add zhangshen op-test

[zhangshen12356]

本次完成以下的单测内容：

• oneflow.matmul
• oneflow.Tensor.matmul
• oneflow.negative
• oneflow.neg
• oneflow.Tensor.negative
• oneflow.nn.AvgPool1d
• oneflow.nn.AvgPool2d
• oneflow.nn.AvgPool3d
• oneflow.nn.AdaptiveAvgPool1d
• oneflow.nn.AdaptiveAvgPool2d
• oneflow.nn.AdaptiveAvgPool3d（PyTorch代码有Bug：Wrong gradients generated by GPU version nn.AdaptiveAvgPool3d pytorch/pytorch#62250，暂时跳过）
• oneflow.nn.MaxPool1d
• oneflow.nn.MaxPool2d
• oneflow.nn.MaxPool3d
• oneflow.repeat（PyTorch没有）
• oneflow.Tensor.repeat
• oneflow.tile
• oneflow.Tensor.tile

[CLAassistant]

All committers have signed the CLA.

[terryzhao127]

@zhangshen12356 nn.AdaptiveAvgPoolXd的bug已经解决

[terryzhao127]

@BBuf @zhangshen12356 nn.AdaptiveAvgPool3d GPU版本存在bug，PyTorch已经在1.10.0的nightly版本修复。因此暂时跳过，直到我们使用上述版本的PyTorch进行测试。

[BBuf]

@BBuf @zhangshen12356 nn.AdaptiveAvgPool3d GPU版本存在bug，PyTorch已经在1.10.0的nightly版本修复。因此暂时跳过，直到我们使用上述版本的PyTorch进行测试。

好，你先处理一下pr冲突。

[terryzhao127]

@zhangshen12356 你可以检查一下旧的test，如果不是使用numpy模拟的，是直接通过hard-coded输入输出测试的，都可以删除了

[zhangshen12356]

@zhangshen12356 你可以检查一下旧的test，如果不是使用numpy模拟的，是直接通过hard-coded输入输出测试的，都可以删除了

好的

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 141.3ms (= 7065.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.3ms (= 6416.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 141.3ms / 128.3ms)

PyTorch resnet50 time: 83.7ms (= 4183.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.6ms (= 3730.9ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.12 (= 83.7ms / 74.6ms)

PyTorch resnet50 time: 58.1ms (= 2906.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 48.1ms (= 2405.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.21 (= 58.1ms / 48.1ms)

PyTorch resnet50 time: 47.1ms (= 2357.1ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 38.6ms (= 1930.5ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.22 (= 47.1ms / 38.6ms)

PyTorch resnet50 time: 43.9ms (= 2194.2ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 38.1ms (= 1907.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.15 (= 43.9ms / 38.1ms)

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.0ms (= 6998.0ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.5ms (= 6422.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 140.0ms / 128.5ms)

PyTorch resnet50 time: 83.8ms (= 4190.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.5ms (= 3724.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 83.8ms / 74.5ms)

PyTorch resnet50 time: 56.3ms (= 2817.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.2ms (= 2362.5ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.19 (= 56.3ms / 47.2ms)

PyTorch resnet50 time: 48.2ms (= 2410.4ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 38.3ms (= 1916.2ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.26 (= 48.2ms / 38.3ms)

PyTorch resnet50 time: 44.7ms (= 2233.2ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 35.9ms (= 1795.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.24 (= 44.7ms / 35.9ms)