API tools faq
paste
Guest User

Untitled

a guest
Aug 3rd, 2017
202
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 45.68 KB | None | 0 0
raw download clone embed print report
  1. sina@sina-Z170X-UD3:~$ cd code/caffe/
  2. sina@sina-Z170X-UD3:~/code/caffe$ ./build/tools/caffe train --solver=examples/alexnetfinetune/solver_sina.prototxt
  3. I0803 12:10:35.137758 2175 caffe.cpp:218] Using GPUs 0
  4. I0803 12:10:35.359722 2175 caffe.cpp:223] GPU 0: GeForce GTX 1070
  5. I0803 12:10:38.265421 2175 solver.cpp:44] Initializing solver from parameters:
  6. test_iter: 280
  7. test_interval: 660
  8. base_lr: 0.001
  9. display: 20
  10. max_iter: 5000
  11. lr_policy: "step"
  12. gamma: 0.1
  13. momentum: 0.9
  14. weight_decay: 0.0005
  15. stepsize: 100000
  16. snapshot: 4999
  17. snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_sinatrain"
  18. solver_mode: GPU
  19. device_id: 0
  20. net: "examples/alexnetfinetune/alexnetsade-bn.prototxt"
  21. train_state {
  22. level: 0
  23. stage: ""
  24. }
  25. type: "SGD"
  26. I0803 12:10:38.265714 2175 solver.cpp:87] Creating training net from net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  27. I0803 12:10:38.286698 2175 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  28. I0803 12:10:38.286759 2175 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  29. I0803 12:10:38.286895 2175 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
  30. I0803 12:10:38.286932 2175 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
  31. I0803 12:10:38.287377 2175 net.cpp:53] Initializing net from parameters:
  32. name: "AlexNet"
  33. state {
  34. phase: TRAIN
  35. level: 0
  36. stage: ""
  37. }
  38. layer {
  39. name: "data"
  40. type: "Data"
  41. top: "data"
  42. top: "label"
  43. include {
  44. phase: TRAIN
  45. }
  46. transform_param {
  47. mirror: true
  48. crop_size: 227
  49. mean_file: "examples/Mydataset_train_lmdb/mean_imagetest.binaryproto"
  50. }
  51. data_param {
  52. source: "examples/Mydataset_train_lmdb"
  53. batch_size: 10
  54. backend: LMDB
  55. }
  56. }
  57. layer {
  58. name: "conv1"
  59. type: "Convolution"
  60. bottom: "data"
  61. top: "conv1"
  62. param {
  63. lr_mult: 1
  64. }
  65. convolution_param {
  66. num_output: 64
  67. bias_term: true
  68. pad: 1
  69. kernel_size: 3
  70. stride: 1
  71. weight_filler {
  72. type: "xavier"
  73. }
  74. }
  75. }
  76. layer {
  77. name: "bn1"
  78. type: "BatchNorm"
  79. bottom: "conv1"
  80. top: "conv1"
  81. param {
  82. lr_mult: 0
  83. decay_mult: 0
  84. }
  85. param {
  86. lr_mult: 0
  87. decay_mult: 0
  88. }
  89. param {
  90. lr_mult: 0
  91. decay_mult: 0
  92. }
  93. }
  94. layer {
  95. name: "scale1"
  96. type: "Scale"
  97. bottom: "conv1"
  98. top: "conv1"
  99. scale_param {
  100. bias_term: true
  101. }
  102. }
  103. layer {
  104. name: "relu1"
  105. type: "ReLU"
  106. bottom: "conv1"
  107. top: "conv1"
  108. }
  109. layer {
  110. name: "norm1"
  111. type: "LRN"
  112. bottom: "conv1"
  113. top: "norm1"
  114. lrn_param {
  115. local_size: 5
  116. alpha: 0.0001
  117. beta: 0.75
  118. }
  119. }
  120. layer {
  121. name: "pool1"
  122. type: "Pooling"
  123. bottom: "norm1"
  124. top: "pool1"
  125. pooling_param {
  126. pool: MAX
  127. kernel_size: 3
  128. stride: 2
  129. }
  130. }
  131. layer {
  132. name: "conv2"
  133. type: "Convolution"
  134. bottom: "pool1"
  135. top: "conv2"
  136. param {
  137. lr_mult: 1
  138. decay_mult: 1
  139. }
  140. param {
  141. lr_mult: 2
  142. decay_mult: 0
  143. }
  144. convolution_param {
  145. num_output: 256
  146. pad: 2
  147. kernel_size: 5
  148. group: 2
  149. weight_filler {
  150. type: "gaussian"
  151. std: 0.01
  152. }
  153. bias_filler {
  154. type: "constant"
  155. value: 0.1
  156. }
  157. }
  158. }
  159. layer {
  160. name: "relu2"
  161. type: "ReLU"
  162. bottom: "conv2"
  163. top: "conv2"
  164. }
  165. layer {
  166. name: "norm2"
  167. type: "LRN"
  168. bottom: "conv2"
  169. top: "norm2"
  170. lrn_param {
  171. local_size: 5
  172. alpha: 0.0001
  173. beta: 0.75
  174. }
  175. }
  176. layer {
  177. name: "pool2"
  178. type: "Pooling"
  179. bottom: "norm2"
  180. top: "pool2"
  181. pooling_param {
  182. pool: MAX
  183. kernel_size: 3
  184. stride: 2
  185. }
  186. }
  187. layer {
  188. name: "conv3"
  189. type: "Convolution"
  190. bottom: "pool2"
  191. top: "conv3"
  192. param {
  193. lr_mult: 1
  194. decay_mult: 1
  195. }
  196. param {
  197. lr_mult: 2
  198. decay_mult: 0
  199. }
  200. convolution_param {
  201. num_output: 384
  202. pad: 1
  203. kernel_size: 3
  204. weight_filler {
  205. type: "gaussian"
  206. std: 0.01
  207. }
  208. bias_filler {
  209. type: "constant"
  210. value: 0
  211. }
  212. }
  213. }
  214. layer {
  215. name: "relu3"
  216. type: "ReLU"
  217. bottom: "conv3"
  218. top: "conv3"
  219. }
  220. layer {
  221. name: "conv4"
  222. type: "Convolution"
  223. bottom: "conv3"
  224. top: "conv4"
  225. param {
  226. lr_mult: 1
  227. decay_mult: 1
  228. }
  229. param {
  230. lr_mult: 2
  231. decay_mult: 0
  232. }
  233. convolution_param {
  234. num_output: 384
  235. pad: 1
  236. kernel_size: 3
  237. group: 2
  238. weight_filler {
  239. type: "gaussian"
  240. std: 0.01
  241. }
  242. bias_filler {
  243. type: "constant"
  244. value: 0.1
  245. }
  246. }
  247. }
  248. layer {
  249. name: "relu4"
  250. type: "ReLU"
  251. bottom: "conv4"
  252. top: "conv4"
  253. }
  254. layer {
  255. name: "conv5"
  256. type: "Convolution"
  257. bottom: "conv4"
  258. top: "conv5"
  259. param {
  260. lr_mult: 1
  261. decay_mult: 1
  262. }
  263. param {
  264. lr_mult: 2
  265. decay_mult: 0
