API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Aug 2nd, 2017
172
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 45.53 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. I0802 16:51:20.823194 2858 caffe.cpp:218] Using GPUs 0
  3. I0802 16:51:20.838037 2858 caffe.cpp:223] GPU 0: GeForce GTX 1070
  4. I0802 16:51:21.037343 2858 solver.cpp:44] Initializing solver from parameters:
  5. test_iter: 56
  6. test_interval: 28
  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. I0802 16:51:21.037497 2858 solver.cpp:87] Creating training net from net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  26. I0802 16:51:21.037703 2858 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  27. I0802 16:51:21.037711 2858 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  28. I0802 16:51:21.037802 2858 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer data
  29. I0802 16:51:21.037811 2858 net.cpp:296] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
  30. I0802 16:51:21.037987 2858 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. transform_param {
  46. mirror: true
  47. crop_size: 227
  48. mean_file: "examples/Mydataset_train_lmdb/mean_imagetest.binaryproto"
  49. }
  50. data_param {
  51. source: "examples/Mydataset_train_lmdb"
  52. batch_size: 256
  53. backend: LMDB
  54. }
  55. }
  56. layer {
  57. name: "conv1"
  58. type: "Convolution"
  59. bottom: "data"
  60. top: "conv1"
  61. param {
  62. lr_mult: 1
  63. }
  64. convolution_param {
  65. num_output: 64
  66. bias_term: true
  67. pad: 1
  68. kernel_size: 3
  69. stride: 1
  70. weight_filler {
  71. type: "xavier"
  72. }
  73. }
  74. }
  75. layer {
  76. name: "bn1"
  77. type: "BatchNorm"
  78. bottom: "conv1"
  79. top: "conv1"
  80. param {
  81. lr_mult: 0
  82. decay_mult: 0
  83. }
  84. param {
  85. lr_mult: 0
  86. decay_mult: 0
  87. }
  88. param {
  89. lr_mult: 0
  90. decay_mult: 0
  91. }
  92. }
  93. layer {
  94. name: "scale1"
  95. type: "Scale"
  96. bottom: "conv1"
  97. top: "conv1"
  98. scale_param {
  99. bias_term: true
  100. }
  101. }
  102. layer {
  103. name: "relu1"
  104. type: "ReLU"
  105. bottom: "conv1"
  106. top: "conv1"
  107. }
  108. layer {
  109. name: "norm1"
  110. type: "LRN"
  111. bottom: "conv1"
  112. top: "norm1"
  113. lrn_param {
  114. local_size: 5
  115. alpha: 0.0001
  116. beta: 0.75
  117. }
  118. }
  119. layer {
  120. name: "pool1"
  121. type: "Pooling"
  122. bottom: "norm1"
  123. top: "pool1"
  124. pooling_param {
  125. pool: MAX
  126. kernel_size: 3
  127. stride: 2
  128. }
  129. }
  130. layer {
  131. name: "conv2"
  132. type: "Convolution"
  133. bottom: "pool1"
  134. top: "conv2"
  135. param {
  136. lr_mult: 1
  137. decay_mult: 1
  138. }
  139. param {
  140. lr_mult: 2
  141. decay_mult: 0
  142. }
  143. convolution_param {
  144. num_output: 256
  145. pad: 2
  146. kernel_size: 5
  147. group: 2
  148. weight_filler {
  149. type: "gaussian"
  150. std: 0.01
  151. }
  152. bias_filler {
  153. type: "constant"
  154. value: 0.1
  155. }
  156. }
  157. }
  158. layer {
  159. name: "relu2"
  160. type: "ReLU"
  161. bottom: "conv2"
  162. top: "conv2"
  163. }
  164. layer {
  165. name: "norm2"
  166. type: "LRN"
  167. bottom: "conv2"
  168. top: "norm2"
  169. lrn_param {
  170. local_size: 5
  171. alpha: 0.0001
  172. beta: 0.75
  173. }
  174. }
  175. layer {
  176. name: "pool2"
  177. type: "Pooling"
  178. bottom: "norm2"
  179. top: "pool2"
  180. pooling_param {
  181. pool: MAX
  182. kernel_size: 3
  183. stride: 2
  184. }
  185. }
  186. layer {
  187. name: "conv3"
  188. type: "Convolution"
  189. bottom: "pool2"
  190. top: "conv3"
  191. param {
  192. lr_mult: 1
  193. decay_mult: 1
  194. }
  195. param {
  196. lr_mult: 2
  197. decay_mult: 0
  198. }
  199. convolution_param {
  200. num_output: 384
  201. pad: 1
  202. kernel_size: 3
  203. weight_filler {
  204. type: "gaussian"
  205. std: 0.01
  206. }
  207. bias_filler {
  208. type: "constant"
  209. value: 0
  210. }
  211. }
  212. }
  213. layer {
  214. name: "relu3"
  215. type: "ReLU"
  216. bottom: "conv3"
  217. top: "conv3"
  218. }
  219. layer {
  220. name: "conv4"
  221. type: "Convolution"
  222. bottom: "conv3"
  223. top: "conv4"
  224. param {
  225. lr_mult: 1
  226. decay_mult: 1
  227. }
  228. param {
  229. lr_mult: 2
  230. decay_mult: 0
  231. }
  232. convolution_param {
  233. num_output: 384
  234. pad: 1
  235. kernel_size: 3
  236. group: 2
  237. weight_filler {
  238. type: "gaussian"
  239. std: 0.01
  240. }
  241. bias_filler {
  242. type: "constant"
  243. value: 0.1
  244. }
  245. }
  246. }
  247. layer {
  248. name: "relu4"
  249. type: "ReLU"
  250. bottom: "conv4"
  251. top: "conv4"
  252. }
  253. layer {
  254. name: "conv5"
  255. type: "Convolution"
  256. bottom: "conv4"
  257. top: "conv5"
  258. param {
  259. lr_mult: 1
  260. decay_mult: 1
  261. }
  262. param {
  263. lr_mult: 2
  264. decay_mult: 0
  265. }
  266. convolution_param {
  267. num_output: 256
  268. pad: 1
  269. kernel_size: 3
  270. group: 2
  271. weight_filler {
  272. type: "gaussian"
  273. std: 0.01
  274. }
  275. bias_filler {
  276. type: "constant"
  277. value: 0.1
  278. }
  279. }
  280. }
  281. layer {
  282. name: "relu5"
  283. type: "ReLU"
  284. bottom: "conv5"
  285. top: "conv5"
  286. }
  287. layer {
  288. name: "pool5"
  289. type: "Pooling"
  290. bottom: "conv5"
  291. top: "pool5"
  292. pooling_param {
  293. pool: MAX
  294. kernel_size: 3
  295. stride: 2
  296. }
  297. }
  298. layer {
  299. name: "fc6"
  300. type: "InnerProduct"
  301. bottom: "pool5"
  302. top: "fc6"
  303. param {
  304. lr_mult: 1
  305. decay_mult: 1
  306. }
  307. param {
  308. lr_mult: 2
  309. decay_mult: 0
  310. }
  311. inner_product_param {
  312. num_output: 4096
  313. weight_filler {
  314. type: "gaussian"
  315. std: 0.005
  316. }
  317. bias_filler {
  318. type: "constant"
  319. value: 0.1
  320. }
  321. }
  322. }
  323. layer {
  324. name: "relu6"
  325. type: "ReLU"
  326. bottom: "fc6"
  327. top: "fc6"
  328. }
  329. layer {
  330. name: "drop6"
  331. type: "Dropout"
  332. bottom: "fc6"
  333. top: "fc6"
  334. dropout_param {
  335. dropout_ratio: 0.5
  336. }
  337. }
  338. layer {
  339. name: "fc7"
  340. type: "InnerProduct"
  341. bottom: "fc6"
