Hey, umap and pynndescent are super cool, but, I'm having major headaches on querying for neighbors in pynndescent. I've been experimenting with different versions of pynndescent and numba but none help. The issue pasted below has the following library versions:
numba: 0.51.2
pynndescent: 0.5.0
I also had the latests:
numba: 0.53.1
pynndescent: 0.5.3
with the same error. Btw, for umap, I changed a ton of ints to int32, whenever you had something like x=some_arr.shape[0]. I don't know why shape[0] was giving trouble but not shape[1].
in
----> 1 index.query(pivoted.iloc[0],k=5)
~/miniconda3/envs/pynn/lib/python3.7/site-packages/pynndescent/pynndescent_.py in query(self, query_data, k, epsilon)
1597 epsilon,
1598 self._visited,
-> 1599 self.search_rng_state,
1600 )
1601 else:
~/miniconda3/envs/pynn/lib/python3.7/site-packages/numba/core/dispatcher.py in _compile_for_args(self, *args, **kws)
413 e.patch_message(msg)
414
--> 415 error_rewrite(e, 'typing')
416 except errors.UnsupportedError as e:
417 # Something unsupported is present in the user code, add help info
~/miniconda3/envs/pynn/lib/python3.7/site-packages/numba/core/dispatcher.py in error_rewrite(e, issue_type)
356 raise e
357 else:
--> 358 reraise(type(e), e, None)
359
360 argtypes = []
~/miniconda3/envs/pynn/lib/python3.7/site-packages/numba/core/utils.py in reraise(tp, value, tb)
78 value = tp()
79 if value.__traceback__ is not tb:
---> 80 raise value.with_traceback(tb)
81 raise value
82
TypingError: Failed in nopython mode pipeline (step: nopython frontend)
No conversion from float32 to array(float32, 1d, C) for 'current_query', defined at None
File "../../../../../miniconda3/envs/pynn/lib/python3.7/site-packages/pynndescent/pynndescent_.py", line 1222:
def search_closure(
Hey, umap and pynndescent are super cool, but, I'm having major headaches on querying for neighbors in pynndescent. I've been experimenting with different versions of pynndescent and numba but none help. The issue pasted below has the following library versions:
numba: 0.51.2 pynndescent: 0.5.0
I also had the latests: numba: 0.53.1 pynndescent: 0.5.3
with the same error. Btw, for umap, I changed a ton of ints to int32, whenever you had something like x=some_arr.shape[0]. I don't know why shape[0] was giving trouble but not shape[1].
**index = NNDescent(pivoted.fillna(-99),metric='euclidean')
index.query(pivoted.iloc[0],k=5)**
TypingError Traceback (most recent call last)