refactor(logs): revert to standard python logs

auton_survival/__init__.py

@@ -473,46 +473,5 @@ class is a composite transform that does both Imputing ***and*** Scaling with
 from .models.cph import DeepCoxPH, DeepRecurrentCoxPH
 from .models.cmhe import DeepCoxMixturesHeterogenousEffects
 
-from loguru import logger
 import warnings
 from tqdm.auto import tqdm
-
-
-logger.disable("auton_survival")
-
-
-def enable_auton_logger(
-    add_logger=False, capture_warnings=False, log_level="INFO"
-):
-    """
-    Enable auton_survival logs
-    """
-    logger.enable("auton_survival")
-    if add_logger:
-        _add_auton_logger(log_level)
-    if capture_warnings:
-        _enable_warnings_capture()
-
-
-def _add_auton_logger(log_level):
-    """
-    Enable auton_survival logs and add a minimal tqdm-friendly loguru logger
-    """
-    logger.remove()
-
-    logger.add(
-        lambda msg: tqdm.write(msg, end=""), colorize=True, level=log_level
-    )
-
-
-def _enable_warnings_capture():
-    """
-    Defer warnings to loguru
-    """
-    showwarning_ = warnings.showwarning
-
-    def showwarning(message, *args, **kwargs):
-        logger.warning(message)
-        showwarning_(message, *args, **kwargs)
-
-    warnings.showwarning = showwarning

auton_survival/experiments.py

@@ -35,7 +35,9 @@
 from sklearn.utils import shuffle
 
 from tqdm import tqdm
-from loguru import logger
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 class SurvivalRegressionCV:
@@ -160,11 +162,11 @@ def fit(self, features, outcomes, horizons, metric="ibs"):
 
         hyper_param_scores = []
         for i, hyper_param in enumerate(self.hyperparam_grid):
-            logger.info("At hyper-param: {}", hyper_param)
+            logger.info(f"At hyper-param: {hyper_param}")
 
             fold_scores = []
             for fold in set(self.folds):
-                logger.info("At fold: {}", fold)
+                logger.info(f"At fold: {fold}")
                 model = SurvivalModel(
                     self.model, random_seed=self.random_seed, **hyper_param
                 )

auton_survival/metrics.py

@@ -32,9 +32,11 @@
 from scipy.optimize import fsolve
 from sklearn.metrics import auc
 from tqdm import tqdm
-from loguru import logger
+import logging
 import warnings
 
+logger = logging.getLogger(__name__)
+
 
 def treatment_effect(
     metric,
@@ -143,8 +145,7 @@ def treatment_effect(
 
     if isinstance(n_bootstrap, int):
         logger.info(
-            "Bootstrapping... {} number of times. This may take a while. Please be Patient...",
-            n_bootstrap,
+            f"Bootstrapping... {n_bootstrap} number of times. This may take a while. Please be Patient...",
         )
 
     is_treated = treatment_indicator.astype(float)
@@ -528,7 +529,7 @@ def _restricted_mean_diff(
     control_weights,
     size_bootstrap=1.0,
     random_seed=None,
-    **kwargs
+    **kwargs,
 ):
     """Compute the difference in the area under the Kaplan Meier curve
     (mean survival time) between control and treatment groups.
@@ -602,7 +603,7 @@ def _survival_at_diff(
     interpolate=True,
     size_bootstrap=1.0,
     random_seed=None,
-    **kwargs
+    **kwargs,
 ):
     """Compute the difference in Kaplan Meier survival function estimates
     between the control and treatment groups at a specified time horizon.
@@ -680,7 +681,7 @@ def _tar(
     interpolate=True,
     size_bootstrap=1.0,
     random_seed=None,
-    **kwargs
+    **kwargs,
 ):
     """Time at Risk (TaR) measures time-to-event at a specified level
     of risk.
@@ -749,7 +750,7 @@ def _hazard_ratio(
     control_weights,
     size_bootstrap=1.0,
     random_seed=None,
-    **kwargs
+    **kwargs,
 ):
     """Train an instance of the Cox Proportional Hazards model and return the
     exp(coefficients) (hazard ratios) of the model.

auton_survival/models/cmhe/__init__.py

@@ -75,7 +75,7 @@
 
 """
 
-from loguru import logger
+import logging
 import numpy as np
 import torch
 
@@ -85,6 +85,8 @@
 
 from auton_survival.utils import _dataframe_to_array
 
+logger = logging.getLogger(__name__)
+
 
 class DeepCoxMixturesHeterogenousEffects:
     """A Deep Cox Mixtures with Heterogenous Effects model.
@@ -149,7 +151,7 @@ def __call__(self):
         else:
             logger.info("An unfitted instance of the CMHE model")
 
-        logger.info("Hidden Layers: {}", self.layers)
+        logger.info(f"Hidden Layers: {self.layers}")
 
     def _preprocess_test_data(self, x, a=None):
         x = _dataframe_to_array(x)

auton_survival/models/cmhe/cmhe_utilities.py

@@ -21,7 +21,7 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 
-from loguru import logger
+import logging
 import torch
 import numpy as np
 
@@ -40,6 +40,8 @@
     smooth_bl_survival,
 )
 
+logger = logging.getLogger(__name__)
+
 
 def get_likelihood(model, breslow_splines, x, t, e, a):
     # Function requires numpy/torch
@@ -300,9 +302,7 @@ def train_cmhe(
 
         losses.append(valcn)
 
-        logger.debug(
-            "Patience: {} | Epoch: {} | Loss: {}", patience_, epoch, valcn
-        )
+        logger.debug(f"Patience: {patience_} | Epoch: {epoch} | Loss: {valcn}")
 
         if valcn > valc:
             patience_ += 1

auton_survival/models/cph/__init__.py

@@ -24,7 +24,7 @@
 r""" Deep Cox Proportional Hazards Model"""
 
 from collections import namedtuple
-from loguru import logger
+import logging
 import torch
 import numpy as np
 import random
@@ -37,6 +37,9 @@
 from auton_survival.models.utils.recurrent_nn_utils import _get_padded_targets
 
