Commit Graph

11292 Commits (ff5fb9d93cc31679cca9eed362a48746e3ca0d0e)
 

Author SHA1 Message Date
baihuawei 9f75418001 fix GPU BroadCast
4 years ago
r1chardf1d0 079a347c04 refactor: composite compute opt framework
4 years ago
mindspore-ci-bot 8e441fce25 !7079 VGG16 remove device_id setting when use GPU
4 years ago
mindspore-ci-bot b6542e3277 !7080 add batch_to_space_nd op for opencl
4 years ago
mindspore-ci-bot 21c5607fca !6971 cudnn inplace optimizer
4 years ago
mindspore-ci-bot ab9cec0531 !7083 Support Categorical distribution
4 years ago
zhengjun10 7e56655bc2 weight format transform modify
4 years ago
mindspore-ci-bot d7b7ba3797 !7054 move dump id to end
4 years ago
mindspore-ci-bot 4cfbf69251 !7084 [MSLITE] Add retinaface model
4 years ago
mindspore-ci-bot 7148109c06 !7088 modify lenet to mixed precision
4 years ago
mindspore-ci-bot b1dd00f3a9 !7092 Support Logistic Distribution
4 years ago
mindspore-ci-bot ec3b5ee195 !7093 Add description of seed that TransformedDistribution uses
4 years ago
mindspore-ci-bot 12094c97ca !7048 Fix some type casting errors in probabilistic programming code
4 years ago
mindspore-ci-bot ccb2e8851c !6782 enable parallel optimizer in auto parallel
4 years ago
minara 5859204046 add reume for training
4 years ago
mindspore-ci-bot d30dece8a9 !6515 Resolve deadlock issue when terminate debugger from UI
4 years ago
mindspore-ci-bot a261def3c9 !7094 C++ API: Move vision ops to new vision files; create transforms namespace
4 years ago
Harshvardhan Gupta e77e816f1d fix overflow dump path
4 years ago
Cathy Wong 2aa26d245c C++ API: Move vision ops to new vision files; create transforms namespace
4 years ago
peixu_ren 3ed22fa2cf Add note regarding to the transformed distribution seed arg
4 years ago
Xun Deng 05a0dac125 add logistic distribution
4 years ago
mindspore-ci-bot 9c79b9d712 !6907 Load input tensors in debugger before suspending execution
4 years ago
root a0874000d8 Fix FusedBatchNormEx and SparseSoftmax for GPU
4 years ago
Ziyan ddc0113058 enable parallel optimizer in auto parallel
4 years ago
guoqi 6606c1e431 modify lenet to mixed precision
4 years ago
Xun Deng 877b561e77 added categorical distribution
4 years ago
guohongzilong 71402b6013 add retinaface model
4 years ago
peixu_ren 1637381fae Fix some type casting errors
4 years ago
Harshvardhan Gupta 7c5e0541ba load inputs before suspending execution in dbg
4 years ago
mindspore-ci-bot c1b9efe8e6 !7050 dataset API docstring: Update/add text examples
4 years ago
mindspore-ci-bot c8d8bef9e6 !7077 code format for nn.layer
4 years ago
mindspore-ci-bot 03f0e64af9 !7049 Fixes to TensorRow conversion and Tensor
4 years ago
Cathy Wong e0aaf847aa dataset API docstring: Update/add text examples
4 years ago
wandongdong 2ba99f347f add batch_to_space op for opencl
4 years ago
mindspore-ci-bot 3ce21658d0 !7069 replace BasicLSTM with DynamicRNN, add export.py
4 years ago
mindspore-ci-bot c5894f71e8 !7067 [MS][LITE][Develop] GPU add new ops named pad
4 years ago
mindspore-ci-bot d113e7a694 !7071 fix securec download links due to mistakes made by openeuler community
4 years ago
caojian05 bfe85189d2 remove device_id setting when use GPU
4 years ago
chenzomi d5ae6fdd84 code format for nn.layer
4 years ago
mindspore-ci-bot 13bf28e538 !7057 fix mssim precision when dtype is uint32.
4 years ago
wang_shaocong 1a6f4340ba add ci test for ocr_mobilenet model
4 years ago
mindspore-ci-bot 6f77ec45f1 !6905 [MD] Enable cache for more leaf datasets
4 years ago
wilfChen b420b6cda7 cudnn inplace optimizer
4 years ago
mindspore-ci-bot b9df01b60e !7022 [AutoParallel]Fix find loss and root reshape bug
4 years ago
gengdongjie 3aad1b5dfe warpctc: replace BasicLSTM with DYNAMICRNN
4 years ago
yanghaoran 82391b7829 fix securec download links due to mistakes made by openeuler community
4 years ago
mindspore-ci-bot 7f94d596c0 !7012 [MD] sampler log update
4 years ago
mindspore-ci-bot d01f721d87 !7013 GraphKernel supports multi-output kernels
4 years ago
mindspore-ci-bot f0e9f6137b !7066 [MS][LITE][DEV]fix fp16 op bug
4 years ago
Pengyongrong 78408c1a99 new ops named pad
4 years ago