do not create a new tensor for inplaced output

[hjchen2]

No description provided.

[hjchen2]

单测会在eager consistent的单测中一起加上来

[github-actions]

CI failed when running job: cpu-module. PR label automerge has been removed

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 136.7ms (= 13668.1ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 139.5ms (= 13952.0ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.02 (= 139.5ms / 136.7ms)

OneFlow resnet50 time: 78.4ms (= 7843.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 82.6ms (= 8260.7ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.05 (= 82.6ms / 78.4ms)

OneFlow resnet50 time: 53.5ms (= 10697.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 57.0ms (= 11401.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.07 (= 57.0ms / 53.5ms)

OneFlow resnet50 time: 42.0ms (= 8396.7ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 47.4ms (= 9479.7ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.13 (= 47.4ms / 42.0ms)

OneFlow resnet50 time: 37.0ms (= 7396.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.8ms (= 7968.6ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.08 (= 39.8ms / 37.0ms)

OneFlow resnet50 time: 148.2ms (= 14815.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 159.3ms (= 15926.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.07 (= 159.3ms / 148.2ms)

OneFlow resnet50 time: 89.7ms (= 8965.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 97.8ms (= 9779.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.09 (= 97.8ms / 89.7ms)

OneFlow resnet50 time: 64.9ms (= 12987.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 73.3ms (= 14650.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.13 (= 73.3ms / 64.9ms)

OneFlow resnet50 time: 61.7ms (= 12336.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 61.7ms (= 12331.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.00 (= 61.7ms / 61.7ms)

OneFlow resnet50 time: 60.6ms (= 12112.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 57.0ms (= 11404.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.94 (= 57.0ms / 60.6ms)