I0415 15:03:37.603461 27311 solver.cpp:42] Solver scaffolding done.
I0415 15:03:37.603549 27311 solver.cpp:247] Solving AlexNet
I0415 15:03:37.603559 27311 solver.cpp:248] Learning Rate Policy: step
I0415 15:03:37.749981 27311 solver.cpp:214] Iteration 0, loss = 5.45141
I0415 15:03:37.750030 27311 solver.cpp:229]     Train net output #0: loss = 5.45141 (* 1 = 5.45141 loss)
I0415 15:03:37.750048 27311 solver.cpp:489] Iteration 0, lr = 0.001
I0415 15:03:38.316994 27311 solver.cpp:214] Iteration 12, loss = 4.23865
I0415 15:03:38.317054 27311 solver.cpp:229]     Train net output #0: loss = 4.23865 (* 1 = 4.23865 loss)
I0415 15:03:38.317068 27311 solver.cpp:489] Iteration 12, lr = 0.001
I0415 15:03:38.920938 27311 solver.cpp:214] Iteration 24, loss = 2.49914
I0415 15:03:38.921000 27311 solver.cpp:229]     Train net output #0: loss = 2.49914 (* 1 = 2.49914 loss)
I0415 15:03:38.921016 27311 solver.cpp:489] Iteration 24, lr = 0.001
I0415 15:03:39.509793 27311 solver.cpp:214] Iteration 36, loss = 3.76504
I0415 15:03:39.509850 27311 solver.cpp:229]     Train net output #0: loss = 3.76504 (* 1 = 3.76504 loss)
I0415 15:03:39.509861 27311 solver.cpp:489] Iteration 36, lr = 0.001
I0415 15:03:40.080806 27311 solver.cpp:214] Iteration 48, loss = 3.74901
I0415 15:03:40.080862 27311 solver.cpp:229]     Train net output #0: loss = 3.74901 (* 1 = 3.74901 loss)
I0415 15:03:40.080878 27311 solver.cpp:489] Iteration 48, lr = 0.001
I0415 15:03:40.643797 27311 solver.cpp:214] Iteration 60, loss = 2.27091
I0415 15:03:40.643849 27311 solver.cpp:229]     Train net output #0: loss = 2.27091 (* 1 = 2.27091 loss)
I0415 15:03:40.643860 27311 solver.cpp:489] Iteration 60, lr = 0.001
I0415 15:03:41.217475 27311 solver.cpp:214] Iteration 72, loss = 2.67078
I0415 15:03:41.217541 27311 solver.cpp:229]     Train net output #0: loss = 2.67078 (* 1 = 2.67078 loss)
I0415 15:03:41.217561 27311 solver.cpp:489] Iteration 72, lr = 0.001
I0415 15:03:41.793390 27311 solver.cpp:214] Iteration 84, loss = 1.77313
I0415 15:03:41.793452 27311 solver.cpp:229]     Train net output #0: loss = 1.77313 (* 1 = 1.77313 loss)
I0415 15:03:41.793468 27311 solver.cpp:489] Iteration 84, lr = 0.001
I0415 15:03:42.362951 27311 solver.cpp:214] Iteration 96, loss = 3.49406
I0415 15:03:42.363004 27311 solver.cpp:229]     Train net output #0: loss = 3.49406 (* 1 = 3.49406 loss)
I0415 15:03:42.363025 27311 solver.cpp:489] Iteration 96, lr = 0.001
I0415 15:03:42.946568 27311 solver.cpp:214] Iteration 108, loss = 2.81601
I0415 15:03:42.946633 27311 solver.cpp:229]     Train net output #0: loss = 2.81601 (* 1 = 2.81601 loss)
I0415 15:03:42.946651 27311 solver.cpp:489] Iteration 108, lr = 0.001
I0415 15:03:43.524155 27311 solver.cpp:214] Iteration 120, loss = 2.85056
I0415 15:03:43.524247 27311 solver.cpp:229]     Train net output #0: loss = 2.85056 (* 1 = 2.85056 loss)
I0415 15:03:43.524265 27311 solver.cpp:489] Iteration 120, lr = 0.001
I0415 15:03:44.100580 27311 solver.cpp:214] Iteration 132, loss = 3.58945
I0415 15:03:44.100646 27311 solver.cpp:229]     Train net output #0: loss = 3.58945 (* 1 = 3.58945 loss)
I0415 15:03:44.100661 27311 solver.cpp:489] Iteration 132, lr = 0.001
F0415 15:03:44.536542 27311 math_functions.cpp:91] Check failed: error == cudaSuccess (4 vs. 0)  unspecified launch failure
*** Check failure stack trace: ***
    @     0x7f01dbd9ddaa  (unknown)
    @     0x7f01dbd9dce4  (unknown)
    @     0x7f01dbd9d6e6  (unknown)
    @     0x7f01dbda0687  (unknown)
    @     0x7f01dc1bb3f5  caffe::caffe_copy<>()
    @     0x7f01dc230232  caffe::BasePrefetchingDataLayer<>::Forward_gpu()
    @     0x7f01dc1d9d6f  caffe::Net<>::ForwardFromTo()
    @     0x7f01dc1da197  caffe::Net<>::ForwardPrefilled()
    @     0x7f01dc20cbe5  caffe::Solver<>::Step()
    @     0x7f01dc20d52f  caffe::Solver<>::Solve()
    @           0x406428  train()
    @           0x404961  main
    @     0x7f01db2afec5  (unknown)
    @           0x404f0d  (unknown)
    @              (nil)  (unknown)
Aborted
wangxiao@gtx-980:~/Downloads/lstm_caffe_master$

---------------------------------------------------------------------------------------------------------------------------------------

怎么破 ？？？求解答。。。。

well, the only word I want to say is : where amazing happens ???

