6 <- fc6
  1174. I0805 02:17:48.872997 3354 net.cpp:369] relu6 -> fc6 (in-place)
  1175. I0805 02:17:48.873232 3354 net.cpp:124] Setting up relu6
  1176. I0805 02:17:48.873237 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1177. I0805 02:17:48.873239 3354 net.cpp:139] Memory required for data: 487135600
  1178. I0805 02:17:48.873241 3354 layer_factory.hpp:77] Creating layer drop6
  1179. I0805 02:17:48.873245 3354 net.cpp:86] Creating Layer drop6
  1180. I0805 02:17:48.873293 3354 net.cpp:408] drop6 <- fc6
  1181. I0805 02:17:48.873311 3354 net.cpp:369] drop6 -> fc6 (in-place)
  1182. I0805 02:17:48.873360 3354 net.cpp:124] Setting up drop6
  1183. I0805 02:17:48.873379 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1184. I0805 02:17:48.873381 3354 net.cpp:139] Memory required for data: 487954800
  1185. I0805 02:17:48.873383 3354 layer_factory.hpp:77] Creating layer fc7
  1186. I0805 02:17:48.873400 3354 net.cpp:86] Creating Layer fc7
  1187. I0805 02:17:48.873404 3354 net.cpp:408] fc7 <- fc6
  1188. I0805 02:17:48.873421 3354 net.cpp:382] fc7 -> fc7
  1189. I0805 02:17:48.952512 3354 net.cpp:124] Setting up fc7
  1190. I0805 02:17:48.952529 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1191. I0805 02:17:48.952533 3354 net.cpp:139] Memory required for data: 488774000
  1192. I0805 02:17:48.952553 3354 layer_factory.hpp:77] Creating layer relu7
  1193. I0805 02:17:48.952559 3354 net.cpp:86] Creating Layer relu7
  1194. I0805 02:17:48.952576 3354 net.cpp:408] relu7 <- fc7
  1195. I0805 02:17:48.952584 3354 net.cpp:369] relu7 -> fc7 (in-place)
  1196. I0805 02:17:48.953184 3354 net.cpp:124] Setting up relu7
  1197. I0805 02:17:48.953193 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1198. I0805 02:17:48.953196 3354 net.cpp:139] Memory required for data: 489593200
