delphi001 / SAAMBE-3D

GNU General Public License v3.0
8 stars 6 forks source link

XGBoost Issue #4

Closed Izhar-Karbat closed 1 year ago

Izhar-Karbat commented 2 years ago

running the code on python 3.9.10 , with xgboost v1.6.2 results in the following exception :

Exception has occurred: XGBoostError [23:52:59] /tmp/abs_40obctay9q/croots/recipe/xgboost-split_1659548945886/work/dmlc-core/src/io/local_filesys.cc:209: Check failed: allow_null: LocalFileSystem::Open "regression.model": No such file or directory Stack trace: [bt] (0) /home/izhar/anaconda3/envs/MDAnalysis/lib/libxgboost.so(+0x98154) [0x7fa42aeac154] [bt] (1) /home/izhar/anaconda3/envs/MDAnalysis/lib/libxgboost.so(dmlc::io::LocalFileSystem::Open(dmlc::io::URI const&, char const, bool)+0x3cb) [0x7fa42b229a6b] [bt] (2) /home/izhar/anaconda3/envs/MDAnalysis/lib/libxgboost.so(dmlc::Stream::Create(char const, char const*, bool)+0x46) [0x7fa42b1fbe56] [bt] (3) /home/izhar/anaconda3/envs/MDAnalysis/lib/libxgboost.so(XGBoosterLoadModel+0x80c) [0x7fa42aec9f3c] [bt] (4) /home/izhar/anaconda3/envs/MDAnalysis/lib/python3.9/lib-dynload/../../libffi.so.8(+0x6a4a) [0x7fa5de73ea4a] [bt] (5) /home/izhar/anaconda3/envs/MDAnalysis/lib/python3.9/lib-dynload/../../libffi.so.8(+0x5fea) [0x7fa5de73dfea] [bt] (6) /home/izhar/anaconda3/envs/MDAnalysis/lib/python3.9/lib-dynload/_ctypes.cpython-39-x86_64-linux-gnu.so(+0x13dc7) [0x7fa5de757dc7] [bt] (7) /home/izhar/anaconda3/envs/MDAnalysis/lib/python3.9/lib-dynload/_ctypes.cpython-39-x86_64-linux-gnu.so(+0x14454) [0x7fa5de758454] [bt] (8) /home/izhar/anaconda3/envs/MDAnalysis/bin/python(_PyObject_MakeTpCall+0x2c0) [0x5567e8caa070]

File "/mnt/NAS/SAAMBE-3D/Mutation_pred.py", line 480, in pred_feature model=xgb.Booster(model_file='regression.model') File "/mnt/NAS/SAAMBE-3D/Mutation_pred.py", line 475, in file_loop return(pred_feature(label,model_type)) File "/mnt/NAS/SAAMBE-3D/Mutation_pred.py", line 561, in pred=file_loop(pdb_id,info[0],info[1],str(info[2]),info[3],model_type)

Izhar-Karbat commented 2 years ago

Update: Solved by downgrading xgboost to version 1.02.