search

scottgroup / snoGloBe

snoGloBe predicts both canonical and noncanonical box C/D snoRNA-RNA interactions using a gradient boosting classifier
GNU General Public License v3.0
3 stars 1 forks source link

scikit-learn installation #1

Closed vetmohit89 closed 1 year ago

vetmohit89 commented 1 year ago

Hello,

I have installed and tried with both version of scikit-learn but it giving me error. I am trying example data now. But it is not working (snoGlobe) [mbansal@login004 example]$ snoglobe.py snoRNA.fa target.txt target.gtf /data/user/mbansal/AG_DKC1_rawdata/small_rna_seq/snoGlobe/snoGloBe/example/chromo_dir/ output.txt Traceback (most recent call last): File "./snoGloBe/bin/snoglobe.py", line 428, in prep_snoglobe() File "./snoGloBe/bin/snoglobe.py", line 382, in prep_snoglobe verbose=verbose, calc_intron=True) File "./snoGloBe/bin/snoglobe.py", line 222, in run_snoglobe make_pred(step, sno_file, target_file, chunksize, nb_threads, outfile, threshold) File "./snoGloBe/bin/predict_mp.py", line 77, in make_pred mdl = pickle.load(open(mdl_path, 'rb')) ModuleNotFoundError: No module named 'sklearn.ensemble.gradient_boosting' this is the error when scikit-learn 1.0.2

When I installed lower version I got following error: ModuleNotFoundError: No module named 'sklearn.ensemble.gradient_boosting'

SinhaI commented 3 months ago

Hi, I am getting ImportError: cannot import name 'logsumexp' error when trying to run snoGloBe. Initially, snoGloBe runs fine and creates a temp directory. My python version is 3.6.15, and scikit-learn Version is 0.24.2. I tried to update python and scikit-learn and encountered another error: Python 3.12.1 scikit-learn Version: 1.5.1 ModuleNotFoundError: No module named 'sklearn.ensemble.gradient_boosting' Any suggestions or comments are highly appreciated.

Thanks.

SinhaI commented 2 months ago

Hi again,

snoGloBe is running fine for me now using the following versions: Python: 3.6.15 scikit-learn: 0.21.3 scipy : 1.1.0

Thanks.