Add proxy env in CI

[jackalcooper]

No description provided.

[github-actions]

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10407/

[github-actions]

Speed stats:

GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.4ms (= 4343.6ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 57.7ms (= 5773.0ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.33 (= 57.7ms / 43.4ms)

OneFlow resnet50 time: 26.6ms (= 2657.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 38.4ms (= 3845.0ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.45 (= 38.4ms / 26.6ms)

OneFlow resnet50 time: 18.2ms (= 3635.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 35.3ms (= 7052.5ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.94 (= 35.3ms / 18.2ms)

OneFlow resnet50 time: 17.7ms (= 3544.9ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 31.0ms (= 6201.8ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.75 (= 31.0ms / 17.7ms)

OneFlow resnet50 time: 16.8ms (= 3350.5ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 29.6ms (= 5912.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.76 (= 29.6ms / 16.8ms)

OneFlow swin dataloader time: 0.200s (= 40.056s / 200, num_workers=1)
PyTorch swin dataloader time: 0.128s (= 25.563s / 200, num_workers=1)
Relative speed: 0.638 (= 0.128s / 0.200s)

OneFlow swin dataloader time: 0.055s (= 11.085s / 200, num_workers=4)
PyTorch swin dataloader time: 0.034s (= 6.724s / 200, num_workers=4)
Relative speed: 0.607 (= 0.034s / 0.055s)

OneFlow swin dataloader time: 0.030s (= 6.064s / 200, num_workers=8)
PyTorch swin dataloader time: 0.017s (= 3.388s / 200, num_workers=8)
Relative speed: 0.559 (= 0.017s / 0.030s)

❌ OneFlow resnet50 time: 49.2ms (= 4924.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.4ms (= 6542.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.33 (= 65.4ms / 49.2ms)

OneFlow resnet50 time: 37.0ms (= 3698.4ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 47.7ms (= 4769.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.29 (= 47.7ms / 37.0ms)

OneFlow resnet50 time: 28.4ms (= 5676.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 40.4ms (= 8088.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.42 (= 40.4ms / 28.4ms)

OneFlow resnet50 time: 24.9ms (= 4987.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 39.0ms (= 7805.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.57 (= 39.0ms / 24.9ms)

OneFlow resnet50 time: 24.1ms (= 4816.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 36.5ms (= 7298.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.52 (= 36.5ms / 24.1ms)

[github-actions]

View latest API docs preview at: https://oneflow-staging.oss-cn-beijing.aliyuncs.com/docs/Oneflow-Inc/oneflow/pr/10407/

[github-actions]

Speed stats:

GPU Name: NVIDIA GeForce RTX 3080 Ti 

❌ OneFlow resnet50 time: 43.7ms (= 4365.8ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 57.5ms (= 5754.8ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.32 (= 57.5ms / 43.7ms)

OneFlow resnet50 time: 26.5ms (= 2650.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 37.5ms (= 3751.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.42 (= 37.5ms / 26.5ms)

OneFlow resnet50 time: 18.4ms (= 3688.3ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 35.3ms (= 7050.1ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.91 (= 35.3ms / 18.4ms)

OneFlow resnet50 time: 17.0ms (= 3398.4ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 31.2ms (= 6240.4ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.84 (= 31.2ms / 17.0ms)

OneFlow resnet50 time: 16.7ms (= 3345.4ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 31.4ms (= 6272.3ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.87 (= 31.4ms / 16.7ms)

OneFlow swin dataloader time: 0.203s (= 40.590s / 200, num_workers=1)
PyTorch swin dataloader time: 0.128s (= 25.610s / 200, num_workers=1)
Relative speed: 0.631 (= 0.128s / 0.203s)

OneFlow swin dataloader time: 0.056s (= 11.164s / 200, num_workers=4)
PyTorch swin dataloader time: 0.032s (= 6.447s / 200, num_workers=4)
Relative speed: 0.577 (= 0.032s / 0.056s)

OneFlow swin dataloader time: 0.032s (= 6.312s / 200, num_workers=8)
PyTorch swin dataloader time: 0.016s (= 3.296s / 200, num_workers=8)
Relative speed: 0.522 (= 0.016s / 0.032s)

❌ OneFlow resnet50 time: 49.3ms (= 4934.4ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.2ms (= 6515.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.32 (= 65.2ms / 49.3ms)

OneFlow resnet50 time: 37.1ms (= 3709.8ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 45.8ms (= 4582.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 45.8ms / 37.1ms)

OneFlow resnet50 time: 28.3ms (= 5659.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 39.8ms (= 7957.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.41 (= 39.8ms / 28.3ms)

OneFlow resnet50 time: 24.6ms (= 4921.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 41.1ms (= 8229.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.67 (= 41.1ms / 24.6ms)

OneFlow resnet50 time: 24.0ms (= 4790.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 35.9ms (= 7176.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.50 (= 35.9ms / 24.0ms)