* update model save load

* first add

* update model save and load

* update train.py

* update LaneNet model saving and loading

* adapt slim to paddle-1.8

* update distillation save and load

* update nas model save and load

* update model load op

* update utils.py

* update load_model_utils.py

* update model saving and loading

* update copyright

* update palette.py

* update solver.py and model_builder.py

* update solver.py

* update infer.py

* update model_builder.py for fitting different aug_method

* update model_builder.py

* update model_builder.py

* update export process

* save best model

* update train.py

* update train.py

* update train.py

* eval for last epoch model

* check model path

* check TEST.TEST_MODEL

* reconsitution humanseg

* first refactor humanseg

* update humanseg.py and add main.py

* add fast scnn

* fix readme

* update model name, add inference time

* fix typos

* fp16_err

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train

* add fp16 and multiple gpus train