oneflow

oneflow copied to clipboard

### Brief Description  it should be: oneflow.index_select(input, dim, index) → Tensor ### Alternatives _No response_

Redempt1onzzZZ

When nan/inf is a parameter of 'oneflow.linspace', there is a core dumped happen

1

## Summary When nan/inf is a parameter of 'oneflow.linspace', there is a core dumped happen, while pytorch/numpy doesn't. ## Code to reproduce bug ```python import oneflow as flow # Create...

Redempt1onzzZZ

negative dimensions should not be allowed in 'oneflow.zeros/ones'

4

## Summary When the oneflow.zeros/ones parameter size contains negative values, the program aborted, there's a 'core dumped'. ## Code to reproduce bug ```python import oneflow as flow tensor = flow.ones((-1,3),...

Redempt1onzzZZ

## Summary when parameter of 'oneflow.eye' is float, it should throw an error or warning. ## Code to reproduce bug ```python import oneflow as flow # Create a 2-D tensor...

Redempt1onzzZZ

When nan/inf is a parameter of 'oneflow.arange', there is a core dumped happen, it should add a data type check here

1

## Summary When nan/inf is a parameter of 'oneflow.arange', there is a core dumped happen. 'numpy.arange' will throw an error. ## Code to reproduce bug ```python import oneflow as flow...

Redempt1onzzZZ

## Summary According to arange's calculation method (Generate numbers in the half open interval [start, stop)), when step is inf, the generated array should contain the start value, but oneflow.arange...

Redempt1onzzZZ

## Summary When oneflow.permute is used for the same dimension operation, there is no crash, but the wrong result is returned. Checks of the same dimension should be added. ##...

Redempt1onzzZZ

## Summary I run a code to test oneflow.var, when I use it to a 2 dimensions tensor, and I set dim parameter equals to -3, core dumped occurs. ##...

Redempt1onzzZZ

## Summary core dumped occurs when an empty array is processed with oneflow.tensordot ## Code to reproduce bug ```python import oneflow as flow import numpy as np a = flow.tensor(np.array([[],[]]),dtype=flow.float32)...

Redempt1onzzZZ

## Summary When nan is an element in tensor, the output of oneflow.max/min will have a value that does not correspond to indice. ## Code to reproduce bug for oneflow.max,...

Redempt1onzzZZ