igor
igor
I don't understand. There is no error.
I didn't notice there was an example dataset. I tried it. I got an empty `test_results.nucpos.bed.gz`. I got 160 lines in `test_results.occpeaks.bed` and `test_results.nucmap_combined.bed`. I didn't see any errors. This...
NucleoATAC version: 0.3.3 (just reinstalled it to be safe) Cython version: 0.23.4 Python version: 2.7.3 Numpy version: 1.12.0 Scipy version: 0.18.1 Operating System: CentOS 6.8 nosetests result: ``` /path/NucleoATAC/nucleoatac/NucleosomeCalling.py:86: RuntimeWarning:...
I am not using a virtual environment. Is that a problem? I am not sure sure how to get just the dependencies, but here is my full `pip list`: ```...
I had trouble using `virtualenv`. After I activate it, `pip` still tries to install to the usual path and not to the local path. Even if I force it to...
I resolved my `virtualenv` issues. This is my `pip list` before running NucleoATAC, which now looks fine to me: ``` appdirs (1.4.0) cycler (0.10.0) Cython (0.25.2) functools32 (3.2.3.post2) matplotlib (2.0.0)...
I got some fails with that. ``` $ python tests.py ....................FF.. ====================================================================== FAIL: Make sure variance calculation is close to what is obtained by simulation ---------------------------------------------------------------------- Traceback (most recent call...
Do you know how I might be able to troubleshoot this? Is there a specific function that might be responsible that I can quickly test without requiring the whole test...
Thanks for the test code. I got an error for sd2: ``` >>> sd2 = signaldist.analStd() nucleoatac/NucleosomeCalling.py:86: RuntimeWarning: invalid value encountered in sqrt return np.sqrt(var) ``` Everything else was fine...
I think you meant sd2, not sd3, right? Here are the latest results: ``` >>> import pyximport; pyximport.install(setup_args={"include_dirs":np.get_include()}) (None, None) >>> from nucleoatac.multinomial_cov import calculateCov >>> >>> flatv = np.ravel(signaldist.vmat.mat)...