|
GNetfc网络运行出现"段错误"
发表于 2019-6-5 18:19:22
浏览:2161
|
回复:1
打印
只看该作者
[复制链接]
楼主
- ==> Preparing data..
- Training and validation data exist already
- ==> Building model from scratch..
- Epoch: 0
- [================================================================>] Step: 1s585ms | To 114/114 ms | Loss: 0.643 | Acc: 66.484% (605/910)
- [================================================================>] Step: 18ms | T 115/115 ms | Loss: 0.561 | Acc: 77.391% (178/230)
- Saving..
- * Saved checkpoint to ./checkpoint/gti_2801_gnetfc.t7
- Epoch: 1
- [================================================================>] Step: 117ms | T 114/114 ms | Loss: 0.543 | Acc: 80.220% (730/910))
- [================================================================>] Step: 19ms | T 115/115 ms | Loss: 0.586 | Acc: 73.478% (169/230)
- ==> Slicing network..
- ==> Absorbing batch normalization..
- ==> Merging gain..
- ==> Dumping CNN parameters into .txt files..
- GTI Model gnetfc: output channel of layer 16 extended from 2 to 256
- ==> Start to generate the chip file..
- - 1559729304260 -140495802324736- 2-:GTI SDK 3.5.0 built on: Sep 7 2018
- - 1559729304263 -140495802324736- 0-:Could not find any available device
- ==> Start to perform batch inference..
- 段错误 (核心已转储)
复制代码
GNet1和GNet18能够正常训练并且推断,GNetfc训练正常,推断出现上述问题,最后定位到调用gClassify出现的问题,但是该函数在封装的so文件,请问这个可能什么情况。是我配置有问题吗?训练情况下需要插入神经棒吗?另外GTI_CHIP_MODE参数是指什么,GNetfc网络的时候,GTI_CHIP_MODE参数为2,其他情况下指为0,我并没有在文档中看到相关的介绍. |
|