-
创建用户
sudo adduser username
-
将用户添加到管理员用户组
sudo adduser username sudo
创建用户
sudo adduser username
将用户添加到管理员用户组
sudo adduser username sudo
name: "Clafifai" | |
# --------------------- Data Layer -------------------------- | |
# input dimension is 36x36x3 | |
layers { | |
name: "data" | |
type: DATA | |
top: "data" | |
top: "label" | |
data_param { | |
source: "cache/lsp/LMDB_train" |
/*------------------------------------------------------------------------------------- | |
* dt 1-d | |
* -----------------------------------------------------------------------------------*/ | |
template <typename Dtype> | |
void MessagePassingLayer<Dtype>::dt1d(const Dtype *src, Dtype *dst, int *ptr, int step, int len, | |
Dtype a_c, Dtype b_c, Dtype a_p, Dtype b_p, Dtype dshift_c, Dtype dshift_p, int dlen) { | |
int *v = new int[len]; | |
float *z = new float[len+1]; | |
int k = 0; |
Mean Pixel Chanel #1: 86.81 | |
Mean Pixel Chanel #2: 76.52 | |
Mean Pixel Chanel #3: 72.62 | |
Training model using gpu id: 0 | |
I0315 04:58:38.202831 26931 caffe.cpp:99] Use GPU with device ID 0 | |
I0315 04:58:38.342545 26931 caffe.cpp:107] Starting Optimization | |
I0315 04:58:38.342646 26931 solver.cpp:32] Initializing solver from parameters: | |
test_iter: 100 | |
test_interval: 5000 | |
base_lr: 0.005 |
deploy_file = '/home/wyang/code/pose/chen-nips14-pose/external/my_models/lsp/lsp_deploy_conv_grabcut.prototxt'; | |
weights_file = '/home/wyang/code/pose/chen-nips14-pose/cache/lsp/fully_conv_net_grabcut.caffemodel'; | |
save_model_file = 'fully_conv_net_grabcut_matcaffe1.caffemodel'; | |
% try | |
% % ------- Set params for old caffe model | |
% load('trans_params.mat'); | |
% addpath('/home/wyang/code/caffe-xianjiec/matlab/caffe'); | |
% set_weights(deploy_file, weights_file, trans_params, save_model_file); |
; Global settings | |
[redshift] | |
temp-day=6500K | |
temp-night=5000 | |
transition=1 | |
;gamma=0.8:0.7:0.8 | |
gamma=1.000:1.000:1.000 | |
location-provider=geoclue | |
;location-provider=manual | |
adjustment-method=vidmode |
acc epoch loss LR | |
0.1052 1 0.019193 0.00025 | |
0.3969 2 0.003348 0.00025 | |
0.5235 3 0.002922 0.00025 | |
0.5888 4 0.002697 0.00025 | |
0.6230 5 0.002564 0.00025 | |
0.6526 6 0.002433 0.00025 | |
0.6769 7 0.002321 0.00025 | |
0.6956 8 0.002223 0.00025 | |
0.7057 9 0.002167 0.00025 |
#!/usr/bin/env bash | |
cd ~ | |
wget https://repo.continuum.io/archive/Anaconda2-4.3.0-Linux-x86_64.sh | |
bash Anaconda2-4.3.0-Linux-x86_64.sh -b | |
echo 'PATH="/home/ubuntu/anaconda2/bin:$PATH"' >> .bashrc | |
. .bashrc | |
jupyter notebook --generate-config |
If you want to use an FTP client to transfer data between local machine and AWS server, you may be instrested in this tutorial.