  1199. I0805 02:17:48.953198 3354 layer_factory.hpp:77] Creating layer drop7
  1200. I0805 02:17:48.953202 3354 net.cpp:86] Creating Layer drop7
  1201. I0805 02:17:48.953204 3354 net.cpp:408] drop7 <- fc7
  1202. I0805 02:17:48.953223 3354 net.cpp:369] drop7 -> fc7 (in-place)
  1203. I0805 02:17:48.953296 3354 net.cpp:124] Setting up drop7
  1204. I0805 02:17:48.953315 3354 net.cpp:131] Top shape: 50 4096 (204800)
  1205. I0805 02:17:48.953320 3354 net.cpp:139] Memory required for data: 490412400
  1206. I0805 02:17:48.953322 3354 layer_factory.hpp:77] Creating layer fc8
  1207. I0805 02:17:48.953330 3354 net.cpp:86] Creating Layer fc8
  1208. I0805 02:17:48.953348 3354 net.cpp:408] fc8 <- fc7
  1209. I0805 02:17:48.953353 3354 net.cpp:382] fc8 -> fc8
  1210. I0805 02:17:48.953526 3354 net.cpp:124] Setting up fc8
  1211. I0805 02:17:48.953532 3354 net.cpp:131] Top shape: 50 4 (200)
  1212. I0805 02:17:48.953534 3354 net.cpp:139] Memory required for data: 490413200
  1213. I0805 02:17:48.953538 3354 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
  1214. I0805 02:17:48.953541 3354 net.cpp:86] Creating Layer fc8_fc8_0_split
  1215. I0805 02:17:48.953544 3354 net.cpp:408] fc8_fc8_0_split <- fc8
  1216. I0805 02:17:48.953562 3354 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_0
  1217. I0805 02:17:48.953567 3354 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_1
  1218. I0805 02:17:48.953619 3354 net.cpp:124] Setting up fc8_fc8_0_split
  1219. I0805 02:17:48.953624 3354 net.cpp:131] Top shape: 50 4 (200)
  1220. I0805 02:17:48.953640 3354 net.cpp:131] Top shape: 50 4 (200)
  1221. I0805 02:17:48.953641 3354 net.cpp:139] Memory required for data: 490414800
  1222. I0805 02:17:48.953644 3354 layer_factory.hpp:77] Creating layer accuracy
  1223. I0805 02:17:48.953665 3354 net.cpp:86] Creating Layer accuracy
  1224. I0805 02:17:48.953667 3354 net.cpp:408] accuracy <- fc8_fc8_0_split_0
  1225. I0805 02:17:48.953671 3354 net.cpp:408] accuracy <- label_data_1_split_0
  1226. I0805 02:17:48.953675 3354 net.cpp:382] accuracy -> accuracy
  1227. I0805 02:17:48.953682 3354 net.cpp:124] Setting up accuracy
  1228. I0805 02:17:48.953686 3354 net.cpp:131] Top shape: (1)
  1229. I0805 02:17:48.953688 3354 net.cpp:139] Memory required for data: 490414804
  1230. I0805 02:17:48.953691 3354 layer_factory.hpp:77] Creating layer loss
  1231. I0805 02:17:48.953696 3354 net.cpp:86] Creating Layer loss
  1232. I0805 02:17:48.953698 3354 net.cpp:408] loss <- fc8_fc8_0_split_1
  1233. I0805 02:17:48.953702 3354 net.cpp:408] loss <- label_data_1_split_1
  1234. I0805 02:17:48.953706 3354 net.cpp:382] loss -> loss
  1235. I0805 02:17:48.953712 3354 layer_factory.hpp:77] Creating layer loss
  1236. I0805 02:17:48.953912 3354 net.cpp:124] Setting up loss
  1237. I0805 02:17:48.953919 3354 net.cpp:131] Top shape: (1)
  1238. I0805 02:17:48.953922 3354 net.cpp:134] with loss weight 1
  1239. I0805 02:17:48.953930 3354 net.cpp:139] Memory required for data: 490414808
  1240. I0805 02:17:48.953934 3354 net.cpp:200] loss needs backward computation.
  1241. I0805 02:17:48.953938 3354 net.cpp:202] accuracy does not need backward computation.
  1242. I0805 02:17:48.953955 3354 net.cpp:200] fc8_fc8_0_split needs backward computation.
