oneflow
oneflow copied to clipboard
Oneflow-Inc
Inplace masked fill
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] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
flow.masked_fill() 接口需要包一下
好的，我看 torch 里只有 torch.Tensor.masked_fill_， 所以这个 flow 下的接口，我也只导出，不添加文档了哈。
CI failed when running job: cpu-module. PR label automerge has been removed
View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/9133/
Speed stats:
GPU Name: GeForce GTX 1080
❌ OneFlow resnet50 time: 139.6ms (= 13964.2ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 160.7ms (= 16069.1ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 160.7ms / 139.6ms)
OneFlow resnet50 time: 85.4ms (= 8543.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.7ms (= 10274.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.20 (= 102.7ms / 85.4ms)
OneFlow resnet50 time: 58.2ms (= 11646.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.0ms (= 15596.0ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.34 (= 78.0ms / 58.2ms)
OneFlow resnet50 time: 44.7ms (= 8943.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.0ms (= 15007.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.68 (= 75.0ms / 44.7ms)
OneFlow resnet50 time: 40.5ms (= 8094.3ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 67.8ms (= 13552.9ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.67 (= 67.8ms / 40.5ms)
Speed stats:
GPU Name: GeForce GTX 1080
❌ OneFlow resnet50 time: 140.6ms (= 14055.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 163.5ms (= 16352.6ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.16 (= 163.5ms / 140.6ms)
OneFlow resnet50 time: 85.9ms (= 8590.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.2ms (= 10120.5ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 101.2ms / 85.9ms)
OneFlow resnet50 time: 58.3ms (= 11659.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 78.3ms (= 15653.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.34 (= 78.3ms / 58.3ms)
OneFlow resnet50 time: 45.3ms (= 9061.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 70.4ms (= 14078.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.55 (= 70.4ms / 45.3ms)
OneFlow resnet50 time: 40.2ms (= 8041.2ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 77.7ms (= 15542.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.93 (= 77.7ms / 40.2ms)
CI failed when running job: Build cu102. PR label automerge has been removed
View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/9133/
Speed stats:
GPU Name: GeForce GTX 1080
❌ OneFlow resnet50 time: 139.6ms (= 13960.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 160.5ms (= 16049.8ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 160.5ms / 139.6ms)
OneFlow resnet50 time: 85.7ms (= 8566.7ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 104.4ms (= 10437.0ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.22 (= 104.4ms / 85.7ms)
OneFlow resnet50 time: 58.1ms (= 11614.8ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 87.8ms (= 17561.1ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.51 (= 87.8ms / 58.1ms)
OneFlow resnet50 time: 45.4ms (= 9087.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 70.8ms (= 14151.2ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.56 (= 70.8ms / 45.4ms)
OneFlow resnet50 time: 40.3ms (= 8050.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 68.8ms (= 13752.1ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.71 (= 68.8ms / 40.3ms)