mindspore2022

584 MB

Tree: c94dea6a51

Author	SHA1	Message	Date
Ziyan	4cbcd8e907	enable use float type learning rate in lars optimizer	6 years ago
zhaozhenlong	9862dea3cf	adapt relu6grad and graphengine modified	6 years ago
shibeiji	468e257a14	add log for kernel runtime in order to trace performance	6 years ago
jonyguo	34e42bd6f9	1. add more log info for dataset & mindrecord, 2. add two new testcase for MindDataset	6 years ago
seatea	6c03542eec	Fix dtype bug for loss_scale and weight_decay. 1.Change dtype of scale to dtype of grad in loss_scale.py; 2.Change dtype of weight_decay to dtype of weight in optimizer.py.	6 years ago
seatea	7b7a6a45a0	Check if the shape of the input of NMSWithMask is (N, 5).	6 years ago
jjfeing	86f5c69995	change parallel complie num: 32->16	6 years ago
GinFung	468dbc3557	Add matmul biasadd fusion pass	6 years ago
zhunaipan	930a1fb0a8	initial version Signed-off-by: leonwanghui <leon.wanghui@huawei.com>	6 years ago