18 Commits (1ea828c30eb9bd35becb3efe5cf55940144bb7be)

Author	SHA1	Message	Date
tianhao zhang	1ea828c30e	fix attention val bug	3 years ago
Hui Zhang	9277fcb8a8	fix attn can not train	3 years ago
Hui Zhang	925abcca23	format	3 years ago
Hui Zhang	feb27e2a84	fuse linear kv	3 years ago
Hui Zhang	46088c0a16	elimiate attn transpose	3 years ago
Hui Zhang	f9e3eaa024	transpose in matmul	3 years ago
tianhao zhang	6745e9dd6b	fix dp init	3 years ago
tianhao zhang	ed80b0e2c3	fix multigpu training test=asr	3 years ago
tianhao zhang	733ec7f2bc	fix conformer multi-gpu training test=asr	3 years ago
huangyuxin	75997d8277	merge	3 years ago
Hui Zhang	e81849277e	att cache for streaming asr	3 years ago
Hui Zhang	fb40602d94	refactor attention cache	3 years ago
huangyuxin	d53e1163a6	update the code, test=asr	4 years ago
huangyuxin	ab16d8ce3c	change default initializer to kaiming_uniform, test=asr	4 years ago
Hui Zhang	3bd87bc379	add wenet lincense	4 years ago
Hui Zhang	38cf56295a	fix reference format	4 years ago
Hui Zhang	c463a00f81	add reference code license	4 years ago
Hui Zhang	e2bcaee4f1	merge deepspeech, parakeet and text_processing into paddlespeech	4 years ago