  1243. I0805 02:17:48.953959 3354 net.cpp:200] fc8 needs backward computation.
  1244. I0805 02:17:48.953963 3354 net.cpp:200] drop7 needs backward computation.
  1245. I0805 02:17:48.953965 3354 net.cpp:200] relu7 needs backward computation.
  1246. I0805 02:17:48.953969 3354 net.cpp:200] fc7 needs backward computation.
  1247. I0805 02:17:48.953972 3354 net.cpp:200] drop6 needs backward computation.
  1248. I0805 02:17:48.953975 3354 net.cpp:200] relu6 needs backward computation.
  1249. I0805 02:17:48.953979 3354 net.cpp:200] fc6 needs backward computation.
  1250. I0805 02:17:48.953981 3354 net.cpp:200] pool5 needs backward computation.
  1251. I0805 02:17:48.953985 3354 net.cpp:200] relu5 needs backward computation.
  1252. I0805 02:17:48.953989 3354 net.cpp:200] conv5 needs backward computation.
  1253. I0805 02:17:48.953991 3354 net.cpp:200] relu4 needs backward computation.
  1254. I0805 02:17:48.953994 3354 net.cpp:200] conv4 needs backward computation.
  1255. I0805 02:17:48.953999 3354 net.cpp:200] relu3 needs backward computation.
  1256. I0805 02:17:48.954001 3354 net.cpp:200] conv3 needs backward computation.
  1257. I0805 02:17:48.954005 3354 net.cpp:200] pool2 needs backward computation.
  1258. I0805 02:17:48.954008 3354 net.cpp:200] norm2 needs backward computation.
  1259. I0805 02:17:48.954012 3354 net.cpp:200] relu2 needs backward computation.
  1260. I0805 02:17:48.954015 3354 net.cpp:200] scale1 needs backward computation.
  1261. I0805 02:17:48.954018 3354 net.cpp:200] bn1 needs backward computation.
  1262. I0805 02:17:48.954021 3354 net.cpp:200] conv2 needs backward computation.
  1263. I0805 02:17:48.954025 3354 net.cpp:200] pool1 needs backward computation.
  1264. I0805 02:17:48.954028 3354 net.cpp:200] norm1 needs backward computation.
  1265. I0805 02:17:48.954031 3354 net.cpp:200] relu1 needs backward computation.
  1266. I0805 02:17:48.954035 3354 net.cpp:200] conv1 needs backward computation.
  1267. I0805 02:17:48.954038 3354 net.cpp:202] label_data_1_split does not need backward computation.
  1268. I0805 02:17:48.954042 3354 net.cpp:202] data does not need backward computation.
  1269. I0805 02:17:48.954044 3354 net.cpp:244] This network produces output accuracy
  1270. I0805 02:17:48.954048 3354 net.cpp:244] This network produces output loss
  1271. I0805 02:17:48.954063 3354 net.cpp:257] Network initialization done.
  1272. I0805 02:17:48.954123 3354 solver.cpp:56] Solver scaffolding done.
  1273. I0805 02:17:48.954715 3354 caffe.cpp:248] Starting Optimization
  1274. I0805 02:17:48.954718 3354 solver.cpp:273] Solving AlexNet
  1275. I0805 02:17:48.954722 3354 solver.cpp:274] Learning Rate Policy: step
  1276. I0805 02:17:48.957973 3354 solver.cpp:331] Iteration 0, Testing net (#0)
  1277. I0805 02:17:49.119514 3354 blocking_queue.cpp:49] Waiting for data
  1278. I0805 02:17:50.964622 3354 solver.cpp:398] Test net output #0: accuracy = 0.195926
  1279. I0805 02:17:50.964660 3354 solver.cpp:398] Test net output #1: loss = 5.40587 (* 1 = 5.40587 loss)
  1280. I0805 02:17:51.079339 3362 data_layer.cpp:73] Restarting data prefetching from start.
  1281. I0805 02:17:51.285264 3354 solver.cpp:219] Iteration 0 (1.26553e-25 iter/s, 2.33049s/20 iters), loss = 1.47625
  1282. I0805 02:17:51.285292 3354 solver.cpp:238] Train net output #0: loss = 1.47625 (* 1 = 1.47625 loss)