 
+logger = logging.getLogger(__name__)
+
+
 DcphModel = namedtuple("DcphModel", ["module", "breslow"])
 
 
@@ -127,7 +130,7 @@ def __call__(self):
         else:
             logger.info("An unfitted instance of the Deep Cox PH model")
 
-        logger.info("Hidden Layers: {}", self.layers)
+        logger.info(f"Hidden Layers: {self.layers}")
 
     def _preprocess_test_data(self, x):
         x = _dataframe_to_array(x)
@@ -209,7 +212,7 @@ def init_torch_model(
             self.initialized = True
         else:
             logger.info(
-                f"""Early initialization selected. Model-specific `fit` parameters will be ignored."""
+                "Early initialization selected. Model-specific `fit` parameters will be ignored."
             )
 
     def fit(
@@ -324,7 +327,7 @@ def predict_time_independent_survival(
     @torch.inference_mode()
     def predict_time_independent_risk(self, x: torch.Tensor) -> torch.Tensor:
         if not self.initialized:
-            logger.warning(
+            raise Exception(
                 "The PyTorch module has not been initialized yet. Please init the "
                 + "model using the `init_torch_model` method on some training data "
                 + "before calling `predict_time_independent_risk`."
@@ -418,7 +421,7 @@ def __call__(self):
                 "An unfitted instance of the Recurrent Deep Cox PH model"
             )
 
-        logger.info("Hidden Layers: {}", self.layers)
+        logger.info(f"Hidden Layers: {self.layers}")
 
     def _gen_torch_model(self, inputdim, optimizer):
         """Helper function to return a torch model."""

auton_survival/models/cph/dcph_utilities.py

@@ -1,7 +1,7 @@
 import torch
 import numpy as np
 import pandas as pd
-from loguru import logger
+import logging
 
 from sksurv.linear_model.coxph import BreslowEstimator
 
@@ -19,6 +19,8 @@
     partial_ll_loss,
 )
 
+logger = logging.getLogger(__name__)
+
 
 def fit_breslow(model, x, t, e):
     return BreslowEstimator().fit(
@@ -119,9 +121,7 @@ def train_dcph(
 
         dics.append(deepcopy(model.state_dict()))
 
-        logger.debug(
-            "Patience: {} | Epoch: {} | Loss: {}", patience_, epoch, valcn
-        )
+        logger.debug(f"Patience: {patience_} | Epoch: {epoch} | Loss: {valcn}")
 
         if valcn > valc:
             patience_ += 1

auton_survival/models/dcm/__init__.py

@@ -50,7 +50,7 @@
 
 """
 
-from loguru import logger
+import logging
 import torch
 import numpy as np
 
@@ -59,6 +59,8 @@
 
 from auton_survival.utils import _dataframe_to_array
 
+logger = logging.getLogger(__name__)
+
 
 class DeepCoxMixtures:
     """A Deep Cox Mixture model.
@@ -116,8 +118,8 @@ def __call__(self):
         else:
             logger.info("An unfitted instance of the Deep Cox Mixtures model")
 
-        logger.info("Number of underlying cox distributions (k): {}", self.k)
-        logger.info("Hidden Layers: {}", self.layers)
+        logger.info(f"Number of underlying cox distributions (k): {self.k}")
+        logger.info(f"Hidden Layers: {self.layers}")
 
     def _preprocess_test_data(self, x):
         x = _dataframe_to_array(x)

auton_survival/models/dcm/dcm_utilities.py

@@ -1,5 +1,5 @@
 import warnings
-from loguru import logger
+import logging
 from matplotlib.pyplot import get
 
 import torch
@@ -24,6 +24,8 @@
     smooth_bl_survival,
 )
 
+logger = logging.getLogger(__name__)
+
 
 def get_likelihood(model, breslow_splines, x, t, e):
     # Function requires numpy/torch
@@ -263,9 +265,7 @@ def train_dcm(
 
         losses.append(valcn)
 
-        logger.debug(
-            "Patience: {} | Epoch: {} | Loss: {}", patience_, epoch, valcn
-        )
+        logger.debug(f"Patience: {patience_} | Epoch: {epoch} | Loss: {valcn}")
 
         if valcn > valc:
             patience_ += 1

auton_survival/models/dsm/__init__.py

@@ -156,7 +156,7 @@
 
 """
 
-from loguru import logger
+import logging
 import torch
 import numpy as np
 
@@ -183,6 +183,9 @@
 __pdoc__["DSMBase"] = False
 
 
+logger = logging.getLogger(__name__)
+
+
 class DSMBase:
     """Base Class for all DSM models"""
 
@@ -547,9 +550,9 @@ def __call__(self):
                 "An unfitted instance of the Deep Survival Machines model"
             )
 
-        logger.info("Number of underlying distributions (k): {}", self.k)
-        logger.info("Hidden Layers: {}", self.layers)
-        logger.info("Distribution Choice: {}", self.dist)
+        logger.info(f"Number of underlying distributions (k): {self.k}")
+        logger.info(f"Hidden Layers: {self.layers}")
+        logger.info(f"Distribution Choice: {self.dist}")
 
 
 class DeepRecurrentSurvivalMachines(DSMBase):

auton_survival/models/dsm/dsm_utilities.py

@@ -37,7 +37,10 @@
 import numpy as np
 
 import gc
-from loguru import logger
+import logging
+
+
+logger = logging.getLogger(__name__)
 
 
 def pretrain_dsm(