Oneflow-Inc/oneflow
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
C++Apache-2.0
Issues
- 0
Inception_v3多卡错误
#10531 opened by LuckySJTU - 2
使用静态图推理时,当输入数量改变,静态图必须要重编吗,请问如何可以不用重编呢
#10524 opened by songh11 - 0
Aborted (core dumped) in `flow.nn.AdaptiveMaxPool1d/AdaptiveMaxPool2d/AdaptiveMaxPool3d`
#10523 opened by x0w3n - 0
Aborted (core dumped) in `flow.nn.AdaptiveAvgPool1d/AdaptiveAvgPool2d/AdaptiveAvgPool3d`
#10522 opened by x0w3n - 0
- 0
- 0
Aborted (core dumped) in `flow.nn.ConvTranspose1d/ConvTranspose2d/ConvTranspose3d`
#10519 opened by x0w3n - 0
Aborted (core dumped) in `oneflow.rand/zeros/ones`
#10518 opened by x0w3n - 0
[Feature Request]: svd operator
#10516 opened by lzhexp - 0
Aborted (core dumped) in oneflow.nn.functional.avg_pool3d
#10512 opened by x0w3n - 0
Aborted (core dumped) in oneflow.nn.functional.avg_pool2d
#10511 opened by x0w3n - 0
Aborted (core dumped) in flow.QatConv3d
#10510 opened by x0w3n - 0
Aborted (core dumped) in flow.QatConv2d
#10509 opened by x0w3n - 0
Aborted (core dumped) in flow.QatConv1d
#10508 opened by x0w3n - 0
Aborted (core dumped) in flow.randn
#10507 opened by x0w3n - 0
Aborted (core dumped) in oneflow.nn.functional.avg_pool1d
#10506 opened by x0w3n - 0
Aborted (core dumped) in oneflow.nn.MaxPool3d
#10505 opened by x0w3n - 0
Aborted (core dumped) in oneflow.nn.MaxPool1d
#10504 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.MaxPool2d
#10503 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.functional.conv_transpose1d
#10502 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.functional.conv_transpose2d
#10500 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.functional.conv_transpose3d
#10501 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.ReflectionPad2d
#10490 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.AdaptiveMaxPool2d
#10491 opened by x0w3n - 0
Aborted (core dumped) in flow.eye
#10492 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.ReflectionPad1d
#10497 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.AdaptiveMaxPool1d
#10498 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.AdaptiveMaxPool3d
#10499 opened by x0w3n - 0
creating GroupNorm with device and dtype throws Exceptions
#10493 opened by fpzh2011 - 0
Aborted (core dumped) in flow.nn.functional.conv3d
#10487 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.functional.conv2d
#10488 opened by x0w3n - 0
Aborted (core dumped) in flow.nn.functional.conv1d
#10489 opened by x0w3n - 0
test
#10486 opened by DriverOwen - 3
oneflow的tensor如何获取其指针
#10483 opened by bjtuwjx - 0
- 0
oneflow.matmul doesn't work for this situation
#10481 opened by Redmept1on - 0
- 0
core dumped occurs when an empty array is processed with oneflow.tensordot and the dims argument is 2
#10479 opened by Redmept1on - 0
oneflow.cast perform differently between cpu and cuda
#10478 opened by Redmept1on - 2
- 0
oneflow.gt/ge/greater/greater_equal perform differently between cpu and cuda.
#10477 opened by Redmept1on - 1
oneflow.max/min perform differently between cpu and cuda.
#10475 opened by Redmept1on - 0
oneflow.amax/amin perform differently from pytorch
#10474 opened by Redmept1on - 1
- 1
oneflow.argmax perform differently between cpu and cuda.
#10472 opened by Redmept1on - 0
oneflow.quantile perform differently between cpu and cuda.
#10471 opened by Redmept1on - 0
- 0
oneflow.log10 has accuracy difference between cpu and cuda
#10468 opened by Redmept1on - 0
oneflow.softmax perform differently between cpu and cuda.
#10469 opened by Redmept1on - 2
When nan is used as the min or max argument to oneflow.clamp, the wrong result is printed
#10467 opened by Redmept1on