  1283. I0805 02:17:51.285316 3354 sgd_solver.cpp:105] Iteration 0, lr = 0.001
  1284. I0805 02:17:57.663884 3354 solver.cpp:219] Iteration 20 (3.13554 iter/s, 6.37849s/20 iters), loss = 1.09913
  1285. I0805 02:17:57.663913 3354 solver.cpp:238] Train net output #0: loss = 1.09913 (* 1 = 1.09913 loss)
  1286. I0805 02:17:57.663931 3354 sgd_solver.cpp:105] Iteration 20, lr = 0.001
  1287. I0805 02:17:58.142580 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1288. I0805 02:18:04.046460 3354 solver.cpp:219] Iteration 40 (3.13359 iter/s, 6.38245s/20 iters), loss = 1.19678
  1289. I0805 02:18:04.046488 3354 solver.cpp:238] Train net output #0: loss = 1.19678 (* 1 = 1.19678 loss)
  1290. I0805 02:18:04.046494 3354 sgd_solver.cpp:105] Iteration 40, lr = 0.001
  1291. I0805 02:18:06.158246 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1292. I0805 02:18:10.428968 3354 solver.cpp:219] Iteration 60 (3.13363 iter/s, 6.38238s/20 iters), loss = 1.18853
  1293. I0805 02:18:10.429024 3354 solver.cpp:238] Train net output #0: loss = 1.18853 (* 1 = 1.18853 loss)
  1294. I0805 02:18:10.429029 3354 sgd_solver.cpp:105] Iteration 60, lr = 0.001
  1295. I0805 02:18:14.424785 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1296. I0805 02:18:16.829213 3354 solver.cpp:219] Iteration 80 (3.12496 iter/s, 6.40009s/20 iters), loss = 1.17246
  1297. I0805 02:18:16.829241 3354 solver.cpp:238] Train net output #0: loss = 1.17246 (* 1 = 1.17246 loss)
  1298. I0805 02:18:16.829246 3354 sgd_solver.cpp:105] Iteration 80, lr = 0.001
  1299. I0805 02:18:22.450965 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1300. I0805 02:18:23.221148 3354 solver.cpp:219] Iteration 100 (3.12901 iter/s, 6.3918s/20 iters), loss = 1.12307
  1301. I0805 02:18:23.221177 3354 solver.cpp:238] Train net output #0: loss = 1.12307 (* 1 = 1.12307 loss)
  1302. I0805 02:18:23.221196 3354 sgd_solver.cpp:105] Iteration 100, lr = 0.001
  1303. I0805 02:18:29.610299 3354 solver.cpp:219] Iteration 120 (3.13025 iter/s, 6.38926s/20 iters), loss = 1.09175
  1304. I0805 02:18:29.610327 3354 solver.cpp:238] Train net output #0: loss = 1.09175 (* 1 = 1.09175 loss)
  1305. I0805 02:18:29.610332 3354 sgd_solver.cpp:105] Iteration 120, lr = 0.001
  1306. I0805 02:18:30.719750 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1307. I0805 02:18:36.019553 3354 solver.cpp:219] Iteration 140 (3.12035 iter/s, 6.40953s/20 iters), loss = 1.18237
  1308. I0805 02:18:36.019580 3354 solver.cpp:238] Train net output #0: loss = 1.18237 (* 1 = 1.18237 loss)
  1309. I0805 02:18:36.019599 3354 sgd_solver.cpp:105] Iteration 140, lr = 0.001
  1310. I0805 02:18:38.764142 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1311. I0805 02:18:42.429481 3354 solver.cpp:219] Iteration 160 (3.12003 iter/s, 6.4102s/20 iters), loss = 1.14376
  1312. I0805 02:18:42.429508 3354 solver.cpp:238] Train net output #0: loss = 1.14376 (* 1 = 1.14376 loss)
  1313. I0805 02:18:42.429527 3354 sgd_solver.cpp:105] Iteration 160, lr = 0.001
  1314. I0805 02:18:47.061050 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1315. I0805 02:18:48.836874 3354 solver.cpp:219] Iteration 180 (3.12126 iter/s, 6.40766s/20 iters), loss = 1.12219
