* Fix eval.py in ocr ctc demo. (#1970)

* Refine readme in cycle gan. (#1972)

* 1. Refine readme
2. Fix some issues

* Refine README.md.

* Fix atention model. (#2059)

* 1. Enable GC strategy in run.sh (#2063)

2. Fix loading checkpoint in filter pruning strategy by adding init_model option into config file.

* Add demo for combined strategies.

* Fix quan.

* Restore args in run.sh