  266. }
  267. convolution_param {
  268. num_output: 256
  269. pad: 1
  270. kernel_size: 3
  271. group: 2
  272. weight_filler {
  273. type: "gaussian"
  274. std: 0.01
  275. }
  276. bias_filler {
  277. type: "constant"
  278. value: 0.1
  279. }
  280. }
  281. }
  282. layer {
  283. name: "relu5"
  284. type: "ReLU"
  285. bottom: "conv5"
  286. top: "conv5"
  287. }
  288. layer {
  289. name: "pool5"
  290. type: "Pooling"
  291. bottom: "conv5"
  292. top: "pool5"
  293. pooling_param {
  294. pool: MAX
  295. kernel_size: 3
  296. stride: 2
  297. }
  298. }
  299. layer {
  300. name: "fc6"
  301. type: "InnerProduct"
  302. bottom: "pool5"
  303. top: "fc6"
  304. param {
  305. lr_mult: 1
  306. decay_mult: 1
  307. }
  308. param {
  309. lr_mult: 2
  310. decay_mult: 0
  311. }
  312. inner_product_param {
  313. num_output: 4096
  314. weight_filler {
  315. type: "gaussian"
  316. std: 0.005
  317. }
  318. bias_filler {
  319. type: "constant"
  320. value: 0.1
  321. }
  322. }
  323. }
  324. layer {
  325. name: "relu6"
  326. type: "ReLU"
  327. bottom: "fc6"
  328. top: "fc6"
  329. }
  330. layer {
  331. name: "drop6"
  332. type: "Dropout"
  333. bottom: "fc6"
  334. top: "fc6"
  335. dropout_param {
  336. dropout_ratio: 0.5
  337. }
  338. }
  339. layer {
  340. name: "fc7"
  341. type: "InnerProduct"
  342. bottom: "fc6"
  343. top: "fc7"
  344. param {
  345. lr_mult: 1
  346. decay_mult: 1
  347. }
  348. param {
  349. lr_mult: 2
  350. decay_mult: 0
  351. }
  352. inner_product_param {
  353. num_output: 4096
  354. weight_filler {
  355. type: "gaussian"
  356. std: 0.005
  357. }
  358. bias_filler {
  359. type: "constant"
  360. value: 0.1
  361. }
  362. }
  363. }
  364. layer {
  365. name: "relu7"
  366. type: "ReLU"
  367. bottom: "fc7"
  368. top: "fc7"
  369. }
  370. layer {
  371. name: "drop7"
  372. type: "Dropout"
  373. bottom: "fc7"
  374. top: "fc7"
  375. dropout_param {
  376. dropout_ratio: 0.5
  377. }
  378. }
  379. layer {
  380. name: "fc8"
  381. type: "InnerProduct"
  382. bottom: "fc7"
  383. top: "fc8"
  384. param {
  385. lr_mult: 1
  386. decay_mult: 1
  387. }
  388. param {
  389. lr_mult: 2
  390. decay_mult: 0
  391. }
  392. inner_product_param {
  393. num_output: 1000
  394. weight_filler {
  395. type: "gaussian"
  396. std: 0.01
  397. }
  398. bias_filler {
  399. type: "constant"
  400. value: 0
  401. }
  402. }
  403. }
  404. layer {
  405. name: "loss"
  406. type: "SoftmaxWithLoss"
  407. bottom: "fc8"
  408. bottom: "label"
  409. top: "loss"
  410. }
  411. I0803 12:10:38.288094 2175 layer_factory.hpp:77] Creating layer data
  412. I0803 12:10:38.463691 2175 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_train_lmdb
  413. I0803 12:10:38.506412 2175 net.cpp:86] Creating Layer data
  414. I0803 12:10:38.506475 2175 net.cpp:382] data -> data
  415. I0803 12:10:38.506530 2175 net.cpp:382] data -> label
  416. I0803 12:10:38.506580 2175 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_train_lmdb/mean_imagetest.binaryproto
  417. I0803 12:10:38.548476 2175 data_layer.cpp:45] output data size: 10,3,227,227
  418. I0803 12:10:38.560668 2175 net.cpp:124] Setting up data
  419. I0803 12:10:38.560693 2175 net.cpp:131] Top shape: 10 3 227 227 (1545870)
  420. I0803 12:10:38.560699 2175 net.cpp:131] Top shape: 10 (10)
  421. I0803 12:10:38.560703 2175 net.cpp:139] Memory required for data: 6183520
  422. I0803 12:10:38.560712 2175 layer_factory.hpp:77] Creating layer conv1
  423. I0803 12:10:38.560735 2175 net.cpp:86] Creating Layer conv1
  424. I0803 12:10:38.560742 2175 net.cpp:408] conv1 <- data
  425. I0803 12:10:38.560756 2175 net.cpp:382] conv1 -> conv1
  426. I0803 12:10:40.714188 2175 net.cpp:124] Setting up conv1
  427. I0803 12:10:40.714244 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  428. I0803 12:10:40.714278 2175 net.cpp:139] Memory required for data: 138097760
  429. I0803 12:10:40.714332 2175 layer_factory.hpp:77] Creating layer bn1
  430. I0803 12:10:40.714362 2175 net.cpp:86] Creating Layer bn1
  431. I0803 12:10:40.714376 2175 net.cpp:408] bn1 <- conv1
  432. I0803 12:10:40.714392 2175 net.cpp:369] bn1 -> conv1 (in-place)
  433. I0803 12:10:40.715009 2175 net.cpp:124] Setting up bn1
  434. I0803 12:10:40.715030 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  435. I0803 12:10:40.715041 2175 net.cpp:139] Memory required for data: 270012000
  436. I0803 12:10:40.715067 2175 layer_factory.hpp:77] Creating layer scale1
  437. I0803 12:10:40.715128 2175 net.cpp:86] Creating Layer scale1
  438. I0803 12:10:40.715138 2175 net.cpp:408] scale1 <- conv1
  439. I0803 12:10:40.715153 2175 net.cpp:369] scale1 -> conv1 (in-place)
  440. I0803 12:10:40.715266 2175 layer_factory.hpp:77] Creating layer scale1
  441. I0803 12:10:40.715734 2175 net.cpp:124] Setting up scale1
  442. I0803 12:10:40.715755 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  443. I0803 12:10:40.715765 2175 net.cpp:139] Memory required for data: 401926240
  444. I0803 12:10:40.715781 2175 layer_factory.hpp:77] Creating layer relu1
  445. I0803 12:10:40.715798 2175 net.cpp:86] Creating Layer relu1
  446. I0803 12:10:40.715809 2175 net.cpp:408] relu1 <- conv1
  447. I0803 12:10:40.715822 2175 net.cpp:369] relu1 -> conv1 (in-place)
  448. I0803 12:10:40.716305 2175 net.cpp:124] Setting up relu1
  449. I0803 12:10:40.716326 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  450. I0803 12:10:40.716336 2175 net.cpp:139] Memory required for data: 533840480
  451. I0803 12:10:40.716346 2175 layer_factory.hpp:77] Creating layer norm1
  452. I0803 12:10:40.716363 2175 net.cpp:86] Creating Layer norm1
  453. I0803 12:10:40.716374 2175 net.cpp:408] norm1 <- conv1
