fix distributed run on built dir

[jackalcooper]

No description provided.

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 136.0ms (= 6799.1ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 125.8ms (= 6291.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.08 (= 136.0ms / 125.8ms)

PyTorch resnet50 time: 82.2ms (= 4107.8ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 73.0ms (= 3647.7ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 82.2ms / 73.0ms)

PyTorch resnet50 time: 57.0ms (= 2851.9ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.6ms (= 2377.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.20 (= 57.0ms / 47.6ms)

PyTorch resnet50 time: 47.5ms (= 2374.5ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 46.2ms (= 2310.9ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.03 (= 47.5ms / 46.2ms)

PyTorch resnet50 time: 45.0ms (= 2247.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 41.9ms (= 2095.4ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.07 (= 45.0ms / 41.9ms)