Dev linalg inv

[simonJJJ]

[zhongshsh]

记得在 /python/oneflow/framework/docstr/__init__.py import doc

[github-actions[bot]]

CI failed when running job: Build cpu. PR label automerge has been removed

[github-actions[bot]]

Code got formatted by CI. Please request CI again if you still want to have this PR merged. If the PR is from a forked repo, please download the patch files from the GitHub Actions web page and apply them locally.

[github-actions[bot]]

CI failed when running job: Build cpu. PR label automerge has been removed

[github-actions[bot]]

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.5ms (= 12852.9ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 142.5ms (= 14250.6ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.11 (= 142.5ms / 128.5ms)

OneFlow resnet50 time: 75.3ms (= 7531.3ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 86.2ms (= 8615.9ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.14 (= 86.2ms / 75.3ms)

OneFlow resnet50 time: 49.1ms (= 9814.5ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 60.3ms (= 12069.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.23 (= 60.3ms / 49.1ms)

OneFlow resnet50 time: 36.5ms (= 7295.8ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 42.5ms (= 8490.6ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.16 (= 42.5ms / 36.5ms)

OneFlow resnet50 time: 28.6ms (= 5718.2ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 43.5ms (= 8702.0ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.52 (= 43.5ms / 28.6ms)

OneFlow swin dataloader time: 0.413s (= 82.575s / 200, num_workers=1)
PyTorch swin dataloader time: 0.150s (= 29.938s / 200, num_workers=1)
Relative speed: 0.363 (= 0.150s / 0.413s)

OneFlow swin dataloader time: 0.112s (= 22.320s / 200, num_workers=4)
PyTorch swin dataloader time: 0.040s (= 8.090s / 200, num_workers=4)
Relative speed: 0.362 (= 0.040s / 0.112s)

OneFlow swin dataloader time: 0.041s (= 8.251s / 200, num_workers=8)
PyTorch swin dataloader time: 0.022s (= 4.464s / 200, num_workers=8)
Relative speed: 0.541 (= 0.022s / 0.041s)

❌ OneFlow resnet50 time: 136.4ms (= 13641.0ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 160.9ms (= 16093.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 160.9ms / 136.4ms)

OneFlow resnet50 time: 85.4ms (= 8535.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.8ms (= 10282.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 102.8ms / 85.4ms)

OneFlow resnet50 time: 58.2ms (= 11638.7ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.6ms (= 15710.5ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.35 (= 78.6ms / 58.2ms)

OneFlow resnet50 time: 46.1ms (= 9217.7ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 71.7ms (= 14344.3ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.56 (= 71.7ms / 46.1ms)

OneFlow resnet50 time: 39.2ms (= 7847.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.6ms (= 14924.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.90 (= 74.6ms / 39.2ms)

[github-actions[bot]]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/8183/