  454. I0803 12:10:40.716389 2175 net.cpp:382] norm1 -> norm1
  455. I0803 12:10:40.717486 2175 net.cpp:124] Setting up norm1
  456. I0803 12:10:40.717516 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  457. I0803 12:10:40.717525 2175 net.cpp:139] Memory required for data: 665754720
  458. I0803 12:10:40.717535 2175 layer_factory.hpp:77] Creating layer pool1
  459. I0803 12:10:40.717552 2175 net.cpp:86] Creating Layer pool1
  460. I0803 12:10:40.717563 2175 net.cpp:408] pool1 <- norm1
  461. I0803 12:10:40.717581 2175 net.cpp:382] pool1 -> pool1
  462. I0803 12:10:40.717699 2175 net.cpp:124] Setting up pool1
  463. I0803 12:10:40.717720 2175 net.cpp:131] Top shape: 10 64 113 113 (8172160)
  464. I0803 12:10:40.717730 2175 net.cpp:139] Memory required for data: 698443360
  465. I0803 12:10:40.717738 2175 layer_factory.hpp:77] Creating layer conv2
  466. I0803 12:10:40.717762 2175 net.cpp:86] Creating Layer conv2
  467. I0803 12:10:40.717774 2175 net.cpp:408] conv2 <- pool1
  468. I0803 12:10:40.717787 2175 net.cpp:382] conv2 -> conv2
  469. I0803 12:10:40.733384 2175 net.cpp:124] Setting up conv2
  470. I0803 12:10:40.733410 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  471. I0803 12:10:40.733415 2175 net.cpp:139] Memory required for data: 829197920
  472. I0803 12:10:40.733430 2175 layer_factory.hpp:77] Creating layer relu2
  473. I0803 12:10:40.733443 2175 net.cpp:86] Creating Layer relu2
  474. I0803 12:10:40.733448 2175 net.cpp:408] relu2 <- conv2
  475. I0803 12:10:40.733456 2175 net.cpp:369] relu2 -> conv2 (in-place)
  476. I0803 12:10:40.733969 2175 net.cpp:124] Setting up relu2
  477. I0803 12:10:40.733983 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  478. I0803 12:10:40.733985 2175 net.cpp:139] Memory required for data: 959952480
  479. I0803 12:10:40.733990 2175 layer_factory.hpp:77] Creating layer norm2
  480. I0803 12:10:40.733999 2175 net.cpp:86] Creating Layer norm2
  481. I0803 12:10:40.734002 2175 net.cpp:408] norm2 <- conv2
  482. I0803 12:10:40.734007 2175 net.cpp:382] norm2 -> norm2
  483. I0803 12:10:40.734232 2175 net.cpp:124] Setting up norm2
  484. I0803 12:10:40.734241 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  485. I0803 12:10:40.734244 2175 net.cpp:139] Memory required for data: 1090707040
  486. I0803 12:10:40.734248 2175 layer_factory.hpp:77] Creating layer pool2
  487. I0803 12:10:40.734269 2175 net.cpp:86] Creating Layer pool2
  488. I0803 12:10:40.734274 2175 net.cpp:408] pool2 <- norm2
  489. I0803 12:10:40.734280 2175 net.cpp:382] pool2 -> pool2
  490. I0803 12:10:40.734321 2175 net.cpp:124] Setting up pool2
  491. I0803 12:10:40.734329 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  492. I0803 12:10:40.734333 2175 net.cpp:139] Memory required for data: 1122819680
  493. I0803 12:10:40.734335 2175 layer_factory.hpp:77] Creating layer conv3
  494. I0803 12:10:40.734345 2175 net.cpp:86] Creating Layer conv3
  495. I0803 12:10:40.734349 2175 net.cpp:408] conv3 <- pool2
  496. I0803 12:10:40.734355 2175 net.cpp:382] conv3 -> conv3
  497. I0803 12:10:40.748950 2175 net.cpp:124] Setting up conv3
  498. I0803 12:10:40.748970 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  499. I0803 12:10:40.748973 2175 net.cpp:139] Memory required for data: 1170988640
  500. I0803 12:10:40.749014 2175 layer_factory.hpp:77] Creating layer relu3
  501. I0803 12:10:40.749023 2175 net.cpp:86] Creating Layer relu3
  502. I0803 12:10:40.749027 2175 net.cpp:408] relu3 <- conv3
  503. I0803 12:10:40.749034 2175 net.cpp:369] relu3 -> conv3 (in-place)
  504. I0803 12:10:40.749212 2175 net.cpp:124] Setting up relu3
  505. I0803 12:10:40.749219 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  506. I0803 12:10:40.749222 2175 net.cpp:139] Memory required for data: 1219157600
  507. I0803 12:10:40.749224 2175 layer_factory.hpp:77] Creating layer conv4
  508. I0803 12:10:40.749233 2175 net.cpp:86] Creating Layer conv4
  509. I0803 12:10:40.749235 2175 net.cpp:408] conv4 <- conv3
  510. I0803 12:10:40.749240 2175 net.cpp:382] conv4 -> conv4
  511. I0803 12:10:40.756110 2175 net.cpp:124] Setting up conv4
  512. I0803 12:10:40.756124 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  513. I0803 12:10:40.756127 2175 net.cpp:139] Memory required for data: 1267326560
  514. I0803 12:10:40.756134 2175 layer_factory.hpp:77] Creating layer relu4
  515. I0803 12:10:40.756139 2175 net.cpp:86] Creating Layer relu4
  516. I0803 12:10:40.756141 2175 net.cpp:408] relu4 <- conv4
  517. I0803 12:10:40.756145 2175 net.cpp:369] relu4 -> conv4 (in-place)
  518. I0803 12:10:40.756290 2175 net.cpp:124] Setting up relu4
  519. I0803 12:10:40.756297 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  520. I0803 12:10:40.756299 2175 net.cpp:139] Memory required for data: 1315495520
  521. I0803 12:10:40.756301 2175 layer_factory.hpp:77] Creating layer conv5
  522. I0803 12:10:40.756307 2175 net.cpp:86] Creating Layer conv5
  523. I0803 12:10:40.756310 2175 net.cpp:408] conv5 <- conv4
  524. I0803 12:10:40.756314 2175 net.cpp:382] conv5 -> conv5
  525. I0803 12:10:40.761378 2175 net.cpp:124] Setting up conv5
  526. I0803 12:10:40.761390 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  527. I0803 12:10:40.761394 2175 net.cpp:139] Memory required for data: 1347608160
  528. I0803 12:10:40.761399 2175 layer_factory.hpp:77] Creating layer relu5
  529. I0803 12:10:40.761406 2175 net.cpp:86] Creating Layer relu5
  530. I0803 12:10:40.761409 2175 net.cpp:408] relu5 <- conv5
  531. I0803 12:10:40.761414 2175 net.cpp:369] relu5 -> conv5 (in-place)
  532. I0803 12:10:40.761597 2175 net.cpp:124] Setting up relu5
  533. I0803 12:10:40.761603 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  534. I0803 12:10:40.761605 2175 net.cpp:139] Memory required for data: 1379720800
