Wei Yang bearpaw

NVIDIA Robotics Research Lab

769 followers · 46 following

NVIDIA
Seattle, WA, USA
01:04 (UTC -08:00)
www.wyang.me
@wei_yang_cv

View GitHub Profile

Recently created

Least recently created

Recently updated

Least recently updated

bearpaw / demo.txt

Created February 17, 2016 07:36

Compute Receptive Field Size

	conv1 5 1
	pool1 2 2
	conv2 5 1
	pool2 2 2
	conv3 5 1
	conv4 9 1
	conv5 1 1
	conv6 1 1
	conv7 9 1
	conv8 13 1

bearpaw / LMDB_val_patches.txt

Last active November 10, 2025 00:49

Read LMDB in python

	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000001.jpg 1
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000002.jpg 1823
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000003.jpg 2903
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000004.jpg 4470
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000005.jpg 4698
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000006.jpg 4797
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000007.jpg 4917
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000008.jpg 5072
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000009.jpg 5177
	./cache/flic_win/LMDB/LMDB_val_patches/casino-royale-00039871_00000001_000010.jpg 5271

bearpaw / demo.m

Last active August 29, 2015 14:27

convert-matcaffe3-model-to-matcaffe

	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);

bearpaw / FLIC-dcnn-training-xjc.log

Last active August 29, 2015 14:26

	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

bearpaw / dt.cpp

Created July 28, 2015 06:54

distance transform


	/*-------------------------------------------------------------------------------------
	* 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;

bearpaw / clarifai.prototxt

Created July 15, 2015 03:32

	name: "Clafifai"
	# --------------------- Data Layer --------------------------
	# input dimension is 36x36x3
	layers {
	name: "data"
	type: DATA
	top: "data"
	top: "label"
	data_param {
	source: "cache/lsp/LMDB_train"

bearpaw / Basic.md

Last active August 29, 2015 14:18

Ubuntu Basic Scripts

创建用户
```
sudo adduser username
```
将用户添加到管理员用户组
```
sudo adduser username sudo
```

bearpaw / Caffe + Ubuntu 12.04 64bit + CUDA 6.5 配置说明.md

Last active March 12, 2020 01:24

Caffe + Ubuntu 12.04 / 14.04 64bit + CUDA 6.5 / 7.0 配置说明

Caffe + Ubuntu 12.04 64bit + CUDA 6.5 配置说明

本步骤能实现用Intel核芯显卡来进行显示，用NVIDIA GPU进行计算。

1. 安装开发所需的依赖包

安装开发所需要的一些基本包

sudo apt-get install build-essential
sudo apt-get install vim cmake git
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev

NewerOlder