fastNLP

25 MB

Tree: 9e29fb2262

Author	SHA1	Message	Date
FengZiYjun	77b3a0c67d	fastNLP high-level interface: - fastNLP interface for sequence labeling works - fastNLP interface for text classification works	7 years ago
FengZiYjun	242e576a30	changes to trainer, tester, preprocessor, etc. - [tester][trainer] add cuda support - [preprocess] fix label2index for padding label seq - update README.md - [test] add test_tester.py - rename "action" to "core"	7 years ago
FengZiYjun	c83008add9	fastnlp.py works, see test/test_fastNLP.py for high-level API	7 years ago
FengZiYjun	fe17f611b6	changes to preprocessor, trainer, inference & seq modeling - [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	7 years ago
FengZiYjun	3d234bf5b2	add model selection (best dev) in Trainer: save the best model during validation; add Inference.	7 years ago
FengZiYjun	301bbdcd1e	add accuracy in POS Tester; optimize evaluation output in Trainer; keep POS pipeline (loader + trainer + tester + saver) OK; add codes borrowed from FudanParser.	7 years ago
FengZiYjun	621b79ee19	update configLoader to load hyper-parameters from file	7 years ago