  535. I0803 12:10:40.761607 2175 layer_factory.hpp:77] Creating layer pool5
  536. I0803 12:10:40.761612 2175 net.cpp:86] Creating Layer pool5
  537. I0803 12:10:40.761615 2175 net.cpp:408] pool5 <- conv5
  538. I0803 12:10:40.761618 2175 net.cpp:382] pool5 -> pool5
  539. I0803 12:10:40.761677 2175 net.cpp:124] Setting up pool5
  540. I0803 12:10:40.761682 2175 net.cpp:131] Top shape: 10 256 28 28 (2007040)
  541. I0803 12:10:40.761698 2175 net.cpp:139] Memory required for data: 1387748960
  542. I0803 12:10:40.761700 2175 layer_factory.hpp:77] Creating layer fc6
  543. I0803 12:10:40.761706 2175 net.cpp:86] Creating Layer fc6
  544. I0803 12:10:40.761709 2175 net.cpp:408] fc6 <- pool5
  545. I0803 12:10:40.761713 2175 net.cpp:382] fc6 -> fc6
  546. I0803 12:10:47.059787 2175 net.cpp:124] Setting up fc6
  547. I0803 12:10:47.059804 2175 net.cpp:131] Top shape: 10 4096 (40960)
  548. I0803 12:10:47.059806 2175 net.cpp:139] Memory required for data: 1387912800
  549. I0803 12:10:47.059830 2175 layer_factory.hpp:77] Creating layer relu6
  550. I0803 12:10:47.059855 2175 net.cpp:86] Creating Layer relu6
  551. I0803 12:10:47.059859 2175 net.cpp:408] relu6 <- fc6
  552. I0803 12:10:47.059864 2175 net.cpp:369] relu6 -> fc6 (in-place)
  553. I0803 12:10:47.060097 2175 net.cpp:124] Setting up relu6
  554. I0803 12:10:47.060103 2175 net.cpp:131] Top shape: 10 4096 (40960)
  555. I0803 12:10:47.060106 2175 net.cpp:139] Memory required for data: 1388076640
  556. I0803 12:10:47.060108 2175 layer_factory.hpp:77] Creating layer drop6
  557. I0803 12:10:47.060132 2175 net.cpp:86] Creating Layer drop6
  558. I0803 12:10:47.060135 2175 net.cpp:408] drop6 <- fc6
  559. I0803 12:10:47.060139 2175 net.cpp:369] drop6 -> fc6 (in-place)
  560. I0803 12:10:47.060207 2175 net.cpp:124] Setting up drop6
  561. I0803 12:10:47.060211 2175 net.cpp:131] Top shape: 10 4096 (40960)
  562. I0803 12:10:47.060256 2175 net.cpp:139] Memory required for data: 1388240480
  563. I0803 12:10:47.060261 2175 layer_factory.hpp:77] Creating layer fc7
  564. I0803 12:10:47.060281 2175 net.cpp:86] Creating Layer fc7
  565. I0803 12:10:47.060283 2175 net.cpp:408] fc7 <- fc6
  566. I0803 12:10:47.060287 2175 net.cpp:382] fc7 -> fc7
  567. I0803 12:10:47.189757 2175 net.cpp:124] Setting up fc7
  568. I0803 12:10:47.189774 2175 net.cpp:131] Top shape: 10 4096 (40960)
  569. I0803 12:10:47.189776 2175 net.cpp:139] Memory required for data: 1388404320
  570. I0803 12:10:47.189797 2175 layer_factory.hpp:77] Creating layer relu7
  571. I0803 12:10:47.189805 2175 net.cpp:86] Creating Layer relu7
  572. I0803 12:10:47.189807 2175 net.cpp:408] relu7 <- fc7
  573. I0803 12:10:47.189826 2175 net.cpp:369] relu7 -> fc7 (in-place)
  574. I0803 12:10:47.190233 2175 net.cpp:124] Setting up relu7
  575. I0803 12:10:47.190240 2175 net.cpp:131] Top shape: 10 4096 (40960)
  576. I0803 12:10:47.190243 2175 net.cpp:139] Memory required for data: 1388568160
  577. I0803 12:10:47.190244 2175 layer_factory.hpp:77] Creating layer drop7
  578. I0803 12:10:47.190269 2175 net.cpp:86] Creating Layer drop7
  579. I0803 12:10:47.190274 2175 net.cpp:408] drop7 <- fc7
  580. I0803 12:10:47.190299 2175 net.cpp:369] drop7 -> fc7 (in-place)
  581. I0803 12:10:47.190349 2175 net.cpp:124] Setting up drop7
  582. I0803 12:10:47.190354 2175 net.cpp:131] Top shape: 10 4096 (40960)
  583. I0803 12:10:47.190356 2175 net.cpp:139] Memory required for data: 1388732000
  584. I0803 12:10:47.190358 2175 layer_factory.hpp:77] Creating layer fc8
  585. I0803 12:10:47.190362 2175 net.cpp:86] Creating Layer fc8
  586. I0803 12:10:47.190364 2175 net.cpp:408] fc8 <- fc7
  587. I0803 12:10:47.190369 2175 net.cpp:382] fc8 -> fc8
  588. I0803 12:10:47.222895 2175 net.cpp:124] Setting up fc8
  589. I0803 12:10:47.222913 2175 net.cpp:131] Top shape: 10 1000 (10000)
  590. I0803 12:10:47.222915 2175 net.cpp:139] Memory required for data: 1388772000
  591. I0803 12:10:47.222936 2175 layer_factory.hpp:77] Creating layer loss
  592. I0803 12:10:47.222942 2175 net.cpp:86] Creating Layer loss
  593. I0803 12:10:47.222946 2175 net.cpp:408] loss <- fc8
  594. I0803 12:10:47.222950 2175 net.cpp:408] loss <- label
  595. I0803 12:10:47.222970 2175 net.cpp:382] loss -> loss
  596. I0803 12:10:47.222978 2175 layer_factory.hpp:77] Creating layer loss
  597. I0803 12:10:47.223251 2175 net.cpp:124] Setting up loss
  598. I0803 12:10:47.223258 2175 net.cpp:131] Top shape: (1)
  599. I0803 12:10:47.223259 2175 net.cpp:134] with loss weight 1
  600. I0803 12:10:47.223287 2175 net.cpp:139] Memory required for data: 1388772004
  601. I0803 12:10:47.223291 2175 net.cpp:200] loss needs backward computation.
  602. I0803 12:10:47.223296 2175 net.cpp:200] fc8 needs backward computation.
  603. I0803 12:10:47.223299 2175 net.cpp:200] drop7 needs backward computation.
  604. I0803 12:10:47.223316 2175 net.cpp:200] relu7 needs backward computation.
  605. I0803 12:10:47.223320 2175 net.cpp:200] fc7 needs backward computation.
  606. I0803 12:10:47.223321 2175 net.cpp:200] drop6 needs backward computation.
  607. I0803 12:10:47.223341 2175 net.cpp:200] relu6 needs backward computation.
  608. I0803 12:10:47.223343 2175 net.cpp:200] fc6 needs backward computation.
  609. I0803 12:10:47.223345 2175 net.cpp:200] pool5 needs backward computation.
  610. I0803 12:10:47.223348 2175 net.cpp:200] relu5 needs backward computation.
  611. I0803 12:10:47.223363 2175 net.cpp:200] conv5 needs backward computation.
  612. I0803 12:10:47.223366 2175 net.cpp:200] relu4 needs backward computation.
  613. I0803 12:10:47.223368 2175 net.cpp:200] conv4 needs backward computation.
  614. I0803 12:10:47.223386 2175 net.cpp:200] relu3 needs backward computation.
  615. I0803 12:10:47.223389 2175 net.cpp:200] conv3 needs backward computation.
  616. I0803 12:10:47.223392 2175 net.cpp:200] pool2 needs backward computation.
  617. I0803 12:10:47.223395 2175 net.cpp:200] norm2 needs backward computation.
  618. I0803 12:10:47.223398 2175 net.cpp:200] relu2 needs backward computation.
  619. I0803 12:10:47.223402 2175 net.cpp:200] conv2 needs backward computation.
  620. I0803 12:10:47.223404 2175 net.cpp:200] pool1 needs backward computation.
  621. I0803 12:10:47.223407 2175 net.cpp:200] norm1 needs backward computation.
  622. I0803 12:10:47.223409 2175 net.cpp:200] relu1 needs backward computation.
  623. I0803 12:10:47.223428 2175 net.cpp:200] scale1 needs backward computation.
  624. I0803 12:10:47.223430 2175 net.cpp:200] bn1 needs backward computation.
  625. I0803 12:10:47.223433 2175 net.cpp:200] conv1 needs backward computation.
  626. I0803 12:10:47.223436 2175 net.cpp:202] data does not need backward computation.
  627. I0803 12:10:47.223439 2175 net.cpp:244] This network produces output loss
  628. I0803 12:10:47.223453 2175 net.cpp:257] Network initialization done.
