根据ENet说明https://github.com/TimoSaemann/ENet/tree/master/Tutorial ,进行训练遇到以下问题
1 No module named spatial_dropout
ENet通过python接口定义了新层spatial_dropout,根据说明直接在终端启动训练,出现“No module named spatial_dropout”,原因是caffe,python层所在文件夹路径,没有添加到系统PYTHONPATH,一个解决方法是:用Python接口来启动训练:
import sys
caffe_root = 'home/f/caffe/'
sys.path.insert(0,caffe_root+'python')
import caffe
caffe.set_device(0)
caffe.set_mode_gpu()
solver = caffe.SGDSolver('/home/xxx/data/solver.prototxt')
solver.solve()
2 drop1_0_3 新建时, img_height == height 验证失败
data层的new_height new_weight不能删掉,根据输入图像设定大小即可。
layer {
name: "data"
type: "DenseImageData"
top: "data"
top: "label"
dense_image_data_param {
source: "ENet/dataset/train_fine_2columns.txt"
batch_size: 4
shuffle: true
new_height: 512
new_width: 1024
label_divide_factor: 8
}
}