http://blog.csdn.net/lucktroy/article/details/9466369
Pylearn2 是建立在Theano之上的一个机器学习库,可以参考:http://deeplearning.net/software/pylearn2/
早前我已经用《Install Theano in Linux without root authority (sodu)》,所以现在也只需要仿照着安装即可。
1. Download:
- $ git clone git://github.com/lisa-lab/pylearn2.git
$ git clone git://github.com/lisa-lab/pylearn2.git
2. 配置环境变量,我一般用.bash_profile, 也可以放在.bashrc里
- export PYLEARN2_DATA_PATH=/path/to/data
export PYLEARN2_DATA_PATH=/path/to/data
3. 安装,进入pylearn2
- $ python setup.py build
$ python setup.py build
- $ python setup.py install --prefix=$PYTHON_HOME
$ python setup.py install --prefix=$PYTHON_HOME
4. 测试,可以看http://deeplearning.net/software/pylearn2/tutorial/index.html#tutorial,首先下载数据集cifar-10: http://www.cs.toronto.edu/~kriz/cifar.html
- $ wget http://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz
$ wget http://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz
- $ tar -zxvf cifar-10*
$ tar -zxvf cifar-10*
4.1. 生成数据子集,进入$PYLEARN2_HOME/
pylearn2/scripts/tutorials/grbm_smd/
- $ python make_dataset.py
$ python make_dataset.py
4.2 训练模型
- $ python ../../train.py cifar_grbm_smd.yaml
$ python ../../train.py cifar_grbm_smd.yaml
4.3 视察模型
- $ python ../../show_weights.py cifar_grbm_smd.pkl
$ python ../../show_weights.py cifar_grbm_smd.pkl