Closed taopeng1100 closed 4 years ago
@taopeng1100: This is a numba bug. Please report this there, but only if you use the newest numba version (otherwise install it and try to reproduce this with the newest version). Give them this link so they see the code that triggers their bug:
@team: We should include numba in the package versions list.
Can you let me know the version of numba that worked for you?
Get Outlook for Androidhttps://aka.ms/ghei36
From: Philipp A. notifications@github.com Sent: Thursday, July 30, 2020 2:06:10 AM To: theislab/scanpy scanpy@noreply.github.com Cc: Peng, Tao tpeng@fredhutch.org; Mention mention@noreply.github.com Subject: Re: [theislab/scanpy] scanpy sc.pp.calculate_qc_metrics(adata, qc_vars=["mt"], inplace=True) (#1341)
@taopeng1100https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_taopeng1100&d=DwMCaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=j6EgtEBZ-6pbDONgnwVzuTHJ6L-gWcikckOhZCwVjPc&m=j6WdQxFMfGL2BOrFUsLxuCIlMEWSPOGMoQ3sfPhU4hw&s=mZvFxBCIH_-96BcNkwcrYuLqIFdGUQVv9-NgmpF6Kyc&e=: This is a numba bug. Please report this there, but only if you use the newest numba version. Give them this link so they see the code that triggers their bug:
https://github.com/theislab/scanpy/blob/2f160ea403d124d237fc2138c0aa0d175fbad22a/scanpy/preprocessing/_qc.py#L402-L428https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_theislab_scanpy_blob_2f160ea403d124d237fc2138c0aa0d175fbad22a_scanpy_preprocessing_-5Fqc.py-23L402-2DL428&d=DwMCaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=j6EgtEBZ-6pbDONgnwVzuTHJ6L-gWcikckOhZCwVjPc&m=j6WdQxFMfGL2BOrFUsLxuCIlMEWSPOGMoQ3sfPhU4hw&s=6bLtPWJSoSBSNQcRvwP4Pp-mHjA7d51Ees_qcvGEJYk&e=
@team: We should include numba in the package versions list.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_theislab_scanpy_issues_1341-23issuecomment-2D666246043&d=DwMCaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=j6EgtEBZ-6pbDONgnwVzuTHJ6L-gWcikckOhZCwVjPc&m=j6WdQxFMfGL2BOrFUsLxuCIlMEWSPOGMoQ3sfPhU4hw&s=AQB0zBmdSewipAgOqteV0t7F27jeGZaU-I4LngYZYus&e=, or unsubscribehttps://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ALCZ3CPX4MWCNAHOFNSYRC3R6EZYFANCNFSM4PMLDJKQ&d=DwMCaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=j6EgtEBZ-6pbDONgnwVzuTHJ6L-gWcikckOhZCwVjPc&m=j6WdQxFMfGL2BOrFUsLxuCIlMEWSPOGMoQ3sfPhU4hw&s=XT2LyZ_ZXTKdnWcyG0sd7w1UgmMmxTpvj_8IWLPy44I&e=.
You can find the newest version on the PyPI page: https://pypi.org/project/numba/
Think this is a duplicate of https://github.com/numba/numba/issues/5955
Thank you @stuartarchibald, it sure is! The error happens when numba tries to JIT-compile top_segment_proportions_sparse_csr
.
Hi Philip, can you kindly help me here? How can I get “sc.pp.calculate_qc_metrics(adata, qc_vars=["mt"], inplace=True)” to work for me?
Maybe downgrade numba for the time being? IDK to which version though. @stuartarchibald has more insight here. Please follow numba/numba#5955 for updates!
I am using Anaconda/Jupyter in my PC. When I try to downgrade numba, I run into issues of numba dependency packages in Anaconda so I am stuck!
@flying-sheep I might know a work around for scanpy whilst we try and fix the root cause in numba, would that help?
Sure, thank you! Care to do a quick PR? Then we can point @taopeng1100 in the direction of installing scanpy’s dev version and everyone’s happy.
@taopeng1100 please reply by GitHub comment and not by email anymore, it spams up this comment section. I always have to remove some junk your email program adds.
@flying-sheep done https://github.com/theislab/scanpy/pull/1359
Okay, it’s merged! @taopeng1100, please install the dev version of scanpy like this, and retry:
pip install git+https://github.com/theislab/scanpy.git
# or
pip install --user git+https://github.com/theislab/scanpy.git
It all works NOW! I appreciate your help!
It all works NOW! I appreciate your help!
It all works NOW! I appreciate your help!
It all works NOW! I appreciate your help!
After I use function
I use
sc.pp.calculate_qc_metrics(adata, qc_vars=["mt"], inplace=True)
and got an error message:The package version is as follows:
scanpy==1.5.1 anndata==0.7.4 umap==0.4.6 numpy==1.18.5 scipy==1.5.0 pandas==1.0.5 scikit-learn==0.23.1 statsmodels==0.11.1 python-igraph==0.7.1+5.3b99dbf6 leidenalg==0.7.0