Michael Penkov comments

Results 254 comments of


                                            Michael Penkov

Using efficient np.linalg.multi_dot

Thank you for your effort @maldil . I see this change has pros and cons. Pros: _maybe_ more efficient Cons: more obscure Given that efficiency is not a prime concern...

AttributeError: 'tuple' object has no attribute 'startswith'

I put in some print statements at various points of the stack trace, and got this: ``` apply_async.options: {'queue': None, 'routing_key': None, 'exchange': None, 'priority': None, 'expires': None, 'serializer': 'json',...

AttributeError: 'tuple' object has no attribute 'startswith'

``` (dbi2) sergeyich:dbi2 misha$ pip freeze | grep celery celery==4.1.1 (dbi2) sergeyich:dbi2 misha$ pip freeze | grep kombu kombu==4.2.1```

AttributeError: 'tuple' object has no attribute 'startswith'

Thank you for the fast reply. I managed to get it working by specifying queue as a keyword argument to the apply_async function. I'm happy to debug via celery.app.routes, but...

Version 2.0.0 significantly slower

Interestingly, cannot reproduce on Py3.9 Linux: ``` (testenv) misha@cabron:~/git/sqlitedict$ python --version Python 3.9.14 (testenv) misha@cabron:~/git/sqlitedict$ time python repro.py real 0m0.107s user 0m0.090s sys 0m0.016s

Probably avoidable fatal S3 race condition

Sure, go for it.

Allow pass in additional low level API keyword arguments to AWS S3 boto3 put_object

Yes, I think so.

Fix `KeyedVectors.add_vectors()` error when use most_similar

Needs a test. The example from the issue description is probably good enough.

poincare model does not work when workers > 1

Please post a stack trace.

docs: Fix a few typos

Thank you @timgates42 !