  629. I0803 12:10:47.223675 2175 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  630. I0803 12:10:47.223681 2175 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  631. I0803 12:10:47.223701 2175 solver.cpp:173] Creating test net (#0) specified by net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  632. I0803 12:10:47.223724 2175 net.cpp:296] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
  633. I0803 12:10:47.223853 2175 net.cpp:53] Initializing net from parameters:
  634. name: "AlexNet"
  635. state {
  636. phase: TEST
  637. }
  638. layer {
  639. name: "data"
  640. type: "Data"
  641. top: "data"
  642. top: "label"
  643. include {
  644. phase: TEST
  645. }
  646. transform_param {
  647. mirror: false
  648. crop_size: 227
  649. mean_file: "examples/Mydataset_test_lmdb/mean_imagetest.binaryproto"
  650. }
  651. data_param {
  652. source: "examples/Mydataset_test_lmdb"
  653. batch_size: 10
  654. backend: LMDB
  655. }
  656. }
  657. layer {
  658. name: "conv1"
  659. type: "Convolution"
  660. bottom: "data"
  661. top: "conv1"
  662. param {
  663. lr_mult: 1
  664. }
  665. convolution_param {
  666. num_output: 64
  667. bias_term: true
  668. pad: 1
  669. kernel_size: 3
  670. stride: 1
  671. weight_filler {
  672. type: "xavier"
  673. }
  674. }
  675. }
  676. layer {
  677. name: "bn1"
  678. type: "BatchNorm"
  679. bottom: "conv1"
  680. top: "conv1"
  681. param {
  682. lr_mult: 0
  683. decay_mult: 0
  684. }
  685. param {
  686. lr_mult: 0
  687. decay_mult: 0
  688. }
  689. param {
  690. lr_mult: 0
  691. decay_mult: 0
  692. }
  693. }
  694. layer {
  695. name: "scale1"
  696. type: "Scale"
  697. bottom: "conv1"
  698. top: "conv1"
  699. scale_param {
  700. bias_term: true
  701. }
  702. }
  703. layer {
  704. name: "relu1"
  705. type: "ReLU"
  706. bottom: "conv1"
  707. top: "conv1"
  708. }
  709. layer {
  710. name: "norm1"
  711. type: "LRN"
  712. bottom: "conv1"
  713. top: "norm1"
  714. lrn_param {
  715. local_size: 5
  716. alpha: 0.0001
  717. beta: 0.75
  718. }
  719. }
  720. layer {
  721. name: "pool1"
  722. type: "Pooling"
  723. bottom: "norm1"
  724. top: "pool1"
  725. pooling_param {
  726. pool: MAX
  727. kernel_size: 3
  728. stride: 2
  729. }
  730. }
  731. layer {
  732. name: "conv2"
  733. type: "Convolution"
  734. bottom: "pool1"
  735. top: "conv2"
  736. param {
  737. lr_mult: 1
  738. decay_mult: 1
  739. }
  740. param {
  741. lr_mult: 2
  742. decay_mult: 0
  743. }
  744. convolution_param {
  745. num_output: 256
  746. pad: 2
  747. kernel_size: 5
  748. group: 2
  749. weight_filler {
  750. type: "gaussian"
  751. std: 0.01
  752. }
  753. bias_filler {
  754. type: "constant"
  755. value: 0.1
  756. }
  757. }
  758. }
  759. layer {
  760. name: "relu2"
  761. type: "ReLU"
  762. bottom: "conv2"
  763. top: "conv2"
  764. }
  765. layer {
  766. name: "norm2"
  767. type: "LRN"
  768. bottom: "conv2"
  769. top: "norm2"
  770. lrn_param {
  771. local_size: 5
  772. alpha: 0.0001
  773. beta: 0.75
  774. }
  775. }
  776. layer {
  777. name: "pool2"
  778. type: "Pooling"
  779. bottom: "norm2"
  780. top: "pool2"
  781. pooling_param {
  782. pool: MAX
  783. kernel_size: 3
  784. stride: 2
  785. }
  786. }
  787. layer {
  788. name: "conv3"
  789. type: "Convolution"
  790. bottom: "pool2"
  791. top: "conv3"
  792. param {
  793. lr_mult: 1
  794. decay_mult: 1
  795. }
  796. param {
  797. lr_mult: 2
  798. decay_mult: 0
  799. }
  800. convolution_param {
  801. num_output: 384
  802. pad: 1
  803. kernel_size: 3
  804. weight_filler {
  805. type: "gaussian"
  806. std: 0.01
  807. }
  808. bias_filler {
  809. type: "constant"
  810. value: 0
  811. }
  812. }
  813. }
  814. layer {
  815. name: "relu3"
  816. type: "ReLU"
  817. bottom: "conv3"
  818. top: "conv3"
  819. }
  820. layer {
  821. name: "conv4"
  822. type: "Convolution"
  823. bottom: "conv3"
  824. top: "conv4"
  825. param {
  826. lr_mult: 1
  827. decay_mult: 1
  828. }
  829. param {
  830. lr_mult: 2
  831. decay_mult: 0
  832. }
  833. convolution_param {
  834. num_output: 384
  835. pad: 1
  836. kernel_size: 3
  837. group: 2
  838. weight_filler {
  839. type: "gaussian"
  840. std: 0.01
  841. }
  842. bias_filler {
  843. type: "constant"
  844. value: 0.1
  845. }
  846. }
  847. }
  848. layer {
  849. name: "relu4"
  850. type: "ReLU"
  851. bottom: "conv4"
  852. top: "conv4"
  853. }
  854. layer {
  855. name: "conv5"
  856. type: "Convolution"
  857. bottom: "conv4"
  858. top: "conv5"
  859. param {
  860. lr_mult: 1
  861. decay_mult: 1
  862. }
  863. param {
  864. lr_mult: 2
  865. decay_mult: 0
  866. }
  867. convolution_param {
  868. num_output: 256
  869. pad: 1
  870. kernel_size: 3
  871. group: 2
  872. weight_filler {
  873. type: "gaussian"
  874. std: 0.01
  875. }
  876. bias_filler {
  877. type: "constant"
  878. value: 0.1
  879. }
  880. }
  881. }
  882. layer {
  883. name: "relu5"
  884. type: "ReLU"
  885. bottom: "conv5"
  886. top: "conv5"
  887. }
  888. layer {
  889. name: "pool5"
  890. type: "Pooling"
  891. bottom: "conv5"
  892. top: "pool5"
  893. pooling_param {
  894. pool: MAX
  895. kernel_size: 3
  896. stride: 2
  897. }
  898. }
  899. layer {
  900. name: "fc6"
  901. type: "InnerProduct"
  902. bottom: "pool5"
  903. top: "fc6"
  904. param {
  905. lr_mult: 1
  906. decay_mult: 1
  907. }
  908. param {
  909. lr_mult: 2
  910. decay_mult: 0
  911. }
  912. inner_product_param {
  913. num_output: 4096
  914. weight_filler {
  915. type: "gaussian"
  916. std: 0.005
  917. }
  918. bias_filler {
  919. type: "constant"
  920. value: 0.1
  921. }
  922. }
  923. }
  924. layer {
  925. name: "relu6"
  926. type: "ReLU"
  927. bottom: "fc6"
  928. top: "fc6"
  929. }
  930. layer {
  931. name: "drop6"
  932. type: "Dropout"
  933. bottom: "fc6"
  934. top: "fc6"
  935. dropout_param {
  936. dropout_ratio: 0.5
  937. }
  938. }
  939. layer {
  940. name: "fc7"
  941. type: "InnerProduct"
  942. bottom: "fc6"
  943. top: "fc7"
  944. param {
  945. lr_mult: 1
  946. decay_mult: 1
  947. }
  948. param {
  949. lr_mult: 2
  950. decay_mult: 0
  951. }
  952. inner_product_param {
  953. num_output: 4096
  954. weight_filler {
  955. type: "gaussian"
  956. std: 0.005
  957. }
  958. bias_filler {
  959. type: "constant"
  960. value: 0.1
  961. }
  962. }
  963. }
  964. layer {
  965. name: "relu7"
  966. type: "ReLU"
  967. bottom: "fc7"
  968. top: "fc7"
  969. }
  970. layer {
  971. name: "drop7"
  972. type: "Dropout"
  973. bottom: "fc7"
  974. top: "fc7"
  975. dropout_param {
  976. dropout_ratio: 0.5
  977. }
  978. }
  979. layer {
  980. name: "fc8"
  981. type: "InnerProduct"
  982. bottom: "fc7"
  983. top: "fc8"
  984. param {
  985. lr_mult: 1
  986. decay_mult: 1
  987. }
  988. param {
  989. lr_mult: 2
  990. decay_mult: 0
  991. }
  992. inner_product_param {
  993. num_output: 1000
  994. weight_filler {
