logs removed

Author: Julien Roussel

examples/benchmark.md

@@ -138,10 +138,6 @@ imputer_mice = imputers.ImputerMICE(groups=("station",), estimator=LinearRegress
 imputer_regressor = imputers.ImputerRegressor(groups=("station",), estimator=LinearRegression())
 ```
 
-```python
-imputer_rpca.fit_transform(df_data)
-```
-
 ```python
 generator_holes = missing_patterns.EmpiricalHoleGenerator(n_splits=2, groups=("station",), subset=cols_to_impute, ratio_masked=ratio_masked)
 ```
@@ -160,10 +156,9 @@ imputer_rpca_opti = hyperparameters.optimize(
     max_evals=10,
     dict_spaces=dict_config_opti
 )
-# imputer_rpca_opti.params_optim = hyperparams_opti
 ```
 
-```python
+```python jupyter={"source_hidden": true}
 dict_config_opti2 = {
     "tau/TEMP": ho.hp.uniform("tau/TEMP", low=.5, high=5),
     "tau/PRES": ho.hp.uniform("tau/PRES", low=.5, high=5),
@@ -194,7 +189,7 @@ dict_imputers = {
     "TSOU": imputer_tsou,
     "TSMLE": imputer_tsmle,
     "RPCA": imputer_rpca,
-    # "RPCA_opti": imputer_rpca_opti,
+    "RPCA_opti": imputer_rpca_opti,
     # "RPCA_opti2": imputer_rpca_opti2,
     # "locf": imputer_locf,
     # "nocb": imputer_nocb,
@@ -230,7 +225,7 @@ results = comparison.compare(df_data)
 results
 ```
 
-```python
+```python jupyter={"source_hidden": true}
 df_plot = results.loc["KL_columnwise",'TEMP']
 plt.barh(df_plot.index, df_plot, color=tab10(0))
 plt.title('TEMP')
@@ -251,8 +246,8 @@ plot.multibar(results.loc["mae"], decimals=1)
 plt.ylabel("mae")
 
 fig.add_subplot(2, 1, 2)
-plot.multibar(results.loc["KL_columnwise"], decimals=1)
-plt.ylabel("KL")
+plot.multibar(results.loc["dist_corr_pattern"], decimals=1)
+plt.ylabel("dist_corr_pattern")
 
 plt.savefig("figures/imputations_benchmark_errors.png")
 plt.show()
@@ -300,10 +295,6 @@ for col in cols_to_impute:
 
 ```
 
-```python
-dfs_imputed
-```
-
 ```python
 # plot.plot_imputations(df_station, dfs_imputed_station)
 

qolmat/benchmark/hyperparameters.py

@@ -1,5 +1,4 @@
 import copy
-import logging
 from typing import Any, Callable, Dict, List, Union
 
 import numpy as np
@@ -13,9 +12,6 @@
 
 from qolmat.benchmark.missing_patterns import _HoleGenerator
 
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.DEBUG)
-
 HyperValue = Union[int, float, str]
 
 
@@ -39,7 +35,6 @@ def fun_obf(args: List[HyperValue]) -> float:
             df_origin = df.copy()
             df_corrupted = df_origin.copy()
             df_corrupted[df_mask] = np.nan
-
             df_imputed = imputer.fit_transform(df_corrupted)
             subset = generator.subset
             fun_metric = metrics.get_metric(metric)

qolmat/benchmark/missing_patterns.py

@@ -1,8 +1,8 @@
 from __future__ import annotations
 
 import functools
-import logging
 from typing import Callable, List, Optional, Tuple, Union
+import warnings
 
 import numpy as np
 import pandas as pd
@@ -11,8 +11,6 @@
 
 from qolmat.utils.exceptions import NoMissingValue, SubsetIsAString
 
-logger = logging.getLogger(__name__)
-
 
 def compute_transition_counts_matrix(states: pd.Series):
     if isinstance(states.iloc[0], tuple):
@@ -305,7 +303,7 @@ def generate_mask(self, X: pd.DataFrame) -> pd.DataFrame:
                     break
 
         if list_failed:
-            logger.warning(f"No place to introduce sampled holes of size {list_failed}!")
+            warnings.warn(f"No place to introduce sampled holes of size {list_failed}!")
         return mask
 
 

qolmat/imputations/em_sampler.py

@@ -1,4 +1,3 @@
-import logging
 from typing import Dict, List, Literal, Optional, Union
 from warnings import WarningMessage
 
@@ -12,8 +11,6 @@
 
 from qolmat.utils import utils
 
-logger = logging.getLogger(__name__)
-
 
 def _gradient_conjugue(A: NDArray, X: NDArray, mask_na: NDArray) -> NDArray:
     """