  342. top: "fc7"
  343. param {
  344. lr_mult: 1
  345. decay_mult: 1
  346. }
  347. param {
  348. lr_mult: 2
  349. decay_mult: 0
  350. }
  351. inner_product_param {
  352. num_output: 4096
  353. weight_filler {
  354. type: "gaussian"
  355. std: 0.005
  356. }
  357. bias_filler {
  358. type: "constant"
  359. value: 0.1
  360. }
  361. }
  362. }
  363. layer {
  364. name: "relu7"
  365. type: "ReLU"
  366. bottom: "fc7"
  367. top: "fc7"
  368. }
  369. layer {
  370. name: "drop7"
  371. type: "Dropout"
  372. bottom: "fc7"
  373. top: "fc7"
  374. dropout_param {
  375. dropout_ratio: 0.5
  376. }
  377. }
  378. layer {
  379. name: "fc8"
  380. type: "InnerProduct"
  381. bottom: "fc7"
  382. top: "fc8"
  383. param {
  384. lr_mult: 1
  385. decay_mult: 1
  386. }
  387. param {
  388. lr_mult: 2
  389. decay_mult: 0
  390. }
  391. inner_product_param {
  392. num_output: 1000
  393. weight_filler {
  394. type: "gaussian"
  395. std: 0.01
  396. }
  397. bias_filler {
  398. type: "constant"
  399. value: 0
  400. }
  401. }
  402. }
  403. layer {
  404. name: "loss"
  405. type: "SoftmaxWithLoss"
  406. bottom: "fc8"
  407. bottom: "label"
  408. top: "loss"
  409. }
  410. I0802 16:51:21.052150 2858 layer_factory.hpp:77] Creating layer data
  411. I0802 16:51:21.061594 2858 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_train_lmdb
  412. I0802 16:51:21.091151 2858 net.cpp:86] Creating Layer data
  413. I0802 16:51:21.091212 2858 net.cpp:382] data -> data
  414. I0802 16:51:21.091265 2858 net.cpp:382] data -> label
  415. I0802 16:51:21.091301 2858 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_train_lmdb/mean_imagetest.binaryproto
  416. I0802 16:51:21.125758 2858 data_layer.cpp:45] output data size: 256,3,227,227
  417. I0802 16:51:21.312355 2858 net.cpp:124] Setting up data
  418. I0802 16:51:21.312403 2858 net.cpp:131] Top shape: 256 3 227 227 (39574272)
  419. I0802 16:51:21.312408 2858 net.cpp:131] Top shape: 256 (256)
  420. I0802 16:51:21.312433 2858 net.cpp:139] Memory required for data: 158298112
  421. I0802 16:51:21.312441 2858 layer_factory.hpp:77] Creating layer conv1
  422. I0802 16:51:21.312459 2858 net.cpp:86] Creating Layer conv1
  423. I0802 16:51:21.312463 2858 net.cpp:408] conv1 <- data
  424. I0802 16:51:21.312500 2858 net.cpp:382] conv1 -> conv1
  425. I0802 16:51:25.388608 2858 net.cpp:124] Setting up conv1
  426. I0802 16:51:25.388659 2858 net.cpp:131] Top shape: 256 64 227 227 (844251136)
  427. I0802 16:51:25.388669 2858 net.cpp:139] Memory required for data: 3535302656
  428. I0802 16:51:25.388715 2858 layer_factory.hpp:77] Creating layer bn1
  429. I0802 16:51:25.388741 2858 net.cpp:86] Creating Layer bn1
  430. I0802 16:51:25.388752 2858 net.cpp:408] bn1 <- conv1
  431. I0802 16:51:25.388772 2858 net.cpp:369] bn1 -> conv1 (in-place)
  432. I0802 16:51:25.389351 2858 net.cpp:124] Setting up bn1
  433. I0802 16:51:25.389370 2858 net.cpp:131] Top shape: 256 64 227 227 (844251136)
  434. I0802 16:51:25.389380 2858 net.cpp:139] Memory required for data: 6912307200
  435. I0802 16:51:25.389405 2858 layer_factory.hpp:77] Creating layer scale1
  436. I0802 16:51:25.389463 2858 net.cpp:86] Creating Layer scale1
  437. I0802 16:51:25.389472 2858 net.cpp:408] scale1 <- conv1
  438. I0802 16:51:25.389487 2858 net.cpp:369] scale1 -> conv1 (in-place)
  439. I0802 16:51:25.389588 2858 layer_factory.hpp:77] Creating layer scale1
  440. I0802 16:51:25.390061 2858 net.cpp:124] Setting up scale1
  441. I0802 16:51:25.390081 2858 net.cpp:131] Top shape: 256 64 227 227 (844251136)
  442. I0802 16:51:25.390090 2858 net.cpp:139] Memory required for data: 10289311744
  443. I0802 16:51:25.390106 2858 layer_factory.hpp:77] Creating layer relu1
  444. I0802 16:51:25.390122 2858 net.cpp:86] Creating Layer relu1
  445. I0802 16:51:25.390132 2858 net.cpp:408] relu1 <- conv1
  446. I0802 16:51:25.390148 2858 net.cpp:369] relu1 -> conv1 (in-place)
  447. I0802 16:51:25.390626 2858 net.cpp:124] Setting up relu1
  448. I0802 16:51:25.390648 2858 net.cpp:131] Top shape: 256 64 227 227 (844251136)
  449. I0802 16:51:25.390660 2858 net.cpp:139] Memory required for data: 13666316288
  450. I0802 16:51:25.390671 2858 layer_factory.hpp:77] Creating layer norm1
  451. I0802 16:51:25.390696 2858 net.cpp:86] Creating Layer norm1
  452. I0802 16:51:25.390707 2858 net.cpp:408] norm1 <- conv1
  453. I0802 16:51:25.390722 2858 net.cpp:382] norm1 -> norm1
  454. I0802 16:51:25.391862 2858 net.cpp:124] Setting up norm1
  455. I0802 16:51:25.391892 2858 net.cpp:131] Top shape: 256 64 227 227 (844251136)
  456. I0802 16:51:25.391901 2858 net.cpp:139] Memory required for data: 17043320832
  457. I0802 16:51:25.391908 2858 layer_factory.hpp:77] Creating layer pool1
  458. I0802 16:51:25.391926 2858 net.cpp:86] Creating Layer pool1
  459. I0802 16:51:25.391937 2858 net.cpp:408] pool1 <- norm1
  460. I0802 16:51:25.391955 2858 net.cpp:382] pool1 -> pool1
  461. I0802 16:51:25.392066 2858 net.cpp:124] Setting up pool1
  462. I0802 16:51:25.392094 2858 net.cpp:131] Top shape: 256 64 113 113 (209207296)
  463. I0802 16:51:25.392102 2858 net.cpp:139] Memory required for data: 17880150016
  464. I0802 16:51:25.392109 2858 layer_factory.hpp:77] Creating layer conv2
  465. I0802 16:51:25.392133 2858 net.cpp:86] Creating Layer conv2
  466. I0802 16:51:25.392145 2858 net.cpp:408] conv2 <- pool1
  467. I0802 16:51:25.392159 2858 net.cpp:382] conv2 -> conv2
  468. I0802 16:51:25.425178 2858 net.cpp:124] Setting up conv2
  469. I0802 16:51:25.425228 2858 net.cpp:131] Top shape: 256 256 113 113 (836829184)
  470. I0802 16:51:25.425236 2858 net.cpp:139] Memory required for data: 21227466752
  471. I0802 16:51:25.425261 2858 layer_factory.hpp:77] Creating layer relu2
  472. I0802 16:51:25.425282 2858 net.cpp:86] Creating Layer relu2
  473. I0802 16:51:25.425290 2858 net.cpp:408] relu2 <- conv2
  474. I0802 16:51:25.425305 2858 net.cpp:369] relu2 -> conv2 (in-place)
  475. I0802 16:51:25.426239 2858 net.cpp:124] Setting up relu2
  476. I0802 16:51:25.426262 2858 net.cpp:131] Top shape: 256 256 113 113 (836829184)
  477. I0802 16:51:25.426268 2858 net.cpp:139] Memory required for data: 24574783488
  478. I0802 16:51:25.426275 2858 layer_factory.hpp:77] Creating layer norm2
  479. I0802 16:51:25.426290 2858 net.cpp:86] Creating Layer norm2
  480. I0802 16:51:25.426297 2858 net.cpp:408] norm2 <- conv2
