print consistent tensor data

[liufengwei0103]

print consistent tensor data
placement = flow.placement("cuda", {0: range(2)})
1.tensor = flow.ones((10, 10), placement=placement, sbp=[flow.sbp.split(0)])

2.tensor = flow.ones((100, 100), placement=placement, sbp=[flow.sbp.split(0)])

3.tensor = flow.ones((0, 100), placement=placement, sbp=[flow.sbp.split(0)])

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 139.8ms (= 6989.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.1ms (= 6403.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 139.8ms / 128.1ms)

PyTorch resnet50 time: 85.3ms (= 4266.3ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.5ms (= 3723.1ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.15 (= 85.3ms / 74.5ms)

PyTorch resnet50 time: 58.1ms (= 2905.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 48.4ms (= 2417.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.20 (= 58.1ms / 48.4ms)

PyTorch resnet50 time: 47.1ms (= 2353.1ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 41.3ms (= 2063.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.14 (= 47.1ms / 41.3ms)

PyTorch resnet50 time: 43.3ms (= 2163.4ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 43.9ms (= 2197.1ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 0.98 (= 43.3ms / 43.9ms)