move scalar to oneflow namespace.

[hjchen2]

No description provided.

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 128.7ms (= 6436.1ms / 50, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 141.1ms (= 7055.0ms / 50, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 141.1ms / 128.7ms)

OneFlow resnet50 time: 75.0ms (= 3748.4ms / 50, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.4ms (= 4222.0ms / 50, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.13 (= 84.4ms / 75.0ms)

OneFlow resnet50 time: 48.8ms (= 2438.9ms / 50, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 60.3ms (= 3014.9ms / 50, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.24 (= 60.3ms / 48.8ms)

OneFlow resnet50 time: 41.2ms (= 2059.9ms / 50, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 47.8ms (= 2389.3ms / 50, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.16 (= 47.8ms / 41.2ms)

OneFlow resnet50 time: 41.5ms (= 2074.6ms / 50, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 43.5ms (= 2176.2ms / 50, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.05 (= 43.5ms / 41.5ms)

OneFlow resnet50 time: 155.9ms (= 7794.6ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 162.4ms (= 8119.5ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.04 (= 162.4ms / 155.9ms)

OneFlow resnet50 time: 102.5ms (= 5124.4ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 107.3ms (= 5367.1ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.05 (= 107.3ms / 102.5ms)

OneFlow resnet50 time: 78.7ms (= 3937.5ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.5ms (= 3777.0ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.96 (= 75.5ms / 78.7ms)

OneFlow resnet50 time: 73.1ms (= 3656.2ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.5ms (= 3725.7ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.02 (= 74.5ms / 73.1ms)

OneFlow resnet50 time: 69.0ms (= 3448.9ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.0ms (= 3251.8ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.94 (= 65.0ms / 69.0ms)