  481. I0802 16:51:25.426311 2858 net.cpp:382] norm2 -> norm2
  482. I0802 16:51:25.426684 2858 net.cpp:124] Setting up norm2
  483. I0802 16:51:25.426702 2858 net.cpp:131] Top shape: 256 256 113 113 (836829184)
  484. I0802 16:51:25.426707 2858 net.cpp:139] Memory required for data: 27922100224
  485. I0802 16:51:25.426712 2858 layer_factory.hpp:77] Creating layer pool2
  486. I0802 16:51:25.426723 2858 net.cpp:86] Creating Layer pool2
  487. I0802 16:51:25.426730 2858 net.cpp:408] pool2 <- norm2
  488. I0802 16:51:25.426743 2858 net.cpp:382] pool2 -> pool2
  489. I0802 16:51:25.426808 2858 net.cpp:124] Setting up pool2
  490. I0802 16:51:25.426821 2858 net.cpp:131] Top shape: 256 256 56 56 (205520896)
  491. I0802 16:51:25.426828 2858 net.cpp:139] Memory required for data: 28744183808
  492. I0802 16:51:25.426833 2858 layer_factory.hpp:77] Creating layer conv3
  493. I0802 16:51:25.426847 2858 net.cpp:86] Creating Layer conv3
  494. I0802 16:51:25.426856 2858 net.cpp:408] conv3 <- pool2
  495. I0802 16:51:25.426870 2858 net.cpp:382] conv3 -> conv3
  496. I0802 16:51:25.447731 2858 net.cpp:124] Setting up conv3
  497. I0802 16:51:25.447755 2858 net.cpp:131] Top shape: 256 384 56 56 (308281344)
  498. I0802 16:51:25.447782 2858 net.cpp:139] Memory required for data: 29977309184
  499. I0802 16:51:25.447793 2858 layer_factory.hpp:77] Creating layer relu3
  500. I0802 16:51:25.447803 2858 net.cpp:86] Creating Layer relu3
  501. I0802 16:51:25.447808 2858 net.cpp:408] relu3 <- conv3
  502. I0802 16:51:25.447816 2858 net.cpp:369] relu3 -> conv3 (in-place)
  503. I0802 16:51:25.448030 2858 net.cpp:124] Setting up relu3
  504. I0802 16:51:25.448040 2858 net.cpp:131] Top shape: 256 384 56 56 (308281344)
  505. I0802 16:51:25.448042 2858 net.cpp:139] Memory required for data: 31210434560
  506. I0802 16:51:25.448047 2858 layer_factory.hpp:77] Creating layer conv4
  507. I0802 16:51:25.448058 2858 net.cpp:86] Creating Layer conv4
  508. I0802 16:51:25.448061 2858 net.cpp:408] conv4 <- conv3
  509. I0802 16:51:25.448070 2858 net.cpp:382] conv4 -> conv4
  510. I0802 16:51:25.458153 2858 net.cpp:124] Setting up conv4
  511. I0802 16:51:25.458170 2858 net.cpp:131] Top shape: 256 384 56 56 (308281344)
  512. I0802 16:51:25.458173 2858 net.cpp:139] Memory required for data: 32443559936
  513. I0802 16:51:25.458181 2858 layer_factory.hpp:77] Creating layer relu4
  514. I0802 16:51:25.458189 2858 net.cpp:86] Creating Layer relu4
  515. I0802 16:51:25.458191 2858 net.cpp:408] relu4 <- conv4
  516. I0802 16:51:25.458195 2858 net.cpp:369] relu4 -> conv4 (in-place)
  517. I0802 16:51:25.458353 2858 net.cpp:124] Setting up relu4
  518. I0802 16:51:25.458360 2858 net.cpp:131] Top shape: 256 384 56 56 (308281344)
  519. I0802 16:51:25.458362 2858 net.cpp:139] Memory required for data: 33676685312
  520. I0802 16:51:25.458365 2858 layer_factory.hpp:77] Creating layer conv5
  521. I0802 16:51:25.458374 2858 net.cpp:86] Creating Layer conv5
  522. I0802 16:51:25.458377 2858 net.cpp:408] conv5 <- conv4
  523. I0802 16:51:25.458384 2858 net.cpp:382] conv5 -> conv5
  524. I0802 16:51:25.464239 2858 net.cpp:124] Setting up conv5
  525. I0802 16:51:25.464249 2858 net.cpp:131] Top shape: 256 256 56 56 (205520896)
  526. I0802 16:51:25.464252 2858 net.cpp:139] Memory required for data: 34498768896
  527. I0802 16:51:25.464257 2858 layer_factory.hpp:77] Creating layer relu5
  528. I0802 16:51:25.464265 2858 net.cpp:86] Creating Layer relu5
  529. I0802 16:51:25.464267 2858 net.cpp:408] relu5 <- conv5
  530. I0802 16:51:25.464272 2858 net.cpp:369] relu5 -> conv5 (in-place)
  531. I0802 16:51:25.464406 2858 net.cpp:124] Setting up relu5
  532. I0802 16:51:25.464411 2858 net.cpp:131] Top shape: 256 256 56 56 (205520896)
  533. I0802 16:51:25.464413 2858 net.cpp:139] Memory required for data: 35320852480
  534. I0802 16:51:25.464416 2858 layer_factory.hpp:77] Creating layer pool5
  535. I0802 16:51:25.464421 2858 net.cpp:86] Creating Layer pool5
  536. I0802 16:51:25.464422 2858 net.cpp:408] pool5 <- conv5
  537. I0802 16:51:25.464426 2858 net.cpp:382] pool5 -> pool5
  538. I0802 16:51:25.464458 2858 net.cpp:124] Setting up pool5
  539. I0802 16:51:25.464463 2858 net.cpp:131] Top shape: 256 256 28 28 (51380224)
  540. I0802 16:51:25.464465 2858 net.cpp:139] Memory required for data: 35526373376
  541. I0802 16:51:25.464468 2858 layer_factory.hpp:77] Creating layer fc6
  542. I0802 16:51:25.464473 2858 net.cpp:86] Creating Layer fc6
  543. I0802 16:51:25.464475 2858 net.cpp:408] fc6 <- pool5
  544. I0802 16:51:25.464479 2858 net.cpp:382] fc6 -> fc6
  545. I0802 16:51:31.718487 2858 net.cpp:124] Setting up fc6
  546. I0802 16:51:31.718506 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  547. I0802 16:51:31.718508 2858 net.cpp:139] Memory required for data: 35530567680
  548. I0802 16:51:31.718533 2858 layer_factory.hpp:77] Creating layer relu6
  549. I0802 16:51:31.718540 2858 net.cpp:86] Creating Layer relu6
  550. I0802 16:51:31.718544 2858 net.cpp:408] relu6 <- fc6
  551. I0802 16:51:31.718564 2858 net.cpp:369] relu6 -> fc6 (in-place)
  552. I0802 16:51:31.718791 2858 net.cpp:124] Setting up relu6
  553. I0802 16:51:31.718797 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  554. I0802 16:51:31.718799 2858 net.cpp:139] Memory required for data: 35534761984
  555. I0802 16:51:31.718801 2858 layer_factory.hpp:77] Creating layer drop6
  556. I0802 16:51:31.718806 2858 net.cpp:86] Creating Layer drop6
  557. I0802 16:51:31.718807 2858 net.cpp:408] drop6 <- fc6
  558. I0802 16:51:31.718825 2858 net.cpp:369] drop6 -> fc6 (in-place)
  559. I0802 16:51:31.718860 2858 net.cpp:124] Setting up drop6
  560. I0802 16:51:31.718896 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  561. I0802 16:51:31.718899 2858 net.cpp:139] Memory required for data: 35538956288
  562. I0802 16:51:31.718901 2858 layer_factory.hpp:77] Creating layer fc7
  563. I0802 16:51:31.718921 2858 net.cpp:86] Creating Layer fc7
  564. I0802 16:51:31.718924 2858 net.cpp:408] fc7 <- fc6
  565. I0802 16:51:31.718940 2858 net.cpp:382] fc7 -> fc7
  566. I0802 16:51:31.848628 2858 net.cpp:124] Setting up fc7
  567. I0802 16:51:31.848644 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  568. I0802 16:51:31.848646 2858 net.cpp:139] Memory required for data: 35543150592
  569. I0802 16:51:31.848654 2858 layer_factory.hpp:77] Creating layer relu7
