Remove gpu_device_num

[liujuncheng]

No description provided.

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8516/

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.0ms (= 12799.5ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.1ms (= 14212.7ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.1ms / 128.0ms)

OneFlow resnet50 time: 74.7ms (= 7468.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.3ms (= 8434.0ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.13 (= 84.3ms / 74.7ms)

OneFlow resnet50 time: 47.8ms (= 9560.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 56.9ms (= 11380.0ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.19 (= 56.9ms / 47.8ms)

OneFlow resnet50 time: 39.2ms (= 7842.2ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 43.0ms (= 8596.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.10 (= 43.0ms / 39.2ms)

OneFlow resnet50 time: 35.5ms (= 7092.1ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.4ms (= 7871.0ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.11 (= 39.4ms / 35.5ms)

OneFlow swin dataloader time: 0.255s (= 50.964s / 200, num_workers=1)
PyTorch swin dataloader time: 0.150s (= 30.032s / 200, num_workers=1)
Relative speed: 0.589 (= 0.150s / 0.255s)

OneFlow swin dataloader time: 0.072s (= 14.408s / 200, num_workers=4)
PyTorch swin dataloader time: 0.045s (= 9.023s / 200, num_workers=4)
Relative speed: 0.626 (= 0.045s / 0.072s)

OneFlow swin dataloader time: 0.040s (= 7.973s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.398s / 200, num_workers=8)
Relative speed: 0.552 (= 0.022s / 0.040s)

❌ OneFlow resnet50 time: 136.5ms (= 13650.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 164.2ms (= 16418.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 164.2ms / 136.5ms)

OneFlow resnet50 time: 83.8ms (= 8375.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 103.6ms (= 10364.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.24 (= 103.6ms / 83.8ms)

OneFlow resnet50 time: 56.3ms (= 11255.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.2ms (= 15649.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.39 (= 78.2ms / 56.3ms)

OneFlow resnet50 time: 46.6ms (= 9323.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.0ms (= 14801.9ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.59 (= 74.0ms / 46.6ms)

OneFlow resnet50 time: 42.1ms (= 8423.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.1ms (= 15422.5ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.83 (= 77.1ms / 42.1ms)