cannot init scalar tensor

Open
#635 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
3/5
Estimated time
1-2 days
Newbie friendliness
45/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Stale
Tech stack
numpy, python
Domain
machine-learning

Research direction

Start with test/python/test_operation.py around line 4015 and trace tensor.from_numpy through singa/tensor.py, especially copy_from_numpy. Run the provided np.array(256, ndmin=0) case and confirm scalar tensors can be created without the shape assertion.

Written by the indexing model from the issue text.

Description

Hi, @dcslin , since some onnx models need scalar-tensor but we can't support.

The error is:

Traceback (most recent call last):
File "../../test/python/test_operation.py", line 4015, in test_tmp
x = tensor.from_numpy(x)
File "/usr/local/lib/python3.5/dist-packages/singa/tensor.py", line 766, in from_numpy
ret.copy_from_numpy(np_array)
File "/usr/local/lib/python3.5/dist-packages/singa/tensor.py", line 307, in copy_from_numpy
assert np_array.size == self.size(), 'tensor shape should be the same'
AssertionError: tensor shape should be the same

please use this test case:

x = np.array(256, ndmin=0).astype(np.int32)
x = tensor.from_numpy(x)
Dominant language
C++
Stars
3.6k
Forks
1.3k
No merged PRs in 30d

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from apache/singa

All issues in apache/singa

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.