  995. type: "gaussian"
  996. std: 0.01
  997. }
  998. bias_filler {
  999. type: "constant"
  1000. value: 0
  1001. }
  1002. }
  1003. }
  1004. layer {
  1005. name: "accuracy"
  1006. type: "Accuracy"
  1007. bottom: "fc8"
  1008. bottom: "label"
  1009. top: "accuracy"
  1010. include {
  1011. phase: TEST
  1012. }
  1013. }
  1014. layer {
  1015. name: "loss"
  1016. type: "SoftmaxWithLoss"
  1017. bottom: "fc8"
  1018. bottom: "label"
  1019. top: "loss"
  1020. }
  1021. I0803 12:10:47.224076 2175 layer_factory.hpp:77] Creating layer data
  1022. I0803 12:10:47.269927 2175 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_test_lmdb
  1023. I0803 12:10:47.280226 2175 net.cpp:86] Creating Layer data
  1024. I0803 12:10:47.280279 2175 net.cpp:382] data -> data
  1025. I0803 12:10:47.280303 2175 net.cpp:382] data -> label
  1026. I0803 12:10:47.280324 2175 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_test_lmdb/mean_imagetest.binaryproto
  1027. I0803 12:10:47.313472 2175 data_layer.cpp:45] output data size: 10,3,227,227
  1028. I0803 12:10:47.334664 2175 net.cpp:124] Setting up data
  1029. I0803 12:10:47.334702 2175 net.cpp:131] Top shape: 10 3 227 227 (1545870)
  1030. I0803 12:10:47.334712 2175 net.cpp:131] Top shape: 10 (10)
  1031. I0803 12:10:47.334718 2175 net.cpp:139] Memory required for data: 6183520
  1032. I0803 12:10:47.334731 2175 layer_factory.hpp:77] Creating layer label_data_1_split
  1033. I0803 12:10:47.334753 2175 net.cpp:86] Creating Layer label_data_1_split
  1034. I0803 12:10:47.334761 2175 net.cpp:408] label_data_1_split <- label
  1035. I0803 12:10:47.334791 2175 net.cpp:382] label_data_1_split -> label_data_1_split_0
  1036. I0803 12:10:47.334811 2175 net.cpp:382] label_data_1_split -> label_data_1_split_1
  1037. I0803 12:10:47.335036 2175 net.cpp:124] Setting up label_data_1_split
  1038. I0803 12:10:47.335067 2175 net.cpp:131] Top shape: 10 (10)
  1039. I0803 12:10:47.335075 2175 net.cpp:131] Top shape: 10 (10)
  1040. I0803 12:10:47.335080 2175 net.cpp:139] Memory required for data: 6183600
  1041. I0803 12:10:47.335088 2175 layer_factory.hpp:77] Creating layer conv1
  1042. I0803 12:10:47.335110 2175 net.cpp:86] Creating Layer conv1
  1043. I0803 12:10:47.335119 2175 net.cpp:408] conv1 <- data
  1044. I0803 12:10:47.335131 2175 net.cpp:382] conv1 -> conv1
  1045. I0803 12:10:47.338120 2175 net.cpp:124] Setting up conv1
  1046. I0803 12:10:47.338152 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  1047. I0803 12:10:47.338197 2175 net.cpp:139] Memory required for data: 138097840
  1048. I0803 12:10:47.338222 2175 layer_factory.hpp:77] Creating layer bn1
  1049. I0803 12:10:47.338238 2175 net.cpp:86] Creating Layer bn1
  1050. I0803 12:10:47.338245 2175 net.cpp:408] bn1 <- conv1
  1051. I0803 12:10:47.338284 2175 net.cpp:369] bn1 -> conv1 (in-place)
  1052. I0803 12:10:47.338843 2175 net.cpp:124] Setting up bn1
  1053. I0803 12:10:47.338861 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  1054. I0803 12:10:47.338870 2175 net.cpp:139] Memory required for data: 270012080
  1055. I0803 12:10:47.338889 2175 layer_factory.hpp:77] Creating layer scale1
  1056. I0803 12:10:47.338907 2175 net.cpp:86] Creating Layer scale1
  1057. I0803 12:10:47.338914 2175 net.cpp:408] scale1 <- conv1
  1058. I0803 12:10:47.338927 2175 net.cpp:369] scale1 -> conv1 (in-place)
  1059. I0803 12:10:47.339010 2175 layer_factory.hpp:77] Creating layer scale1
  1060. I0803 12:10:47.339423 2175 net.cpp:124] Setting up scale1
  1061. I0803 12:10:47.339439 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  1062. I0803 12:10:47.339448 2175 net.cpp:139] Memory required for data: 401926320
  1063. I0803 12:10:47.339462 2175 layer_factory.hpp:77] Creating layer relu1
  1064. I0803 12:10:47.339473 2175 net.cpp:86] Creating Layer relu1
  1065. I0803 12:10:47.339485 2175 net.cpp:408] relu1 <- conv1
  1066. I0803 12:10:47.339496 2175 net.cpp:369] relu1 -> conv1 (in-place)
  1067. I0803 12:10:47.339871 2175 net.cpp:124] Setting up relu1
  1068. I0803 12:10:47.339887 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  1069. I0803 12:10:47.339893 2175 net.cpp:139] Memory required for data: 533840560
  1070. I0803 12:10:47.339901 2175 layer_factory.hpp:77] Creating layer norm1
  1071. I0803 12:10:47.339912 2175 net.cpp:86] Creating Layer norm1
  1072. I0803 12:10:47.339921 2175 net.cpp:408] norm1 <- conv1
  1073. I0803 12:10:47.339933 2175 net.cpp:382] norm1 -> norm1
  1074. I0803 12:10:47.340322 2175 net.cpp:124] Setting up norm1
  1075. I0803 12:10:47.340337 2175 net.cpp:131] Top shape: 10 64 227 227 (32978560)
  1076. I0803 12:10:47.340343 2175 net.cpp:139] Memory required for data: 665754800
  1077. I0803 12:10:47.340349 2175 layer_factory.hpp:77] Creating layer pool1
  1078. I0803 12:10:47.340360 2175 net.cpp:86] Creating Layer pool1
  1079. I0803 12:10:47.340366 2175 net.cpp:408] pool1 <- norm1
  1080. I0803 12:10:47.340379 2175 net.cpp:382] pool1 -> pool1
  1081. I0803 12:10:47.340472 2175 net.cpp:124] Setting up pool1
  1082. I0803 12:10:47.340487 2175 net.cpp:131] Top shape: 10 64 113 113 (8172160)
  1083. I0803 12:10:47.340492 2175 net.cpp:139] Memory required for data: 698443440
  1084. I0803 12:10:47.340498 2175 layer_factory.hpp:77] Creating layer conv2
  1085. I0803 12:10:47.340517 2175 net.cpp:86] Creating Layer conv2
  1086. I0803 12:10:47.340523 2175 net.cpp:408] conv2 <- pool1
  1087. I0803 12:10:47.340538 2175 net.cpp:382] conv2 -> conv2
  1088. I0803 12:10:47.347864 2175 net.cpp:124] Setting up conv2
  1089. I0803 12:10:47.347887 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  1090. I0803 12:10:47.347894 2175 net.cpp:139] Memory required for data: 829198000
  1091. I0803 12:10:47.347909 2175 layer_factory.hpp:77] Creating layer relu2
  1092. I0803 12:10:47.347920 2175 net.cpp:86] Creating Layer relu2
  1093. I0803 12:10:47.347924 2175 net.cpp:408] relu2 <- conv2
  1094. I0803 12:10:47.347932 2175 net.cpp:369] relu2 -> conv2 (in-place)
  1095. I0803 12:10:47.348479 2175 net.cpp:124] Setting up relu2
  1096. I0803 12:10:47.348491 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  1097. I0803 12:10:47.348496 2175 net.cpp:139] Memory required for data: 959952560
  1098. I0803 12:10:47.348502 2175 layer_factory.hpp:77] Creating layer norm2
  1099. I0803 12:10:47.348511 2175 net.cpp:86] Creating Layer norm2
  1100. I0803 12:10:47.348516 2175 net.cpp:408] norm2 <- conv2
  1101. I0803 12:10:47.348523 2175 net.cpp:382] norm2 -> norm2
  1102. I0803 12:10:47.348768 2175 net.cpp:124] Setting up norm2
  1103. I0803 12:10:47.348778 2175 net.cpp:131] Top shape: 10 256 113 113 (32688640)
  1104. I0803 12:10:47.348783 2175 net.cpp:139] Memory required for data: 1090707120
  1105. I0803 12:10:47.348786 2175 layer_factory.hpp:77] Creating layer pool2
  1106. I0803 12:10:47.348793 2175 net.cpp:86] Creating Layer pool2
  1107. I0803 12:10:47.348798 2175 net.cpp:408] pool2 <- norm2
