1ndigo
1ndigo
Thanks for explaining.
Higher precision printing is indeed important to me because sometimes a lot of manual testing is required, and it can provide a more intuitive observation of the results of the...
i met the same issue, any solution?
Thanks a lot, it worked as expected!
oneflow.argmin has this issue too ```python import oneflow as flow import numpy as np x1 = flow.tensor(np.array([[float('inf'), 0, -1, float('nan'), 5]], dtype=np.float32)) x1 = x1.cuda() y1 = flow.argmin(x1,dim=0) print(y1.device,y1) x1...
when inf is step, there is no core dumped
oneflow.median has the same bug ```python import oneflow as flow import numpy as np x1 = flow.tensor(np.array([[float('inf'), 0, -1, float('nan'), 5]], dtype=np.float32)) x1 = x1.cuda() y1 = flow.median(x1,dim=1) print(y1) x1...
Sorry to bother, it's obvious that my issues are ignored, maybe some confirmation or attention needed. @levi131
已提交commit进行修改，望审阅 #263
I notice that in `/mlx/mlx/backend/cpu/inverse.cpp`, in the func `general_inv`, current src code simply judge the info!=0, maybe adding an extra judgement if (info > 0) can solve this problem. My...