  1316. I0805 02:18:48.836901 3354 solver.cpp:238] Train net output #0: loss = 1.12219 (* 1 = 1.12219 loss)
  1317. I0805 02:18:48.836906 3354 sgd_solver.cpp:105] Iteration 180, lr = 0.001
  1318. I0805 02:18:55.106952 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1319. I0805 02:18:55.252671 3354 solver.cpp:219] Iteration 200 (3.11718 iter/s, 6.41606s/20 iters), loss = 0.991293
  1320. I0805 02:18:55.252702 3354 solver.cpp:238] Train net output #0: loss = 0.991293 (* 1 = 0.991293 loss)
  1321. I0805 02:18:55.252722 3354 sgd_solver.cpp:105] Iteration 200, lr = 0.001
  1322. I0805 02:19:01.675238 3354 solver.cpp:219] Iteration 220 (3.11389 iter/s, 6.42282s/20 iters), loss = 1.00901
  1323. I0805 02:19:01.675266 3354 solver.cpp:238] Train net output #0: loss = 1.00901 (* 1 = 1.00901 loss)
  1324. I0805 02:19:01.675287 3354 sgd_solver.cpp:105] Iteration 220, lr = 0.001
  1325. I0805 02:19:03.417192 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1326. I0805 02:19:08.092980 3354 solver.cpp:219] Iteration 240 (3.11623 iter/s, 6.418s/20 iters), loss = 0.976785
  1327. I0805 02:19:08.093008 3354 solver.cpp:238] Train net output #0: loss = 0.976785 (* 1 = 0.976785 loss)
  1328. I0805 02:19:08.093014 3354 sgd_solver.cpp:105] Iteration 240, lr = 0.001
  1329. I0805 02:19:11.480767 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1330. I0805 02:19:14.534917 3354 solver.cpp:219] Iteration 260 (3.10453 iter/s, 6.44219s/20 iters), loss = 0.978768
  1331. I0805 02:19:14.534945 3354 solver.cpp:238] Train net output #0: loss = 0.978768 (* 1 = 0.978768 loss)
  1332. I0805 02:19:14.534965 3354 sgd_solver.cpp:105] Iteration 260, lr = 0.001
  1333. I0805 02:19:19.823832 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1334. I0805 02:19:20.977524 3354 solver.cpp:219] Iteration 280 (3.10421 iter/s, 6.44286s/20 iters), loss = 1.0599
  1335. I0805 02:19:20.977550 3354 solver.cpp:238] Train net output #0: loss = 1.0599 (* 1 = 1.0599 loss)
  1336. I0805 02:19:20.977569 3354 sgd_solver.cpp:105] Iteration 280, lr = 0.001
  1337. I0805 02:19:27.415933 3354 solver.cpp:219] Iteration 300 (3.10624 iter/s, 6.43866s/20 iters), loss = 0.888842
  1338. I0805 02:19:27.416128 3354 solver.cpp:238] Train net output #0: loss = 0.888842 (* 1 = 0.888842 loss)
  1339. I0805 02:19:27.416151 3354 sgd_solver.cpp:105] Iteration 300, lr = 0.001
  1340. I0805 02:19:27.902052 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1341. I0805 02:19:33.852607 3354 solver.cpp:219] Iteration 320 (3.10715 iter/s, 6.43676s/20 iters), loss = 0.975844
  1342. I0805 02:19:33.852635 3354 solver.cpp:238] Train net output #0: loss = 0.975844 (* 1 = 0.975844 loss)
  1343. I0805 02:19:33.852654 3354 sgd_solver.cpp:105] Iteration 320, lr = 0.001
  1344. I0805 02:19:35.978494 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1345. I0805 02:19:40.275147 3354 solver.cpp:219] Iteration 340 (3.11392 iter/s, 6.42278s/20 iters), loss = 0.933279
  1346. I0805 02:19:40.275174 3354 solver.cpp:238] Train net output #0: loss = 0.933279 (* 1 = 0.933279 loss)
  1347. I0805 02:19:40.275194 3354 sgd_solver.cpp:105] Iteration 340, lr = 0.001
  1348. I0805 02:19:44.293598 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1349. I0805 02:19:46.716472 3354 solver.cpp:219] Iteration 360 (3.10484 iter/s, 6.44156s/20 iters), loss = 0.960518
