Dev logical_and/or modules

[hengzi]

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 138.1ms (= 6904.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 125.7ms (= 6286.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 138.1ms / 125.7ms)

PyTorch resnet50 time: 82.3ms (= 4115.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 72.8ms (= 3639.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 82.3ms / 72.8ms)

PyTorch resnet50 time: 56.0ms (= 2798.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.5ms (= 2375.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.18 (= 56.0ms / 47.5ms)

PyTorch resnet50 time: 48.5ms (= 2427.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 44.2ms (= 2207.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 48.5ms / 44.2ms)

PyTorch resnet50 time: 43.8ms (= 2187.9ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.4ms (= 1971.7ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.11 (= 43.8ms / 39.4ms)