  570. I0802 16:51:31.848675 2858 net.cpp:86] Creating Layer relu7
  571. I0802 16:51:31.848678 2858 net.cpp:408] relu7 <- fc7
  572. I0802 16:51:31.848697 2858 net.cpp:369] relu7 -> fc7 (in-place)
  573. I0802 16:51:31.849135 2858 net.cpp:124] Setting up relu7
  574. I0802 16:51:31.849146 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  575. I0802 16:51:31.849148 2858 net.cpp:139] Memory required for data: 35547344896
  576. I0802 16:51:31.849150 2858 layer_factory.hpp:77] Creating layer drop7
  577. I0802 16:51:31.849154 2858 net.cpp:86] Creating Layer drop7
  578. I0802 16:51:31.849171 2858 net.cpp:408] drop7 <- fc7
  579. I0802 16:51:31.849174 2858 net.cpp:369] drop7 -> fc7 (in-place)
  580. I0802 16:51:31.849221 2858 net.cpp:124] Setting up drop7
  581. I0802 16:51:31.849225 2858 net.cpp:131] Top shape: 256 4096 (1048576)
  582. I0802 16:51:31.849227 2858 net.cpp:139] Memory required for data: 35551539200
  583. I0802 16:51:31.849230 2858 layer_factory.hpp:77] Creating layer fc8
  584. I0802 16:51:31.849248 2858 net.cpp:86] Creating Layer fc8
  585. I0802 16:51:31.849251 2858 net.cpp:408] fc8 <- fc7
  586. I0802 16:51:31.849256 2858 net.cpp:382] fc8 -> fc8
  587. I0802 16:51:31.881816 2858 net.cpp:124] Setting up fc8
  588. I0802 16:51:31.881834 2858 net.cpp:131] Top shape: 256 1000 (256000)
  589. I0802 16:51:31.881835 2858 net.cpp:139] Memory required for data: 35552563200
  590. I0802 16:51:31.881857 2858 layer_factory.hpp:77] Creating layer loss
  591. I0802 16:51:31.881878 2858 net.cpp:86] Creating Layer loss
  592. I0802 16:51:31.881881 2858 net.cpp:408] loss <- fc8
  593. I0802 16:51:31.881898 2858 net.cpp:408] loss <- label
  594. I0802 16:51:31.881903 2858 net.cpp:382] loss -> loss
  595. I0802 16:51:31.881927 2858 layer_factory.hpp:77] Creating layer loss
  596. I0802 16:51:31.882881 2858 net.cpp:124] Setting up loss
  597. I0802 16:51:31.882890 2858 net.cpp:131] Top shape: (1)
  598. I0802 16:51:31.882894 2858 net.cpp:134] with loss weight 1
  599. I0802 16:51:31.882920 2858 net.cpp:139] Memory required for data: 35552563204
  600. I0802 16:51:31.882925 2858 net.cpp:200] loss needs backward computation.
  601. I0802 16:51:31.882930 2858 net.cpp:200] fc8 needs backward computation.
  602. I0802 16:51:31.882946 2858 net.cpp:200] drop7 needs backward computation.
  603. I0802 16:51:31.882949 2858 net.cpp:200] relu7 needs backward computation.
  604. I0802 16:51:31.882953 2858 net.cpp:200] fc7 needs backward computation.
  605. I0802 16:51:31.882971 2858 net.cpp:200] drop6 needs backward computation.
  606. I0802 16:51:31.882973 2858 net.cpp:200] relu6 needs backward computation.
  607. I0802 16:51:31.882977 2858 net.cpp:200] fc6 needs backward computation.
  608. I0802 16:51:31.882979 2858 net.cpp:200] pool5 needs backward computation.
  609. I0802 16:51:31.882982 2858 net.cpp:200] relu5 needs backward computation.
  610. I0802 16:51:31.882999 2858 net.cpp:200] conv5 needs backward computation.
  611. I0802 16:51:31.883002 2858 net.cpp:200] relu4 needs backward computation.
  612. I0802 16:51:31.883004 2858 net.cpp:200] conv4 needs backward computation.
  613. I0802 16:51:31.883008 2858 net.cpp:200] relu3 needs backward computation.
  614. I0802 16:51:31.883024 2858 net.cpp:200] conv3 needs backward computation.
  615. I0802 16:51:31.883028 2858 net.cpp:200] pool2 needs backward computation.
  616. I0802 16:51:31.883031 2858 net.cpp:200] norm2 needs backward computation.
  617. I0802 16:51:31.883034 2858 net.cpp:200] relu2 needs backward computation.
  618. I0802 16:51:31.883038 2858 net.cpp:200] conv2 needs backward computation.
  619. I0802 16:51:31.883041 2858 net.cpp:200] pool1 needs backward computation.
  620. I0802 16:51:31.883056 2858 net.cpp:200] norm1 needs backward computation.
  621. I0802 16:51:31.883087 2858 net.cpp:200] relu1 needs backward computation.
  622. I0802 16:51:31.883091 2858 net.cpp:200] scale1 needs backward computation.
  623. I0802 16:51:31.883095 2858 net.cpp:200] bn1 needs backward computation.
  624. I0802 16:51:31.883110 2858 net.cpp:200] conv1 needs backward computation.
  625. I0802 16:51:31.883112 2858 net.cpp:202] data does not need backward computation.
  626. I0802 16:51:31.883116 2858 net.cpp:244] This network produces output loss
  627. I0802 16:51:31.883142 2858 net.cpp:257] Network initialization done.
  628. I0802 16:51:31.883363 2858 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: examples/alexnetfinetune/alexnetsade-bn.prototxt
  629. I0802 16:51:31.883368 2858 upgrade_proto.cpp:80] Successfully upgraded batch norm layers using deprecated params.
  630. I0802 16:51:31.883388 2858 solver.cpp:173] Creating test net (#0) specified by net file: examples/alexnetfinetune/alexnetsade-bn.prototxt
  631. I0802 16:51:31.883411 2858 net.cpp:296] The NetState phase (1) differed from the phase (0) specified by a rule in layer data
  632. I0802 16:51:31.883577 2858 net.cpp:53] Initializing net from parameters:
  633. name: "AlexNet"
  634. state {
  635. phase: TEST
  636. }
  637. layer {
  638. name: "data"
  639. type: "Data"
  640. top: "data"
  641. top: "label"
  642. include {
  643. phase: TEST
  644. }
  645. transform_param {
  646. mirror: false
  647. crop_size: 227
  648. mean_file: "examples/Mydataset_test_lmdb/mean_imagetest.binaryproto"
  649. }
  650. data_param {
  651. source: "examples/Mydataset_test_lmdb"
  652. batch_size: 50
  653. backend: LMDB
  654. }
  655. }
  656. layer {
  657. name: "conv1"
  658. type: "Convolution"
  659. bottom: "data"
  660. top: "conv1"
  661. param {
  662. lr_mult: 1
  663. }
  664. convolution_param {
  665. num_output: 64
  666. bias_term: true
  667. pad: 1
  668. kernel_size: 3
  669. stride: 1
  670. weight_filler {
  671. type: "xavier"
  672. }
  673. }
  674. }
  675. layer {
  676. name: "bn1"
  677. type: "BatchNorm"
  678. bottom: "conv1"
  679. top: "conv1"
  680. param {
  681. lr_mult: 0
  682. decay_mult: 0
  683. }
  684. param {
  685. lr_mult: 0
  686. decay_mult: 0
  687. }
  688. param {
  689. lr_mult: 0
  690. decay_mult: 0
  691. }
  692. }
  693. layer {
  694. name: "scale1"
  695. type: "Scale"
  696. bottom: "conv1"
  697. top: "conv1"
  698. scale_param {
  699. bias_term: true
  700. }
  701. }
  702. layer {
  703. name: "relu1"
  704. type: "ReLU"
  705. bottom: "conv1"
  706. top: "conv1"
  707. }
  708. layer {
  709. name: "norm1"
