XgboostError

[peter-WeiZhang]

XGBoostError: [14:36:23] C:\Users\Administrator\workspace\xgboost-win64_release_1.0.0\dmlc-core\src\io\local_filesys.cc:209: Check failed: allow_null: LocalFileSystem::Open "C:\Users\ZW\AppData\Local\Temp\tmp_mu9qwkg": Permission denied

I don't know how to solve this, I have tried to open jupyter notebook with root, but still not work, thanks!

[peter-WeiZhang]

XGBoostError Traceback (most recent call last)
in
1 automl = AutoML() # total train limit = 1 hour
----> 2 automl.fit(X_train, y_train)

d:\anaconda3\envs\mljar\lib\site-packages\supervised\automl.py in fit(self, X_train, y_train, X_validation, y_validation)
1122
1123 except Exception as e:
-> 1124 raise e
1125 finally:
1126 if self._X_train_path is not None:

d:\anaconda3\envs\mljar\lib\site-packages\supervised\automl.py in fit(self, X_train, y_train, X_validation, y_validation)
1052 self._time_start[self._fit_level] = start
1053 for params in tuner.default_params(len(self._models)):
-> 1054 self.train_model(params)
1055 self._time_spend["default_algorithms"] = np.round(time.time() - start, 2)
1056

d:\anaconda3\envs\mljar\lib\site-packages\supervised\automl.py in train_model(self, params)
553 raise AutoMLException(f"Cannot create directory {model_path}")
554
--> 555 mf.train(model_path)
556
557 mf.save(model_path)

d:\anaconda3\envs\mljar\lib\site-packages\supervised\model_framework.py in train(self, model_path)
142 self.callbacks.on_iteration_start()
143
--> 144 learner.fit(X_train, y_train, X_validation, y_validation, log_to_file)
145
146 self.callbacks.on_iteration_end(

d:\anaconda3\envs\mljar\lib\site-packages\supervised\algorithms\xgboost.py in fit(self, X, y, X_validation, y_validation, log_to_file)
111 # dmlc/xgboost#5474
112 with tempfile.NamedTemporaryFile() as tmp:
--> 113 self.model.save_model(tmp.name)
114 del self.model
115 self.model = xgb.Booster()

d:\anaconda3\envs\mljar\lib\site-packages\xgboost\core.py in save_model(self, fname)
1495 if isinstance(fname, (STRING_TYPES, os_PathLike)): # assume file name
1496 _check_call(_LIB.XGBoosterSaveModel(
-> 1497 self.handle, c_str(os_fspath(fname))))
1498 else:
1499 raise TypeError("fname must be a string or os_PathLike")

d:\anaconda3\envs\mljar\lib\site-packages\xgboost\core.py in _check_call(ret)
187 """
188 if ret != 0:
--> 189 raise XGBoostError(py_str(_LIB.XGBGetLastError()))
190
191

XGBoostError: [14:36:23] C:\Users\Administrator\workspace\xgboost-win64_release_1.0.0\dmlc-core\src\io\local_filesys.cc:209: Check failed: allow_null: LocalFileSystem::Open "C:\Users\ZW\AppData\Local\Temp\tmp_mu9qwkg": Permission denied

[pplonski]

When training xgboost Im saving it to temporary file and reloading to reduce RAM consumption. Maybe there is smarter way to reduce the RAM?

The error means that your operating system doesnt allow to create temporary file because of permissions. Have you try to run it as a script?

[peter-WeiZhang]

When training xgboost Im saving it to temporary file and reloading to reduce RAM consumption. Maybe there is smarter way to reduce the RAM?

The error means that your operating system doesnt allow to create temporary file because of permissions. Have you try to run it as a script?

I have tried to download as python script in jupyter and ran it with VS Code, it reports the same error.

[pplonski]

Thank you for checking! I will fix this with #127 issue.

[pplonski]

Fixed in release 0.6.1. Closing the issue.