ZZK comments

Results 31 comments of

ZZK

Fix BroadcastElemenetwiseUnary launch error

不连续的情况下，用pack需要保证连续。但是这里判断是否连续比较麻烦，后续用unroll展开处理

Add Mario oneflow Backend

> add a oneflow version badge in `README.md`? like PyTorch Sure，it look like this: ![image](https://user-images.githubusercontent.com/42901638/195038180-3a9ddad9-8bdc-46d7-b6c8-a5b4feca0714.png)

AttributeError: module 'oneflow.nn' has no attribute 'DataParallel'

pytorch现在都是推荐用 DDP了，而不是DP

oneflow.nn.MovingAverageMinMaxObserver produces "Aborted (core dumped)" bug

Try to use this: ```python input1 = oneflow.rand(9, 40, 20, 10,dtype=oneflow.float64) input2 = oneflow.randint(0,5,(1,),dtype=oneflow.int64) mod = oneflow.nn.MovingAverageMinMaxObserver(training= True, stop_update_after_iters=1, quantization_formula= 'google', quantization_bit=8, quantization_scheme='symmetric', momentum= 0.95).double() # Add .double() output =...

oneflow.nn.Fold may produce inconsistent results when runned w/ and w/o requires_grad_(True)

![image](https://user-images.githubusercontent.com/42901638/181176901-fcbf2c5e-b1f3-451a-b268-b7a80c00a0e5.png) ![image](https://user-images.githubusercontent.com/42901638/181177493-0b1667d8-7d40-4e97-b73d-c86053fa6fc1.png) I just follow your script, and get different result of `count`. It seems like a Floating point error by accident, I think it is not a Bug? Or...

Cannot run oneflow.roi_align() properly

Currently OneFlow only support ROIAlign Operator in CUDA, you can set the tensor `device` property as "cuda"

oneflow.nn.conv2D causes "core dumped"

You can change to use this: ```python import oneflow input = oneflow.rand(4,10,32,32,dtype=oneflow.float64) m = oneflow.nn.Conv2d(10,10,3).double() output = m(input) ``` `.double()` will change the parameter to float64 dtype(Default is float32 dtype....

ZZK