- add Loss, Optimizer
- change Trainer & Tester initialization interface: two styles of definition provided
- handle Optimizer construction and loss function definition in a hard manner
- add argparse in task-specific scripts. (seq_labeling.py & text_classify.py)
- seq_labeling.py & text_classify.py work

- fastNLP interface for sequence labeling works
- fastNLP interface for text classification works

- [tester][trainer] add cuda support
- [preprocess] fix label2index for padding label seq
- update README.md
- [test] add test_tester.py
- rename "action" to "core"

- [trainer]rename "batchify" to "make_batch" in trainer
- [trainer]pack (batch_x_pad, seq_len) into batch_x in make_batch for seq labeling, because seq length before pad is needed to make masks
- [trainer]unpack it in data_forward
- [model]shorten model definition
- [inference]build inference class. test_POS_pipeline.py is OK to infer
- [preprocessor]handle pickles in a nicer manner
- [FastNLP] add fastNLP.py as high-level API, not finished yet

optimize evaluation output in Trainer;
keep POS pipeline (loader + trainer + tester + saver) OK;
add codes borrowed from FudanParser.