  710. type: "LRN"
  711. bottom: "conv1"
  712. top: "norm1"
  713. lrn_param {
  714. local_size: 5
  715. alpha: 0.0001
  716. beta: 0.75
  717. }
  718. }
  719. layer {
  720. name: "pool1"
  721. type: "Pooling"
  722. bottom: "norm1"
  723. top: "pool1"
  724. pooling_param {
  725. pool: MAX
  726. kernel_size: 3
  727. stride: 2
  728. }
  729. }
  730. layer {
  731. name: "conv2"
  732. type: "Convolution"
  733. bottom: "pool1"
  734. top: "conv2"
  735. param {
  736. lr_mult: 1
  737. decay_mult: 1
  738. }
  739. param {
  740. lr_mult: 2
  741. decay_mult: 0
  742. }
  743. convolution_param {
  744. num_output: 256
  745. pad: 2
  746. kernel_size: 5
  747. group: 2
  748. weight_filler {
  749. type: "gaussian"
  750. std: 0.01
  751. }
  752. bias_filler {
  753. type: "constant"
  754. value: 0.1
  755. }
  756. }
  757. }
  758. layer {
  759. name: "relu2"
  760. type: "ReLU"
  761. bottom: "conv2"
  762. top: "conv2"
  763. }
  764. layer {
  765. name: "norm2"
  766. type: "LRN"
  767. bottom: "conv2"
  768. top: "norm2"
  769. lrn_param {
  770. local_size: 5
  771. alpha: 0.0001
  772. beta: 0.75
  773. }
  774. }
  775. layer {
  776. name: "pool2"
  777. type: "Pooling"
  778. bottom: "norm2"
  779. top: "pool2"
  780. pooling_param {
  781. pool: MAX
  782. kernel_size: 3
  783. stride: 2
  784. }
  785. }
  786. layer {
  787. name: "conv3"
  788. type: "Convolution"
  789. bottom: "pool2"
  790. top: "conv3"
  791. param {
  792. lr_mult: 1
  793. decay_mult: 1
  794. }
  795. param {
  796. lr_mult: 2
  797. decay_mult: 0
  798. }
  799. convolution_param {
  800. num_output: 384
  801. pad: 1
  802. kernel_size: 3
  803. weight_filler {
  804. type: "gaussian"
  805. std: 0.01
  806. }
  807. bias_filler {
  808. type: "constant"
  809. value: 0
  810. }
  811. }
  812. }
  813. layer {
  814. name: "relu3"
  815. type: "ReLU"
  816. bottom: "conv3"
  817. top: "conv3"
  818. }
  819. layer {
  820. name: "conv4"
  821. type: "Convolution"
  822. bottom: "conv3"
  823. top: "conv4"
  824. param {
  825. lr_mult: 1
  826. decay_mult: 1
  827. }
  828. param {
  829. lr_mult: 2
  830. decay_mult: 0
  831. }
  832. convolution_param {
  833. num_output: 384
  834. pad: 1
  835. kernel_size: 3
  836. group: 2
  837. weight_filler {
  838. type: "gaussian"
  839. std: 0.01
  840. }
  841. bias_filler {
  842. type: "constant"
  843. value: 0.1
  844. }
  845. }
  846. }
  847. layer {
  848. name: "relu4"
  849. type: "ReLU"
  850. bottom: "conv4"
  851. top: "conv4"
  852. }
  853. layer {
  854. name: "conv5"
  855. type: "Convolution"
  856. bottom: "conv4"
  857. top: "conv5"
  858. param {
  859. lr_mult: 1
  860. decay_mult: 1
  861. }
  862. param {
  863. lr_mult: 2
  864. decay_mult: 0
  865. }
  866. convolution_param {
  867. num_output: 256
  868. pad: 1
  869. kernel_size: 3
  870. group: 2
  871. weight_filler {
  872. type: "gaussian"
  873. std: 0.01
  874. }
  875. bias_filler {
  876. type: "constant"
  877. value: 0.1
  878. }
  879. }
  880. }
  881. layer {
  882. name: "relu5"
  883. type: "ReLU"
  884. bottom: "conv5"
  885. top: "conv5"
  886. }
  887. layer {
  888. name: "pool5"
  889. type: "Pooling"
  890. bottom: "conv5"
  891. top: "pool5"
  892. pooling_param {
  893. pool: MAX
  894. kernel_size: 3
  895. stride: 2
  896. }
  897. }
  898. layer {
  899. name: "fc6"
  900. type: "InnerProduct"
  901. bottom: "pool5"
  902. top: "fc6"
  903. param {
  904. lr_mult: 1
  905. decay_mult: 1
  906. }
  907. param {
  908. lr_mult: 2
  909. decay_mult: 0
  910. }
  911. inner_product_param {
  912. num_output: 4096
  913. weight_filler {
  914. type: "gaussian"
  915. std: 0.005
  916. }
  917. bias_filler {
  918. type: "constant"
  919. value: 0.1
  920. }
  921. }
  922. }
  923. layer {
  924. name: "relu6"
  925. type: "ReLU"
  926. bottom: "fc6"
  927. top: "fc6"
  928. }
  929. layer {
  930. name: "drop6"
  931. type: "Dropout"
  932. bottom: "fc6"
  933. top: "fc6"
  934. dropout_param {
  935. dropout_ratio: 0.5
  936. }
  937. }
  938. layer {
  939. name: "fc7"
  940. type: "InnerProduct"
  941. bottom: "fc6"
  942. top: "fc7"
  943. param {
  944. lr_mult: 1
  945. decay_mult: 1
  946. }
  947. param {
  948. lr_mult: 2
  949. decay_mult: 0
  950. }
  951. inner_product_param {
  952. num_output: 4096
  953. weight_filler {
  954. type: "gaussian"
  955. std: 0.005
  956. }
  957. bias_filler {
  958. type: "constant"
  959. value: 0.1
  960. }
  961. }
  962. }
  963. layer {
  964. name: "relu7"
  965. type: "ReLU"
  966. bottom: "fc7"
  967. top: "fc7"
  968. }
  969. layer {
  970. name: "drop7"
  971. type: "Dropout"
  972. bottom: "fc7"
  973. top: "fc7"
  974. dropout_param {
  975. dropout_ratio: 0.5
  976. }
  977. }
  978. layer {
  979. name: "fc8"
  980. type: "InnerProduct"
  981. bottom: "fc7"
  982. top: "fc8"
  983. param {
  984. lr_mult: 1
  985. decay_mult: 1
  986. }
  987. param {
  988. lr_mult: 2
  989. decay_mult: 0
  990. }
  991. inner_product_param {
  992. num_output: 1000
  993. weight_filler {
  994. type: "gaussian"
  995. std: 0.01
  996. }
  997. bias_filler {
  998. type: "constant"
  999. value: 0
  1000. }
  1001. }
  1002. }
  1003. layer {
  1004. name: "accuracy"
  1005. type: "Accuracy"
  1006. bottom: "fc8"
  1007. bottom: "label"
  1008. top: "accuracy"
  1009. include {
  1010. phase: TEST
  1011. }
  1012. }
  1013. layer {
  1014. name: "loss"
  1015. type: "SoftmaxWithLoss"
  1016. bottom: "fc8"
  1017. bottom: "label"
  1018. top: "loss"
  1019. }
  1020. I0802 16:51:31.883865 2858 layer_factory.hpp:77] Creating layer data
  1021. I0802 16:51:31.903520 2858 db_lmdb.cpp:35] Opened lmdb examples/Mydataset_test_lmdb
  1022. I0802 16:51:31.915539 2858 net.cpp:86] Creating Layer data
  1023. I0802 16:51:31.915576 2858 net.cpp:382] data -> data
  1024. I0802 16:51:31.915601 2858 net.cpp:382] data -> label
  1025. I0802 16:51:31.915618 2858 data_transformer.cpp:25] Loading mean file from: examples/Mydataset_test_lmdb/mean_imagetest.binaryproto
  1026. I0802 16:51:31.949586 2858 data_layer.cpp:45] output data size: 50,3,227,227
  1027. I0802 16:51:32.009131 2858 net.cpp:124] Setting up data
  1028. I0802 16:51:32.009177 2858 net.cpp:131] Top shape: 50 3 227 227 (7729350)
  1029. I0802 16:51:32.009181 2858 net.cpp:131] Top shape: 50 (50)
