[primitive] add matmul test

[guo-ran]

No description provided.

[github-actions]

Speed stats:

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8137/

[github-actions]

CI failed when running job: cuda-benchmark. PR label automerge has been removed

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8137/

[github-actions]

Speed stats:

GPU Name: NVIDIA GeForce GTX 1080 

❌ OneFlow resnet50 time: 129.1ms (= 12910.3ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.8ms (= 14279.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.8ms / 129.1ms)

OneFlow resnet50 time: 78.8ms (= 7884.2ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.4ms (= 8438.7ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.07 (= 84.4ms / 78.8ms)

OneFlow resnet50 time: 52.7ms (= 10534.9ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 58.9ms (= 11786.9ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.12 (= 58.9ms / 52.7ms)

OneFlow resnet50 time: 44.0ms (= 8798.6ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 52.6ms (= 10527.5ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.20 (= 52.6ms / 44.0ms)

OneFlow resnet50 time: 38.2ms (= 7637.2ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 35.7ms (= 7146.7ms / 200, input_shape=[1, 3, 224, 224])
❌ Relative speed: 0.94 (= 35.7ms / 38.2ms)

OneFlow swin dataloader time: 0.246s (= 49.286s / 200, num_workers=1)
PyTorch swin dataloader time: 0.151s (= 30.289s / 200, num_workers=1)
Relative speed: 0.615 (= 0.151s / 0.246s)

OneFlow swin dataloader time: 0.068s (= 13.654s / 200, num_workers=4)
PyTorch swin dataloader time: 0.043s (= 8.534s / 200, num_workers=4)
Relative speed: 0.625 (= 0.043s / 0.068s)

OneFlow swin dataloader time: 0.037s (= 7.480s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.418s / 200, num_workers=8)
Relative speed: 0.591 (= 0.022s / 0.037s)

❌ OneFlow resnet50 time: 145.0ms (= 14498.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 169.3ms (= 16925.9ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 169.3ms / 145.0ms)

OneFlow resnet50 time: 103.0ms (= 10302.1ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 117.7ms (= 11774.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.14 (= 117.7ms / 103.0ms)

OneFlow resnet50 time: 77.1ms (= 15425.9ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 89.0ms (= 17804.6ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.15 (= 89.0ms / 77.1ms)

OneFlow resnet50 time: 63.6ms (= 12728.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 67.4ms (= 13485.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
❌ Relative speed: 1.06 (= 67.4ms / 63.6ms)

OneFlow resnet50 time: 55.5ms (= 11108.0ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.4ms (= 15681.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.41 (= 78.4ms / 55.5ms)