I restart my pc and run the code again and it worked ......

caffe 训练时，出现错误：Check failed: error == cudaSuccess (4 vs. 0) unspecified launch failure的更多相关文章

  1. caffe运行错误： im2col.cu:61] Check failed: error == cudaSuccess (8 vs. 0) invalid device function

    错误: im2col.cu:61] Check failed: error == cudaSuccess (8 vs. 0)  invalid device function 原因:由于Makefil ...

  2. 配置SSD-caffe测试时出现“Check failed: error == cudaSuccess (10 vs. 0) invalid device ordinal”解决方案

    这是由于GPU数量不匹配造成的,如果训练自己的数据,那么我们只需要将solver.prototxt文件中的device_id项改为自己的GPU块数,一块就是0,两块就是1,以此类推. 但是SSD配置时 ...

  3. 【CUDA开发】 Check failed: error == cudaSuccess (8 vs. 0) invalid device function

    最近在复现R-CNN一系列的实验时,配置代码环境真是花费了不少时间.由于对MATLAB不熟悉,实验采用的都是github上rbg大神的Python版本.在配置Faster R-CNN时,编译没有问题, ...

  4. Caffe 分类问题 Check failed: error == cudaSuccess (2 vs. 0) out of memory

    如果图片过大,需要适当缩小batch_size的值,否则使用GPU时可能超出其缓存大小而报错

  5. Check failed: status == CUBLAS_STATUS_SUCCESS (11 vs. 0) CUBLAS_STATUS_MAPPING_ERROR

    I0930 21:23:15.115576 30918 solver.cpp:281] Learning Rate Policy: multistepF0930 21:23:17.263314 310 ...

  6. check failed status == cudnn_status_success (4 vs. 0) cudnn_status_internal_error

    Check failed: error == cudaSuccess (30 vs. 0) unknown error  这个有可能是显存不足造成的,或者网络参数不对造成的 check failed ...

  7. 目标检测faster rcnn error == cudaSuccess (2 vs. 0) out of memory

    想尝试 更深更强的网络,或者自己写了一个费显存的层,发现1080 ti的11G显存不够用了,老师报显存不够怎么办? Check failed: error == cudaSuccess (2 vs. ...

  8. 发布到远程存储库时遇到错误: Git failed with a fatal error.

    正在推送 master发布到远程存储库时遇到错误: Git failed with a fatal error.Authentication failed for 'http://1212121xxx ...

  9. windows7下解决caffe check failed registry.count(type) == 1(0 vs. 1) unknown layer type问题

    在Windows7下调用vs2013生成的Caffe静态库时经常会提示Check failed: registry.count(type) == 1 (0 vs. 1) Unknown layer t ...

随机推荐

  1. ios 检测应用程序升级问题

    app 上其实已经有自动检测我们版本的功能.  其实我也觉得对于一个程序员来说检测功能让,系统来维护更合适和合理.开发者只要告诉苹果即可. 然而今天老大非要实现自己版本更新的问题,因此也查找了相关的资 ...

  2. C++11的new concepts (move semantic)

    MoveConstructible 和MoveAssignable MoveConstructible Specifies that an instance of the type can be mo ...

  3. C#线程状态简析

    在C# 语言世界中,当我们创建一个新的子线程,该线程状态为unstarted, 子线程开始之后,子线程状态为Running,IsAlive 为true: 线程响应 Thread.Start 并开始运行 ...

  4. nslayoutConstraint

    1.vfl的正确编写格式 NSDictionary *dict1 = NSDictionaryOfVariableBindings(_boxV,_headerL,_imageV,_backBtn,_d ...

  5. 将salt取到的数据处理

    #!/usr/bin/env python #coding:utf-8 import json with open('minfo') as f,open('minfoMiddle','w') as f ...

  6. magento 全页缓存

    全页缓存能极大的提高网页的响应速率,但magento社区的商用全页缓存工具都很贵,但是有一款免费的叫ezzoom,虽然免费但是功能强大,能够对产品.分类.特定模块进行缓存,还支持页面某些部分的动态加载 ...

  7. 传智springMVC笔记

    springmvc 第一天 springmvc的基础知识 课程安排: 第一天:springmvc的基础知识 什么是springmvc? springmvc框架原理(掌握) 前端控制器.处理器映射器.处 ...

  8. HPCC 登录总结

    最近开始做NGS的分析,数据明显更大,在自己的机子上面做有些不现实了,需要登录高性能计算机. 1. 目录结构: home directory: /auto/rcf-40/USERNAME -- onl ...

  9. 关于FireFox类VIM插件。VimPerator

    说实话现在一直非常喜欢VIM的那种操控方式.当然我不是一个程序员,只是觉得处理文本的时候,那种效率很快.而且不用来回再鼠标跟键盘之间的切换也让我感觉很舒服. 以前一直在想如果开网页的时候也能那样得话不 ...

  10. Java 性能优化实战记录(3)--JVM OOM的分析和原因追查

    前言: C/C++的程序员渴望Java的自由, Java程序员期许C/C++的约束. 其实那里都是围城, 外面的人想进来, 里面的人想出去. 背景: 作为Java程序员, 除了享受垃圾回收机制带来的便 ...