  1108. I0803 12:10:47.348803 2175 net.cpp:382] pool2 -> pool2
  1109. I0803 12:10:47.348870 2175 net.cpp:124] Setting up pool2
  1110. I0803 12:10:47.348878 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  1111. I0803 12:10:47.348881 2175 net.cpp:139] Memory required for data: 1122819760
  1112. I0803 12:10:47.348886 2175 layer_factory.hpp:77] Creating layer conv3
  1113. I0803 12:10:47.348894 2175 net.cpp:86] Creating Layer conv3
  1114. I0803 12:10:47.348899 2175 net.cpp:408] conv3 <- pool2
  1115. I0803 12:10:47.348906 2175 net.cpp:382] conv3 -> conv3
  1116. I0803 12:10:47.360764 2175 net.cpp:124] Setting up conv3
  1117. I0803 12:10:47.360780 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  1118. I0803 12:10:47.360783 2175 net.cpp:139] Memory required for data: 1170988720
  1119. I0803 12:10:47.360791 2175 layer_factory.hpp:77] Creating layer relu3
  1120. I0803 12:10:47.360798 2175 net.cpp:86] Creating Layer relu3
  1121. I0803 12:10:47.360801 2175 net.cpp:408] relu3 <- conv3
  1122. I0803 12:10:47.360808 2175 net.cpp:369] relu3 -> conv3 (in-place)
  1123. I0803 12:10:47.360967 2175 net.cpp:124] Setting up relu3
  1124. I0803 12:10:47.360975 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  1125. I0803 12:10:47.360977 2175 net.cpp:139] Memory required for data: 1219157680
  1126. I0803 12:10:47.360980 2175 layer_factory.hpp:77] Creating layer conv4
  1127. I0803 12:10:47.360988 2175 net.cpp:86] Creating Layer conv4
  1128. I0803 12:10:47.360991 2175 net.cpp:408] conv4 <- conv3
  1129. I0803 12:10:47.360996 2175 net.cpp:382] conv4 -> conv4
  1130. I0803 12:10:47.368546 2175 net.cpp:124] Setting up conv4
  1131. I0803 12:10:47.368559 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  1132. I0803 12:10:47.368561 2175 net.cpp:139] Memory required for data: 1267326640
  1133. I0803 12:10:47.368582 2175 layer_factory.hpp:77] Creating layer relu4
  1134. I0803 12:10:47.368605 2175 net.cpp:86] Creating Layer relu4
  1135. I0803 12:10:47.368609 2175 net.cpp:408] relu4 <- conv4
  1136. I0803 12:10:47.368614 2175 net.cpp:369] relu4 -> conv4 (in-place)
  1137. I0803 12:10:47.368779 2175 net.cpp:124] Setting up relu4
  1138. I0803 12:10:47.368785 2175 net.cpp:131] Top shape: 10 384 56 56 (12042240)
  1139. I0803 12:10:47.368801 2175 net.cpp:139] Memory required for data: 1315495600
  1140. I0803 12:10:47.368803 2175 layer_factory.hpp:77] Creating layer conv5
  1141. I0803 12:10:47.368827 2175 net.cpp:86] Creating Layer conv5
  1142. I0803 12:10:47.368831 2175 net.cpp:408] conv5 <- conv4
  1143. I0803 12:10:47.368835 2175 net.cpp:382] conv5 -> conv5
  1144. I0803 12:10:47.375073 2175 net.cpp:124] Setting up conv5
  1145. I0803 12:10:47.375088 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  1146. I0803 12:10:47.375092 2175 net.cpp:139] Memory required for data: 1347608240
  1147. I0803 12:10:47.375113 2175 layer_factory.hpp:77] Creating layer relu5
  1148. I0803 12:10:47.375118 2175 net.cpp:86] Creating Layer relu5
  1149. I0803 12:10:47.375135 2175 net.cpp:408] relu5 <- conv5
  1150. I0803 12:10:47.375141 2175 net.cpp:369] relu5 -> conv5 (in-place)
  1151. I0803 12:10:47.375282 2175 net.cpp:124] Setting up relu5
  1152. I0803 12:10:47.375288 2175 net.cpp:131] Top shape: 10 256 56 56 (8028160)
  1153. I0803 12:10:47.375289 2175 net.cpp:139] Memory required for data: 1379720880
  1154. I0803 12:10:47.375291 2175 layer_factory.hpp:77] Creating layer pool5
  1155. I0803 12:10:47.375310 2175 net.cpp:86] Creating Layer pool5
  1156. I0803 12:10:47.375313 2175 net.cpp:408] pool5 <- conv5
  1157. I0803 12:10:47.375331 2175 net.cpp:382] pool5 -> pool5
  1158. I0803 12:10:47.375396 2175 net.cpp:124] Setting up pool5
  1159. I0803 12:10:47.375399 2175 net.cpp:131] Top shape: 10 256 28 28 (2007040)
  1160. I0803 12:10:47.375401 2175 net.cpp:139] Memory required for data: 1387749040
  1161. I0803 12:10:47.375403 2175 layer_factory.hpp:77] Creating layer fc6
  1162. I0803 12:10:47.375408 2175 net.cpp:86] Creating Layer fc6
  1163. I0803 12:10:47.375411 2175 net.cpp:408] fc6 <- pool5
  1164. I0803 12:10:47.375414 2175 net.cpp:382] fc6 -> fc6
  1165. I0803 12:10:53.939147 2175 net.cpp:124] Setting up fc6
  1166. I0803 12:10:53.970624 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1167. I0803 12:10:53.970654 2175 net.cpp:139] Memory required for data: 1387912880
  1168. I0803 12:10:54.020473 2175 layer_factory.hpp:77] Creating layer relu6
  1169. I0803 12:10:54.029845 2175 net.cpp:86] Creating Layer relu6
  1170. I0803 12:10:54.029889 2175 net.cpp:408] relu6 <- fc6
  1171. I0803 12:10:54.029919 2175 net.cpp:369] relu6 -> fc6 (in-place)
  1172. I0803 12:10:54.952005 2175 net.cpp:124] Setting up relu6
  1173. I0803 12:10:54.952059 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1174. I0803 12:10:54.952069 2175 net.cpp:139] Memory required for data: 1388076720
  1175. I0803 12:10:54.952081 2175 layer_factory.hpp:77] Creating layer drop6
  1176. I0803 12:10:54.952183 2175 net.cpp:86] Creating Layer drop6
  1177. I0803 12:10:54.952199 2175 net.cpp:408] drop6 <- fc6
  1178. I0803 12:10:54.952221 2175 net.cpp:369] drop6 -> fc6 (in-place)
  1179. I0803 12:10:54.952363 2175 net.cpp:124] Setting up drop6
  1180. I0803 12:10:54.952380 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1181. I0803 12:10:54.952388 2175 net.cpp:139] Memory required for data: 1388240560
  1182. I0803 12:10:54.952399 2175 layer_factory.hpp:77] Creating layer fc7
  1183. I0803 12:10:54.952422 2175 net.cpp:86] Creating Layer fc7
  1184. I0803 12:10:54.952433 2175 net.cpp:408] fc7 <- fc6
  1185. I0803 12:10:54.952450 2175 net.cpp:382] fc7 -> fc7
  1186. I0803 12:10:55.168879 2175 net.cpp:124] Setting up fc7
  1187. I0803 12:10:55.168905 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1188. I0803 12:10:55.168921 2175 net.cpp:139] Memory required for data: 1388404400
  1189. I0803 12:10:55.168929 2175 layer_factory.hpp:77] Creating layer relu7
  1190. I0803 12:10:55.168952 2175 net.cpp:86] Creating Layer relu7
  1191. I0803 12:10:55.168956 2175 net.cpp:408] relu7 <- fc7
  1192. I0803 12:10:55.168962 2175 net.cpp:369] relu7 -> fc7 (in-place)
  1193. I0803 12:10:55.213413 2175 net.cpp:124] Setting up relu7
  1194. I0803 12:10:55.213459 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1195. I0803 12:10:55.213466 2175 net.cpp:139] Memory required for data: 1388568240
  1196. I0803 12:10:55.213479 2175 layer_factory.hpp:77] Creating layer drop7
  1197. I0803 12:10:55.213507 2175 net.cpp:86] Creating Layer drop7
  1198. I0803 12:10:55.213522 2175 net.cpp:408] drop7 <- fc7
  1199. I0803 12:10:55.213541 2175 net.cpp:369] drop7 -> fc7 (in-place)
  1200. I0803 12:10:55.213641 2175 net.cpp:124] Setting up drop7
  1201. I0803 12:10:55.213655 2175 net.cpp:131] Top shape: 10 4096 (40960)
  1202. I0803 12:10:55.213662 2175 net.cpp:139] Memory required for data: 1388732080
  1203. I0803 12:10:55.213670 2175 layer_factory.hpp:77] Creating layer fc8