  1030. I0802 16:51:32.009186 2858 net.cpp:139] Memory required for data: 30917600
  1031. I0802 16:51:32.009212 2858 layer_factory.hpp:77] Creating layer label_data_1_split
  1032. I0802 16:51:32.009222 2858 net.cpp:86] Creating Layer label_data_1_split
  1033. I0802 16:51:32.009227 2858 net.cpp:408] label_data_1_split <- label
  1034. I0802 16:51:32.009232 2858 net.cpp:382] label_data_1_split -> label_data_1_split_0
  1035. I0802 16:51:32.009241 2858 net.cpp:382] label_data_1_split -> label_data_1_split_1
  1036. I0802 16:51:32.009376 2858 net.cpp:124] Setting up label_data_1_split
  1037. I0802 16:51:32.009400 2858 net.cpp:131] Top shape: 50 (50)
  1038. I0802 16:51:32.009403 2858 net.cpp:131] Top shape: 50 (50)
  1039. I0802 16:51:32.009407 2858 net.cpp:139] Memory required for data: 30918000
  1040. I0802 16:51:32.009428 2858 layer_factory.hpp:77] Creating layer conv1
  1041. I0802 16:51:32.009438 2858 net.cpp:86] Creating Layer conv1
  1042. I0802 16:51:32.009441 2858 net.cpp:408] conv1 <- data
  1043. I0802 16:51:32.009446 2858 net.cpp:382] conv1 -> conv1
  1044. I0802 16:51:32.012914 2858 net.cpp:124] Setting up conv1
  1045. I0802 16:51:32.012955 2858 net.cpp:131] Top shape: 50 64 227 227 (164892800)
  1046. I0802 16:51:32.012959 2858 net.cpp:139] Memory required for data: 690489200
  1047. I0802 16:51:32.012984 2858 layer_factory.hpp:77] Creating layer bn1
  1048. I0802 16:51:32.012991 2858 net.cpp:86] Creating Layer bn1
  1049. I0802 16:51:32.012995 2858 net.cpp:408] bn1 <- conv1
  1050. I0802 16:51:32.013000 2858 net.cpp:369] bn1 -> conv1 (in-place)
  1051. I0802 16:51:32.013191 2858 net.cpp:124] Setting up bn1
  1052. I0802 16:51:32.013196 2858 net.cpp:131] Top shape: 50 64 227 227 (164892800)
  1053. I0802 16:51:32.013213 2858 net.cpp:139] Memory required for data: 1350060400
  1054. I0802 16:51:32.013221 2858 layer_factory.hpp:77] Creating layer scale1
  1055. I0802 16:51:32.013226 2858 net.cpp:86] Creating Layer scale1
  1056. I0802 16:51:32.013228 2858 net.cpp:408] scale1 <- conv1
  1057. I0802 16:51:32.013233 2858 net.cpp:369] scale1 -> conv1 (in-place)
  1058. I0802 16:51:32.013275 2858 layer_factory.hpp:77] Creating layer scale1
  1059. I0802 16:51:32.013424 2858 net.cpp:124] Setting up scale1
  1060. I0802 16:51:32.013429 2858 net.cpp:131] Top shape: 50 64 227 227 (164892800)
  1061. I0802 16:51:32.013433 2858 net.cpp:139] Memory required for data: 2009631600
  1062. I0802 16:51:32.013438 2858 layer_factory.hpp:77] Creating layer relu1
  1063. I0802 16:51:32.013443 2858 net.cpp:86] Creating Layer relu1
  1064. I0802 16:51:32.013447 2858 net.cpp:408] relu1 <- conv1
  1065. I0802 16:51:32.013450 2858 net.cpp:369] relu1 -> conv1 (in-place)
  1066. I0802 16:51:32.013588 2858 net.cpp:124] Setting up relu1
  1067. I0802 16:51:32.013594 2858 net.cpp:131] Top shape: 50 64 227 227 (164892800)
  1068. I0802 16:51:32.013597 2858 net.cpp:139] Memory required for data: 2669202800
  1069. I0802 16:51:32.013600 2858 layer_factory.hpp:77] Creating layer norm1
  1070. I0802 16:51:32.013604 2858 net.cpp:86] Creating Layer norm1
  1071. I0802 16:51:32.013607 2858 net.cpp:408] norm1 <- conv1
  1072. I0802 16:51:32.013612 2858 net.cpp:382] norm1 -> norm1
  1073. I0802 16:51:32.013787 2858 net.cpp:124] Setting up norm1
  1074. I0802 16:51:32.013794 2858 net.cpp:131] Top shape: 50 64 227 227 (164892800)
  1075. I0802 16:51:32.013797 2858 net.cpp:139] Memory required for data: 3328774000
  1076. I0802 16:51:32.013800 2858 layer_factory.hpp:77] Creating layer pool1
  1077. I0802 16:51:32.013806 2858 net.cpp:86] Creating Layer pool1
  1078. I0802 16:51:32.013809 2858 net.cpp:408] pool1 <- norm1
  1079. I0802 16:51:32.013813 2858 net.cpp:382] pool1 -> pool1
  1080. I0802 16:51:32.013840 2858 net.cpp:124] Setting up pool1
  1081. I0802 16:51:32.013845 2858 net.cpp:131] Top shape: 50 64 113 113 (40860800)
  1082. I0802 16:51:32.013849 2858 net.cpp:139] Memory required for data: 3492217200
  1083. I0802 16:51:32.013851 2858 layer_factory.hpp:77] Creating layer conv2
  1084. I0802 16:51:32.013859 2858 net.cpp:86] Creating Layer conv2
  1085. I0802 16:51:32.013862 2858 net.cpp:408] conv2 <- pool1
  1086. I0802 16:51:32.013867 2858 net.cpp:382] conv2 -> conv2
  1087. I0802 16:51:32.017405 2858 net.cpp:124] Setting up conv2
  1088. I0802 16:51:32.017417 2858 net.cpp:131] Top shape: 50 256 113 113 (163443200)
  1089. I0802 16:51:32.017421 2858 net.cpp:139] Memory required for data: 4145990000
  1090. I0802 16:51:32.017431 2858 layer_factory.hpp:77] Creating layer relu2
  1091. I0802 16:51:32.017436 2858 net.cpp:86] Creating Layer relu2
  1092. I0802 16:51:32.017439 2858 net.cpp:408] relu2 <- conv2
  1093. I0802 16:51:32.017444 2858 net.cpp:369] relu2 -> conv2 (in-place)
  1094. I0802 16:51:32.017783 2858 net.cpp:124] Setting up relu2
  1095. I0802 16:51:32.017791 2858 net.cpp:131] Top shape: 50 256 113 113 (163443200)
  1096. I0802 16:51:32.017794 2858 net.cpp:139] Memory required for data: 4799762800
  1097. I0802 16:51:32.017798 2858 layer_factory.hpp:77] Creating layer norm2
  1098. I0802 16:51:32.017803 2858 net.cpp:86] Creating Layer norm2
  1099. I0802 16:51:32.017807 2858 net.cpp:408] norm2 <- conv2
  1100. I0802 16:51:32.017810 2858 net.cpp:382] norm2 -> norm2
  1101. I0802 16:51:32.017953 2858 net.cpp:124] Setting up norm2
  1102. I0802 16:51:32.017959 2858 net.cpp:131] Top shape: 50 256 113 113 (163443200)
  1103. I0802 16:51:32.017961 2858 net.cpp:139] Memory required for data: 5453535600
  1104. I0802 16:51:32.017964 2858 layer_factory.hpp:77] Creating layer pool2
  1105. I0802 16:51:32.017969 2858 net.cpp:86] Creating Layer pool2
  1106. I0802 16:51:32.017999 2858 net.cpp:408] pool2 <- norm2
  1107. I0802 16:51:32.018003 2858 net.cpp:382] pool2 -> pool2
  1108. I0802 16:51:32.018060 2858 net.cpp:124] Setting up pool2
  1109. I0802 16:51:32.018065 2858 net.cpp:131] Top shape: 50 256 56 56 (40140800)
  1110. I0802 16:51:32.018067 2858 net.cpp:139] Memory required for data: 5614098800
  1111. I0802 16:51:32.018081 2858 layer_factory.hpp:77] Creating layer conv3
  1112. I0802 16:51:32.018087 2858 net.cpp:86] Creating Layer conv3
  1113. I0802 16:51:32.018105 2858 net.cpp:408] conv3 <- pool2
  1114. I0802 16:51:32.018107 2858 net.cpp:382] conv3 -> conv3
  1115. I0802 16:51:32.026690 2858 net.cpp:124] Setting up conv3
  1116. I0802 16:51:32.026705 2858 net.cpp:131] Top shape: 50 384 56 56 (60211200)
