add emoji in speed test output, make it continue-on-error

[daquexian]

把 speed test 设成 continue-on-error: true，不让 ci 失败
在它的输出里添加 ✔️ 和 ❌ 展示本项速度是否达标

[github-actions]

Speed stats:

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 128.8ms (= 6442.3ms / 50, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.9ms (= 7145.0ms / 50, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.9ms / 128.8ms)

OneFlow resnet50 time: 75.4ms (= 3772.3ms / 50, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.6ms (= 4231.8ms / 50, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.12 (= 84.6ms / 75.4ms)

OneFlow resnet50 time: 48.0ms (= 2401.4ms / 50, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 58.6ms (= 2929.4ms / 50, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.22 (= 58.6ms / 48.0ms)

OneFlow resnet50 time: 44.4ms (= 2221.5ms / 50, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 49.0ms (= 2448.1ms / 50, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.10 (= 49.0ms / 44.4ms)

OneFlow resnet50 time: 40.9ms (= 2043.5ms / 50, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 41.6ms (= 2077.5ms / 50, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.02 (= 41.6ms / 40.9ms)

OneFlow resnet50 time: 160.8ms (= 8041.7ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 158.7ms (= 7933.0ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.99 (= 158.7ms / 160.8ms)

OneFlow resnet50 time: 102.0ms (= 5102.3ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 99.9ms (= 4995.8ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.98 (= 99.9ms / 102.0ms)

OneFlow resnet50 time: 78.3ms (= 3916.6ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 82.3ms (= 4116.0ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.05 (= 82.3ms / 78.3ms)

OneFlow resnet50 time: 67.5ms (= 3376.4ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 60.7ms (= 3035.0ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.90 (= 60.7ms / 67.5ms)

OneFlow resnet50 time: 67.0ms (= 3349.4ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 63.9ms (= 3192.9ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.95 (= 63.9ms / 67.0ms)

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 129.2ms (= 6460.6ms / 50, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 143.2ms (= 7161.2ms / 50, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 143.2ms / 129.2ms)

OneFlow resnet50 time: 74.9ms (= 3744.4ms / 50, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 83.4ms (= 4168.5ms / 50, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 83.4ms / 74.9ms)

OneFlow resnet50 time: 49.5ms (= 2473.8ms / 50, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 59.4ms (= 2969.3ms / 50, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.20 (= 59.4ms / 49.5ms)

OneFlow resnet50 time: 40.4ms (= 2018.0ms / 50, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 49.8ms (= 2490.8ms / 50, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.23 (= 49.8ms / 40.4ms)

OneFlow resnet50 time: 39.7ms (= 1984.7ms / 50, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 44.4ms (= 2220.0ms / 50, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.12 (= 44.4ms / 39.7ms)

OneFlow resnet50 time: 159.2ms (= 7961.0ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 166.1ms (= 8306.4ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.04 (= 166.1ms / 159.2ms)

OneFlow resnet50 time: 101.5ms (= 5077.0ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 106.8ms (= 5337.8ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.05 (= 106.8ms / 101.5ms)

OneFlow resnet50 time: 84.8ms (= 4240.0ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 80.2ms (= 4010.2ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.95 (= 80.2ms / 84.8ms)

OneFlow resnet50 time: 71.1ms (= 3553.7ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 71.1ms (= 3555.7ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.00 (= 71.1ms / 71.1ms)

OneFlow resnet50 time: 68.4ms (= 3417.8ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 65.1ms (= 3256.2ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.95 (= 65.1ms / 68.4ms)