Closed bhclowers closed 3 years ago
Quick update. I installed the package on a separate Windows 10 machine running Python 3.7. From there I was able run the example file without issue (https://raw.githubusercontent.com/OverLordGoldDragon/ssqueezepy/master/examples/extracting_ridges.py). However, the examples on the README.md for the ridge extraction still throw an error with respect to an incorrect tuple size. This should get me going though.
Thanks for the report.
Assuming you ran without os.environ['SSQ_GPU'] = '1'
, this is a numba issue. A culprit to my mind is loose type handling in ssqueezepy
, which is taken care of fine on my system, but maybe not everywhere; an idea is to explicitly cast values via int
and float
where applicable, see if it helps.
Worth trying os.environ['SSQ_PARALLEL'] = '0'
, but this'll be much slower. If anything other than ridge extraction fails, can try os.environ['SSQ_GPU'] = '1'
to run via CuPy rather than numba.
On README, code was written in an earlier version and I've not gotten around to update it, but comparing with extracting_ridges.py
it shouldn't be hard to fix (a PR is welcome).
In case this helps someone else, here are the examples extracted into an ipynb.
I am able to get the examples shown in the front page to work fine, however, all attempts towards ridge extraction fail on my system. The frustrating part is that I don't even get a traceback, only total kernel failure. I've tried through the command line and through a jupyter notebook (I realize they are linking to the same kernel). Any ideas on how I might rectify this situation?
I've attempted to run the examples on the readme for ridge extraciton and observed the same result when running the core example. On the latter I can't proceed past line 63 (https://raw.githubusercontent.com/OverLordGoldDragon/ssqueezepy/master/examples/extracting_ridges.py).
Synchrosqueezing is an intriguing concept and am interested to try it out on our data.
System Profile: OSX 10.13.6 (High Sierra) Python 3.8 (Anaconda, Fresh Install)
Install/Dependency Trace: Downloading ssqueezepy-0.6.1-py3-none-any.whl (121 kB) |████████████████████████████████| 121 kB 1.2 MB/s Requirement already satisfied: numba in /opt/anaconda3/lib/python3.8/site-packages (from ssqueezepy) (0.53.1) Requirement already satisfied: scipy in /opt/anaconda3/lib/python3.8/site-packages (from ssqueezepy) (1.6.2) Requirement already satisfied: matplotlib in /opt/anaconda3/lib/python3.8/site-packages (from ssqueezepy) (3.3.4) Requirement already satisfied: numpy in /opt/anaconda3/lib/python3.8/site-packages (from ssqueezepy) (1.20.1) Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /opt/anaconda3/lib/python3.8/site-packages (from matplotlib->ssqueezepy) (2.4.7) Requirement already satisfied: pillow>=6.2.0 in /opt/anaconda3/lib/python3.8/site-packages (from matplotlib->ssqueezepy) (8.2.0) Requirement already satisfied: kiwisolver>=1.0.1 in /opt/anaconda3/lib/python3.8/site-packages (from matplotlib->ssqueezepy) (1.3.1) Requirement already satisfied: cycler>=0.10 in /opt/anaconda3/lib/python3.8/site-packages (from matplotlib->ssqueezepy) (0.10.0) Requirement already satisfied: python-dateutil>=2.1 in /opt/anaconda3/lib/python3.8/site-packages (from matplotlib->ssqueezepy) (2.8.1) Requirement already satisfied: six in /opt/anaconda3/lib/python3.8/site-packages (from cycler>=0.10->matplotlib->ssqueezepy) (1.15.0) Requirement already satisfied: llvmlite<0.37,>=0.36.0rc1 in /opt/anaconda3/lib/python3.8/site-packages (from numba->ssqueezepy) (0.36.0) Requirement already satisfied: setuptools in /opt/anaconda3/lib/python3.8/site-packages (from numba->ssqueezepy) (52.0.0.post20210125) Installing collected packages: ssqueezepy Successfully installed ssqueezepy-0.6.1