config file

[zseder]

argparse is okay right now, but there are a lot of options and there will be a lot more later, so we need a config infrastructure

• arguments should be there
• new configs about possible datasets (eng+ner, hun+pos, etc)

[zseder]

there is pylearn2's yaml config, we should consider that first, maybe it will be complicated first, but more reusable by others. Maybe this will bring more changes in the code, but for later good