* add IMPALA algorithm and some common utils

* update README.md

* refactor files structure of impala algorithm; seperate numpy utils from utils

* add hyper parameter scheduler module; add entropy and lr scheduler in impala

* clip reward in atari wrapper instead of learner side; fix codestyle

* add benchmark result of impala; refine code of impala example; add obs_format in atari_wrappers

* Update README.md

* add a3c algorithm, A2C example and rl_utils

* require training in single gpu/cpu

* only check cpu/gpu num in learner

* refine Readme

* update impala benchmark picture; update Readme

* add benchmark result of A2C

* move get_params/set_params in agent_base

* fix shell script cannot run in ubuntu

* refine comment and document

* Update README.md

* Update README.md