  1204. I0803 12:10:55.213688 2175 net.cpp:86] Creating Layer fc8
  1205. I0803 12:10:55.213696 2175 net.cpp:408] fc8 <- fc7
  1206. I0803 12:10:55.213712 2175 net.cpp:382] fc8 -> fc8
  1207. I0803 12:10:55.271644 2175 net.cpp:124] Setting up fc8
  1208. I0803 12:10:55.271661 2175 net.cpp:131] Top shape: 10 1000 (10000)
  1209. I0803 12:10:55.271664 2175 net.cpp:139] Memory required for data: 1388772080
  1210. I0803 12:10:55.271672 2175 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
  1211. I0803 12:10:55.322099 2175 net.cpp:86] Creating Layer fc8_fc8_0_split
  1212. I0803 12:10:55.322120 2175 net.cpp:408] fc8_fc8_0_split <- fc8
  1213. I0803 12:10:55.322129 2175 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_0
  1214. I0803 12:10:55.322139 2175 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_1
  1215. I0803 12:10:55.322221 2175 net.cpp:124] Setting up fc8_fc8_0_split
  1216. I0803 12:10:55.322228 2175 net.cpp:131] Top shape: 10 1000 (10000)
  1217. I0803 12:10:55.322232 2175 net.cpp:131] Top shape: 10 1000 (10000)
  1218. I0803 12:10:55.322234 2175 net.cpp:139] Memory required for data: 1388852080
  1219. I0803 12:10:55.322237 2175 layer_factory.hpp:77] Creating layer accuracy
  1220. I0803 12:10:55.327911 2175 net.cpp:86] Creating Layer accuracy
  1221. I0803 12:10:55.327934 2175 net.cpp:408] accuracy <- fc8_fc8_0_split_0
  1222. I0803 12:10:55.327941 2175 net.cpp:408] accuracy <- label_data_1_split_0
  1223. I0803 12:10:55.327951 2175 net.cpp:382] accuracy -> accuracy
  1224. I0803 12:10:55.327967 2175 net.cpp:124] Setting up accuracy
  1225. I0803 12:10:55.327975 2175 net.cpp:131] Top shape: (1)
  1226. I0803 12:10:55.327978 2175 net.cpp:139] Memory required for data: 1388852084
  1227. I0803 12:10:55.327983 2175 layer_factory.hpp:77] Creating layer loss
  1228. I0803 12:10:55.336427 2175 net.cpp:86] Creating Layer loss
  1229. I0803 12:10:55.336453 2175 net.cpp:408] loss <- fc8_fc8_0_split_1
  1230. I0803 12:10:55.336462 2175 net.cpp:408] loss <- label_data_1_split_1
  1231. I0803 12:10:55.336469 2175 net.cpp:382] loss -> loss
  1232. I0803 12:10:55.341104 2175 layer_factory.hpp:77] Creating layer loss
  1233. I0803 12:10:55.341733 2175 net.cpp:124] Setting up loss
  1234. I0803 12:10:55.341776 2175 net.cpp:131] Top shape: (1)
  1235. I0803 12:10:55.341780 2175 net.cpp:134] with loss weight 1
  1236. I0803 12:10:55.353785 2175 net.cpp:139] Memory required for data: 1388852088
  1237. I0803 12:10:55.353804 2175 net.cpp:200] loss needs backward computation.
  1238. I0803 12:10:55.353821 2175 net.cpp:202] accuracy does not need backward computation.
  1239. I0803 12:10:55.353832 2175 net.cpp:200] fc8_fc8_0_split needs backward computation.
  1240. I0803 12:10:55.353842 2175 net.cpp:200] fc8 needs backward computation.
  1241. I0803 12:10:55.353850 2175 net.cpp:200] drop7 needs backward computation.
  1242. I0803 12:10:55.353859 2175 net.cpp:200] relu7 needs backward computation.
  1243. I0803 12:10:55.353868 2175 net.cpp:200] fc7 needs backward computation.
  1244. I0803 12:10:55.353873 2175 net.cpp:200] drop6 needs backward computation.
  1245. I0803 12:10:55.353881 2175 net.cpp:200] relu6 needs backward computation.
  1246. I0803 12:10:55.353888 2175 net.cpp:200] fc6 needs backward computation.
  1247. I0803 12:10:55.353895 2175 net.cpp:200] pool5 needs backward computation.
  1248. I0803 12:10:55.353904 2175 net.cpp:200] relu5 needs backward computation.
  1249. I0803 12:10:55.353911 2175 net.cpp:200] conv5 needs backward computation.
  1250. I0803 12:10:55.353920 2175 net.cpp:200] relu4 needs backward computation.
  1251. I0803 12:10:55.353927 2175 net.cpp:200] conv4 needs backward computation.
  1252. I0803 12:10:55.353935 2175 net.cpp:200] relu3 needs backward computation.
  1253. I0803 12:10:55.353943 2175 net.cpp:200] conv3 needs backward computation.
  1254. I0803 12:10:55.353955 2175 net.cpp:200] pool2 needs backward computation.
  1255. I0803 12:10:55.353962 2175 net.cpp:200] norm2 needs backward computation.
  1256. I0803 12:10:55.353971 2175 net.cpp:200] relu2 needs backward computation.
  1257. I0803 12:10:55.353978 2175 net.cpp:200] conv2 needs backward computation.
  1258. I0803 12:10:55.353988 2175 net.cpp:200] pool1 needs backward computation.
  1259. I0803 12:10:55.354003 2175 net.cpp:200] norm1 needs backward computation.
  1260. I0803 12:10:55.354012 2175 net.cpp:200] relu1 needs backward computation.
  1261. I0803 12:10:55.354019 2175 net.cpp:200] scale1 needs backward computation.
  1262. I0803 12:10:55.354027 2175 net.cpp:200] bn1 needs backward computation.
  1263. I0803 12:10:55.354035 2175 net.cpp:200] conv1 needs backward computation.
  1264. I0803 12:10:55.354045 2175 net.cpp:202] label_data_1_split does not need backward computation.
  1265. I0803 12:10:55.354055 2175 net.cpp:202] data does not need backward computation.
  1266. I0803 12:10:55.354063 2175 net.cpp:244] This network produces output accuracy
  1267. I0803 12:10:55.354075 2175 net.cpp:244] This network produces output loss
  1268. I0803 12:10:55.354123 2175 net.cpp:257] Network initialization done.
  1269. I0803 12:10:55.385907 2175 solver.cpp:56] Solver scaffolding done.
  1270. I0803 12:10:55.402374 2175 caffe.cpp:248] Starting Optimization
  1271. I0803 12:10:55.402422 2175 solver.cpp:273] Solving AlexNet
  1272. I0803 12:10:55.402431 2175 solver.cpp:274] Learning Rate Policy: step
  1273. I0803 12:10:55.789188 2175 solver.cpp:331] Iteration 0, Testing net (#0)
  1274. I0803 12:10:57.610282 2175 blocking_queue.cpp:49] Waiting for data
  1275. I0803 12:11:36.445051 2228 data_layer.cpp:73] Restarting data prefetching from start.
  1276. I0803 12:11:36.613025 2175 solver.cpp:398] Test net output #0: accuracy = 0
  1277. I0803 12:11:36.613050 2175 solver.cpp:398] Test net output #1: loss = 6.80939 (* 1 = 6.80939 loss)
  1278. F0803 12:11:36.613711 2175 syncedmem.cpp:71] Check failed: error == cudaSuccess (2 vs. 0) out of memory
  1279. *** Check failure stack trace: ***
  1280. @ 0x7f8d2b64d5cd google::LogMessage::Fail()
  1281. @ 0x7f8d2b64f433 google::LogMessage::SendToLog()
  1282. @ 0x7f8d2b64d15b google::LogMessage::Flush()
  1283. @ 0x7f8d2b64fe1e google::LogMessageFatal::~LogMessageFatal()
  1284. @ 0x7f8d2bc84ad8 caffe::SyncedMemory::mutable_gpu_data()
  1285. @ 0x7f8d2bdf90f2 caffe::Blob<>::mutable_gpu_data()
  1286. @ 0x7f8d2be2ce48 caffe::CuDNNConvolutionLayer<>::Forward_gpu()
  1287. @ 0x7f8d2bc938b1 caffe::Net<>::ForwardFromTo()
  1288. @ 0x7f8d2bc939b7 caffe::Net<>::Forward()
  1289. @ 0x7f8d2bdef348 caffe::Solver<>::Step()
  1290. @ 0x7f8d2bdefeea caffe::Solver<>::Solve()
  1291. @ 0x40aeb4 train()
  1292. @ 0x4075a8 main
  1293. @ 0x7f8d29de4830 __libc_start_main
  1294. @ 0x407e79 _start
  1295. @ (nil) (unknown)
  1296. Aborted (core dumped)
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
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!