  1117. I0802 16:51:32.026707 2858 net.cpp:139] Memory required for data: 5854943600
  1118. I0802 16:51:32.026727 2858 layer_factory.hpp:77] Creating layer relu3
  1119. I0802 16:51:32.026748 2858 net.cpp:86] Creating Layer relu3
  1120. I0802 16:51:32.026751 2858 net.cpp:408] relu3 <- conv3
  1121. I0802 16:51:32.026772 2858 net.cpp:369] relu3 -> conv3 (in-place)
  1122. I0802 16:51:32.026926 2858 net.cpp:124] Setting up relu3
  1123. I0802 16:51:32.026932 2858 net.cpp:131] Top shape: 50 384 56 56 (60211200)
  1124. I0802 16:51:32.026933 2858 net.cpp:139] Memory required for data: 6095788400
  1125. I0802 16:51:32.026937 2858 layer_factory.hpp:77] Creating layer conv4
  1126. I0802 16:51:32.026958 2858 net.cpp:86] Creating Layer conv4
  1127. I0802 16:51:32.026962 2858 net.cpp:408] conv4 <- conv3
  1128. I0802 16:51:32.026965 2858 net.cpp:382] conv4 -> conv4
  1129. I0802 16:51:32.034037 2858 net.cpp:124] Setting up conv4
  1130. I0802 16:51:32.034052 2858 net.cpp:131] Top shape: 50 384 56 56 (60211200)
  1131. I0802 16:51:32.034054 2858 net.cpp:139] Memory required for data: 6336633200
  1132. I0802 16:51:32.034060 2858 layer_factory.hpp:77] Creating layer relu4
  1133. I0802 16:51:32.034080 2858 net.cpp:86] Creating Layer relu4
  1134. I0802 16:51:32.034085 2858 net.cpp:408] relu4 <- conv4
  1135. I0802 16:51:32.034106 2858 net.cpp:369] relu4 -> conv4 (in-place)
  1136. I0802 16:51:32.034279 2858 net.cpp:124] Setting up relu4
  1137. I0802 16:51:32.034286 2858 net.cpp:131] Top shape: 50 384 56 56 (60211200)
  1138. I0802 16:51:32.034287 2858 net.cpp:139] Memory required for data: 6577478000
  1139. I0802 16:51:32.034289 2858 layer_factory.hpp:77] Creating layer conv5
  1140. I0802 16:51:32.034312 2858 net.cpp:86] Creating Layer conv5
  1141. I0802 16:51:32.034314 2858 net.cpp:408] conv5 <- conv4
  1142. I0802 16:51:32.034318 2858 net.cpp:382] conv5 -> conv5
  1143. I0802 16:51:32.040576 2858 net.cpp:124] Setting up conv5
  1144. I0802 16:51:32.040592 2858 net.cpp:131] Top shape: 50 256 56 56 (40140800)
  1145. I0802 16:51:32.040611 2858 net.cpp:139] Memory required for data: 6738041200
  1146. I0802 16:51:32.040618 2858 layer_factory.hpp:77] Creating layer relu5
  1147. I0802 16:51:32.040626 2858 net.cpp:86] Creating Layer relu5
  1148. I0802 16:51:32.040629 2858 net.cpp:408] relu5 <- conv5
  1149. I0802 16:51:32.040635 2858 net.cpp:369] relu5 -> conv5 (in-place)
  1150. I0802 16:51:32.040817 2858 net.cpp:124] Setting up relu5
  1151. I0802 16:51:32.040823 2858 net.cpp:131] Top shape: 50 256 56 56 (40140800)
  1152. I0802 16:51:32.040825 2858 net.cpp:139] Memory required for data: 6898604400
  1153. I0802 16:51:32.040828 2858 layer_factory.hpp:77] Creating layer pool5
  1154. I0802 16:51:32.040848 2858 net.cpp:86] Creating Layer pool5
  1155. I0802 16:51:32.040849 2858 net.cpp:408] pool5 <- conv5
  1156. I0802 16:51:32.040853 2858 net.cpp:382] pool5 -> pool5
  1157. I0802 16:51:32.040920 2858 net.cpp:124] Setting up pool5
  1158. I0802 16:51:32.040925 2858 net.cpp:131] Top shape: 50 256 28 28 (10035200)
  1159. I0802 16:51:32.040927 2858 net.cpp:139] Memory required for data: 6938745200
  1160. I0802 16:51:32.040930 2858 layer_factory.hpp:77] Creating layer fc6
  1161. I0802 16:51:32.040936 2858 net.cpp:86] Creating Layer fc6
  1162. I0802 16:51:32.040938 2858 net.cpp:408] fc6 <- pool5
  1163. I0802 16:51:32.040942 2858 net.cpp:382] fc6 -> fc6
  1164. I0802 16:52:04.265542 2858 net.cpp:124] Setting up fc6
  1165. I0802 16:52:07.463881 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1166. I0802 16:52:07.463922 2858 net.cpp:139] Memory required for data: 6939564400
  1167. I0802 16:52:08.317765 2858 layer_factory.hpp:77] Creating layer relu6
  1168. I0802 16:52:08.367188 2858 net.cpp:86] Creating Layer relu6
  1169. I0802 16:52:08.424886 2858 net.cpp:408] relu6 <- fc6
  1170. I0802 16:52:08.596489 2858 net.cpp:369] relu6 -> fc6 (in-place)
  1171. I0802 16:52:16.592660 2858 net.cpp:124] Setting up relu6
  1172. I0802 16:52:16.592710 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1173. I0802 16:52:16.592722 2858 net.cpp:139] Memory required for data: 6940383600
  1174. I0802 16:52:16.607672 2858 layer_factory.hpp:77] Creating layer drop6
  1175. I0802 16:52:16.630497 2858 net.cpp:86] Creating Layer drop6
  1176. I0802 16:52:16.630540 2858 net.cpp:408] drop6 <- fc6
  1177. I0802 16:52:16.630563 2858 net.cpp:369] drop6 -> fc6 (in-place)
  1178. I0802 16:52:16.676877 2858 net.cpp:124] Setting up drop6
  1179. I0802 16:52:16.676910 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1180. I0802 16:52:16.676918 2858 net.cpp:139] Memory required for data: 6941202800
  1181. I0802 16:52:16.676931 2858 layer_factory.hpp:77] Creating layer fc7
  1182. I0802 16:52:16.682255 2858 net.cpp:86] Creating Layer fc7
  1183. I0802 16:52:16.682273 2858 net.cpp:408] fc7 <- fc6
  1184. I0802 16:52:16.682291 2858 net.cpp:382] fc7 -> fc7
  1185. I0802 16:52:17.076421 2858 net.cpp:124] Setting up fc7
  1186. I0802 16:52:17.076830 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1187. I0802 16:52:17.076838 2858 net.cpp:139] Memory required for data: 6942022000
  1188. I0802 16:52:17.076848 2858 layer_factory.hpp:77] Creating layer relu7
  1189. I0802 16:52:17.076869 2858 net.cpp:86] Creating Layer relu7
  1190. I0802 16:52:17.076887 2858 net.cpp:408] relu7 <- fc7
  1191. I0802 16:52:17.076892 2858 net.cpp:369] relu7 -> fc7 (in-place)
  1192. I0802 16:52:17.127794 2858 net.cpp:124] Setting up relu7
  1193. I0802 16:52:17.127849 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1194. I0802 16:52:17.127859 2858 net.cpp:139] Memory required for data: 6942841200
  1195. I0802 16:52:17.127871 2858 layer_factory.hpp:77] Creating layer drop7
  1196. I0802 16:52:17.127898 2858 net.cpp:86] Creating Layer drop7
  1197. I0802 16:52:17.127913 2858 net.cpp:408] drop7 <- fc7
  1198. I0802 16:52:17.127929 2858 net.cpp:369] drop7 -> fc7 (in-place)
  1199. I0802 16:52:17.128054 2858 net.cpp:124] Setting up drop7
  1200. I0802 16:52:17.128072 2858 net.cpp:131] Top shape: 50 4096 (204800)
  1201. I0802 16:52:17.128082 2858 net.cpp:139] Memory required for data: 6943660400
  1202. I0802 16:52:17.128088 2858 layer_factory.hpp:77] Creating layer fc8
  1203. I0802 16:52:17.128105 2858 net.cpp:86] Creating Layer fc8
  1204. I0802 16:52:17.128116 2858 net.cpp:408] fc8 <- fc7
  1205. I0802 16:52:17.128137 2858 net.cpp:382] fc8 -> fc8