  1350. I0805 02:19:46.716506 3354 solver.cpp:238] Train net output #0: loss = 0.960518 (* 1 = 0.960518 loss)
  1351. I0805 02:19:46.716524 3354 sgd_solver.cpp:105] Iteration 360, lr = 0.001
  1352. I0805 02:19:52.412041 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1353. I0805 02:19:53.205533 3354 solver.cpp:219] Iteration 380 (3.082 iter/s, 6.48929s/20 iters), loss = 0.909624
  1354. I0805 02:19:53.205565 3354 solver.cpp:238] Train net output #0: loss = 0.909624 (* 1 = 0.909624 loss)
  1355. I0805 02:19:53.205585 3354 sgd_solver.cpp:105] Iteration 380, lr = 0.001
  1356. I0805 02:19:59.693739 3354 solver.cpp:219] Iteration 400 (3.08241 iter/s, 6.48843s/20 iters), loss = 0.821957
  1357. I0805 02:19:59.693917 3354 solver.cpp:238] Train net output #0: loss = 0.821957 (* 1 = 0.821957 loss)
  1358. I0805 02:19:59.693939 3354 sgd_solver.cpp:105] Iteration 400, lr = 0.001
  1359. I0805 02:20:00.816355 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1360. I0805 02:20:06.136162 3354 solver.cpp:219] Iteration 420 (3.10438 iter/s, 6.44252s/20 iters), loss = 0.957007
  1361. I0805 02:20:06.136193 3354 solver.cpp:238] Train net output #0: loss = 0.957007 (* 1 = 0.957007 loss)
  1362. I0805 02:20:06.136211 3354 sgd_solver.cpp:105] Iteration 420, lr = 0.001
  1363. I0805 02:20:08.895287 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1364. I0805 02:20:12.605448 3354 solver.cpp:219] Iteration 440 (3.09143 iter/s, 6.46951s/20 iters), loss = 0.939899
  1365. I0805 02:20:12.605476 3354 solver.cpp:238] Train net output #0: loss = 0.939899 (* 1 = 0.939899 loss)
  1366. I0805 02:20:12.605495 3354 sgd_solver.cpp:105] Iteration 440, lr = 0.001
  1367. I0805 02:20:17.296435 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1368. I0805 02:20:19.085794 3354 solver.cpp:219] Iteration 460 (3.08615 iter/s, 6.48057s/20 iters), loss = 0.858006
  1369. I0805 02:20:19.085822 3354 solver.cpp:238] Train net output #0: loss = 0.858006 (* 1 = 0.858006 loss)
  1370. I0805 02:20:19.085841 3354 sgd_solver.cpp:105] Iteration 460, lr = 0.001
  1371. I0805 02:20:25.370455 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1372. I0805 02:20:25.511584 3354 solver.cpp:219] Iteration 480 (3.11235 iter/s, 6.426s/20 iters), loss = 0.826491
  1373. I0805 02:20:25.511615 3354 solver.cpp:238] Train net output #0: loss = 0.826491 (* 1 = 0.826491 loss)
  1374. I0805 02:20:25.511634 3354 sgd_solver.cpp:105] Iteration 480, lr = 0.001
  1375. I0805 02:20:31.960383 3354 solver.cpp:219] Iteration 500 (3.10125 iter/s, 6.44901s/20 iters), loss = 0.767649
  1376. I0805 02:20:31.960554 3354 solver.cpp:238] Train net output #0: loss = 0.767649 (* 1 = 0.767649 loss)
  1377. I0805 02:20:31.960577 3354 sgd_solver.cpp:105] Iteration 500, lr = 0.001
  1378. I0805 02:20:33.766893 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1379. I0805 02:20:38.403030 3354 solver.cpp:219] Iteration 520 (3.10427 iter/s, 6.44273s/20 iters), loss = 0.809904
  1380. I0805 02:20:38.403059 3354 solver.cpp:238] Train net output #0: loss = 0.809904 (* 1 = 0.809904 loss)
  1381. I0805 02:20:38.403079 3354 sgd_solver.cpp:105] Iteration 520, lr = 0.001
  1382. I0805 02:20:41.798969 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1383. I0805 02:20:44.858299 3354 solver.cpp:219] Iteration 540 (3.09815 iter/s, 6.45547s/20 iters), loss = 0.744726
