Overlap function met error using multi-cpus

[ttbond]

Hi~
Pyranges is a really helpful python pacakge for my research! Many thanks!
But I have a problem, that is, when I use overlap function, the following error is asserted:

File "/data/home/mySrc/xxx.py", line 168, in <module>  
    counter.countExon('baby1')  
  File "/data/home/mySrc/xxx.py", line 145, in countExoner  
    if self.checkAnyGeneRela(tmpPrList):  
  File "/data/home/mySrc/xxx.py", line 125, in checkAnyGeneRela  
    overRel = refGenePr.overlap(myPr, nb_cpu=self.threads)  # here, self.threads was setted as 2  
  File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/pyranges.py", line 3494, in overlap  
    dfs = pyrange_apply(_overlap, self, other, **kwargs)  
  File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 291, in pyrange_apply  
    df, odf = make_binary_sparse(kwargs, df, odf)  
  File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 134, in make_binary_sparse  
    odf = make_sparse(odf)  
  File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 113, in make_sparse  
    if "Strand" in df:  
TypeError: argument of type 'ray._raylet.ObjectRef' is not iterable

Actually, I tried different versions of python enviroment and none of them works (with different problems related to ray).
Thanks again for developing such a valuable package. Look for your favorable reply.

[endrebak]

I’ll have a look

…

On Fri, 17 Mar 2023 at 03:41, ttbond ***@***.***> wrote: Hi~ Pyranges is a really helpful python pacakge for my research! Many thanks! But I have a problem, that is, when I use overlap function, the following error is asserted: File "/data/home/mySrc/xxx.py", line 168, in <module> counter.countExon('baby1') File "/data/home/mySrc/xxx.py", line 145, in countExoner if self.checkAnyGeneRela(tmpPrList): File "/data/home/mySrc/xxx.py", line 125, in checkAnyGeneRela overRel = refGenePr.overlap(myPr, nb_cpu=self.threads) # here, self.threads was setted as 2 File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/pyranges.py", line 3494, in overlap dfs = pyrange_apply(_overlap, self, other, **kwargs) File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 291, in pyrange_apply df, odf = make_binary_sparse(kwargs, df, odf) File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 134, in make_binary_sparse odf = make_sparse(odf) File "/data/home/xutun/miniconda3/envs/myEnv/lib/python3.9/site-packages/pyranges/multithreaded.py", line 113, in make_sparse if "Strand" in df: TypeError: argument of type 'ray._raylet.ObjectRef' is not iterable Actually, I tried different versions of python enviroment and none of them works (with different problems related to ray). — Reply to this email directly, view it on GitHub <#298>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEHURURCMPRPNIALMHJOFJLW4PFMNANCNFSM6AAAAAAV57LTSQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[endrebak]

Ray has changed and there is little to be gained from using multiple CPUs.

I wonder if I should consider removing the GPU argument instead of fixing this bug.