  1206. I0802 16:52:17.330720 2858 net.cpp:124] Setting up fc8
  1207. I0802 16:52:17.330772 2858 net.cpp:131] Top shape: 50 1000 (50000)
  1208. I0802 16:52:17.330781 2858 net.cpp:139] Memory required for data: 6943860400
  1209. I0802 16:52:17.330806 2858 layer_factory.hpp:77] Creating layer fc8_fc8_0_split
  1210. I0802 16:52:17.420990 2858 net.cpp:86] Creating Layer fc8_fc8_0_split
  1211. I0802 16:52:17.421036 2858 net.cpp:408] fc8_fc8_0_split <- fc8
  1212. I0802 16:52:17.421066 2858 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_0
  1213. I0802 16:52:17.421098 2858 net.cpp:382] fc8_fc8_0_split -> fc8_fc8_0_split_1
  1214. I0802 16:52:17.433820 2858 net.cpp:124] Setting up fc8_fc8_0_split
  1215. I0802 16:52:17.433867 2858 net.cpp:131] Top shape: 50 1000 (50000)
  1216. I0802 16:52:17.433878 2858 net.cpp:131] Top shape: 50 1000 (50000)
  1217. I0802 16:52:17.433887 2858 net.cpp:139] Memory required for data: 6944260400
  1218. I0802 16:52:17.433907 2858 layer_factory.hpp:77] Creating layer accuracy
  1219. I0802 16:52:17.490643 2858 net.cpp:86] Creating Layer accuracy
  1220. I0802 16:52:17.490690 2858 net.cpp:408] accuracy <- fc8_fc8_0_split_0
  1221. I0802 16:52:17.490710 2858 net.cpp:408] accuracy <- label_data_1_split_0
  1222. I0802 16:52:17.490727 2858 net.cpp:382] accuracy -> accuracy
  1223. I0802 16:52:17.525151 2858 net.cpp:124] Setting up accuracy
  1224. I0802 16:52:17.525197 2858 net.cpp:131] Top shape: (1)
  1225. I0802 16:52:17.525207 2858 net.cpp:139] Memory required for data: 6944260404
  1226. I0802 16:52:17.525218 2858 layer_factory.hpp:77] Creating layer loss
  1227. I0802 16:52:17.525286 2858 net.cpp:86] Creating Layer loss
  1228. I0802 16:52:17.525297 2858 net.cpp:408] loss <- fc8_fc8_0_split_1
  1229. I0802 16:52:17.525315 2858 net.cpp:408] loss <- label_data_1_split_1
  1230. I0802 16:52:17.629770 2858 net.cpp:382] loss -> loss
  1231. I0802 16:52:17.822856 2858 layer_factory.hpp:77] Creating layer loss
  1232. I0802 16:52:18.057018 2858 net.cpp:124] Setting up loss
  1233. I0802 16:52:18.057070 2858 net.cpp:131] Top shape: (1)
  1234. I0802 16:52:18.057077 2858 net.cpp:134] with loss weight 1
  1235. I0802 16:52:18.161309 2858 net.cpp:139] Memory required for data: 6944260408
  1236. I0802 16:52:18.161334 2858 net.cpp:200] loss needs backward computation.
  1237. I0802 16:52:18.161356 2858 net.cpp:202] accuracy does not need backward computation.
  1238. I0802 16:52:18.161370 2858 net.cpp:200] fc8_fc8_0_split needs backward computation.
  1239. I0802 16:52:18.161378 2858 net.cpp:200] fc8 needs backward computation.
  1240. I0802 16:52:18.161389 2858 net.cpp:200] drop7 needs backward computation.
  1241. I0802 16:52:18.161402 2858 net.cpp:200] relu7 needs backward computation.
  1242. I0802 16:52:18.161413 2858 net.cpp:200] fc7 needs backward computation.
  1243. I0802 16:52:18.161423 2858 net.cpp:200] drop6 needs backward computation.
  1244. I0802 16:52:18.161433 2858 net.cpp:200] relu6 needs backward computation.
  1245. I0802 16:52:18.161445 2858 net.cpp:200] fc6 needs backward computation.
  1246. I0802 16:52:18.161454 2858 net.cpp:200] pool5 needs backward computation.
  1247. I0802 16:52:18.161464 2858 net.cpp:200] relu5 needs backward computation.
  1248. I0802 16:52:18.161475 2858 net.cpp:200] conv5 needs backward computation.
  1249. I0802 16:52:18.172785 2858 net.cpp:200] relu4 needs backward computation.
  1250. I0802 16:52:18.172827 2858 net.cpp:200] conv4 needs backward computation.
  1251. I0802 16:52:18.214609 2858 net.cpp:200] relu3 needs backward computation.
  1252. I0802 16:52:18.219866 2858 net.cpp:200] conv3 needs backward computation.
  1253. I0802 16:52:18.219916 2858 net.cpp:200] pool2 needs backward computation.
  1254. I0802 16:52:18.219926 2858 net.cpp:200] norm2 needs backward computation.
  1255. I0802 16:52:18.219935 2858 net.cpp:200] relu2 needs backward computation.
  1256. I0802 16:52:18.219943 2858 net.cpp:200] conv2 needs backward computation.
  1257. I0802 16:52:18.219959 2858 net.cpp:200] pool1 needs backward computation.
  1258. I0802 16:52:18.229528 2858 net.cpp:200] norm1 needs backward computation.
  1259. I0802 16:52:18.229575 2858 net.cpp:200] relu1 needs backward computation.
  1260. I0802 16:52:18.229585 2858 net.cpp:200] scale1 needs backward computation.
  1261. I0802 16:52:18.229595 2858 net.cpp:200] bn1 needs backward computation.
  1262. I0802 16:52:18.229602 2858 net.cpp:200] conv1 needs backward computation.
  1263. I0802 16:52:18.229616 2858 net.cpp:202] label_data_1_split does not need backward computation.
  1264. I0802 16:52:18.229629 2858 net.cpp:202] data does not need backward computation.
  1265. I0802 16:52:18.229641 2858 net.cpp:244] This network produces output accuracy
  1266. I0802 16:52:18.229655 2858 net.cpp:244] This network produces output loss
  1267. I0802 16:52:18.229758 2858 net.cpp:257] Network initialization done.
  1268. I0802 16:52:18.840412 2858 solver.cpp:56] Solver scaffolding done.
  1269. I0802 16:52:18.998049 2858 caffe.cpp:248] Starting Optimization
  1270. I0802 16:52:18.998095 2858 solver.cpp:273] Solving AlexNet
  1271. I0802 16:52:18.998103 2858 solver.cpp:274] Learning Rate Policy: step
  1272. I0802 16:52:19.448653 2858 solver.cpp:331] Iteration 0, Testing net (#0)
  1273. F0802 16:52:22.157035 2858 syncedmem.cpp:71] Check failed: error == cudaSuccess (2 vs. 0) out of memory
  1274. *** Check failure stack trace: ***
  1275. @ 0x7f856777c5cd google::LogMessage::Fail()
  1276. @ 0x7f856777e433 google::LogMessage::SendToLog()
  1277. @ 0x7f856777c15b google::LogMessage::Flush()
  1278. @ 0x7f856777ee1e google::LogMessageFatal::~LogMessageFatal()
  1279. @ 0x7f8567db3ad8 caffe::SyncedMemory::mutable_gpu_data()
  1280. @ 0x7f8567f280f2 caffe::Blob<>::mutable_gpu_data()
  1281. @ 0x7f8567f5be48 caffe::CuDNNConvolutionLayer<>::Forward_gpu()
  1282. @ 0x7f8567dc28b1 caffe::Net<>::ForwardFromTo()
  1283. @ 0x7f8567dc29b7 caffe::Net<>::Forward()
  1284. @ 0x7f8567f1acb2 caffe::Solver<>::Test()
  1285. @ 0x7f8567f1b6ce caffe::Solver<>::TestAll()
  1286. @ 0x7f8567f1ec27 caffe::Solver<>::Step()
  1287. @ 0x7f8567f1eeea caffe::Solver<>::Solve()
  1288. @ 0x40aeb4 train()
  1289. @ 0x4075a8 main
  1290. @ 0x7f8565f13830 __libc_start_main
  1291. @ 0x407e79 _start
  1292. @ (nil) (unknown)
  1293. Aborted (core dumped)
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!