  1384. I0805 02:20:44.858327 3354 solver.cpp:238] Train net output #0: loss = 0.744726 (* 1 = 0.744726 loss)
  1385. I0805 02:20:44.858346 3354 sgd_solver.cpp:105] Iteration 540, lr = 0.001
  1386. I0805 02:20:50.157848 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1387. I0805 02:20:51.305193 3354 solver.cpp:219] Iteration 560 (3.10217 iter/s, 6.4471s/20 iters), loss = 0.758878
  1388. I0805 02:20:51.305222 3354 solver.cpp:238] Train net output #0: loss = 0.758878 (* 1 = 0.758878 loss)
  1389. I0805 02:20:51.305240 3354 sgd_solver.cpp:105] Iteration 560, lr = 0.001
  1390. I0805 02:20:57.777695 3354 solver.cpp:219] Iteration 580 (3.0899 iter/s, 6.4727s/20 iters), loss = 0.636537
  1391. I0805 02:20:57.777724 3354 solver.cpp:238] Train net output #0: loss = 0.636537 (* 1 = 0.636537 loss)
  1392. I0805 02:20:57.777745 3354 sgd_solver.cpp:105] Iteration 580, lr = 0.001
  1393. I0805 02:20:58.271718 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1394. I0805 02:21:04.231633 3354 solver.cpp:219] Iteration 600 (3.09879 iter/s, 6.45413s/20 iters), loss = 0.752477
  1395. I0805 02:21:04.231822 3354 solver.cpp:238] Train net output #0: loss = 0.752477 (* 1 = 0.752477 loss)
  1396. I0805 02:21:04.231869 3354 sgd_solver.cpp:105] Iteration 600, lr = 0.001
  1397. I0805 02:21:06.627866 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1398. I0805 02:21:07.148818 3354 solver.cpp:331] Iteration 610, Testing net (#0)
  1399. I0805 02:21:09.101388 3354 solver.cpp:398] Test net output #0: accuracy = 0.166296
  1400. I0805 02:21:09.101428 3354 solver.cpp:398] Test net output #1: loss = 2.17123 (* 1 = 2.17123 loss)
  1401. I0805 02:21:12.643909 3354 solver.cpp:219] Iteration 620 (2.37744 iter/s, 8.41239s/20 iters), loss = 0.724874
  1402. I0805 02:21:12.643936 3354 solver.cpp:238] Train net output #0: loss = 0.724874 (* 1 = 0.724874 loss)
  1403. I0805 02:21:12.643956 3354 sgd_solver.cpp:105] Iteration 620, lr = 0.001
  1404. I0805 02:21:16.677302 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1405. I0805 02:21:19.090366 3354 solver.cpp:219] Iteration 640 (3.10239 iter/s, 6.44665s/20 iters), loss = 0.696024
  1406. I0805 02:21:19.090395 3354 solver.cpp:238] Train net output #0: loss = 0.696024 (* 1 = 0.696024 loss)
  1407. I0805 02:21:19.090415 3354 sgd_solver.cpp:105] Iteration 640, lr = 0.001
  1408. I0805 02:21:24.768528 3361 data_layer.cpp:73] Restarting data prefetching from start.
  1409. I0805 02:21:25.534693 3354 solver.cpp:219] Iteration 660 (3.10342 iter/s, 6.44451s/20 iters), loss = 0.636167
  1410. I0805 02:21:25.534720 3354 solver.cpp:238] Train net output #0: loss = 0.636167 (* 1 = 0.636167 loss)
  1411. I0805 02:21:25.534725 3354 sgd_solver.cpp:105] Iteration 660, lr = 0.001
  1412. ^CI0805 02:21:26.827807 3354 solver.cpp:448] Snapshotting to binary proto file models/bvlc_alexnet/caffe_alexnet_sinatrain_iter_665.caffemodel
  1413. I0805 02:21:27.386745 3354 sgd_solver.cpp:273] Snapshotting solver state to binary proto file models/bvlc_alexnet/caffe_alexnet_sinatrain_iter_665.solverstate
  1414. I0805 02:21:27.628018 3354 solver.cpp:295] Optimization stopped early.
  1415. I0805 02:21:27.628033 3354 caffe.cpp:259] 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!