Add verbose argument to fklearn validator (#155)

* Add verbose argument to fklearn validator

* Fix lint

* Update CHANGELOG

* Fix lint

* Fix tests

* Bump version

CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## [1.22.0] -  2020-11-18
+- **Enhancement**
+    - Add verbose method to `validator` and `parallel_validator`
+
 ## [1.21.0] - 2020-10-02
 - **Enhancement**
     - Now transformers can create a new column instead of replace the input

src/fklearn/resources/VERSION

@@ -1 +1 @@
-1.21.0
+1.22.0

src/fklearn/validation/validator.py

@@ -12,6 +12,7 @@
 
 from fklearn.types import EvalFnType, LearnerFnType, LogType
 from fklearn.types import SplitterFnType, ValidatorReturnType, PerturbFnType
+from tqdm import tqdm
 
 
 def validator_iteration(data: pd.DataFrame,
@@ -20,7 +21,8 @@ def validator_iteration(data: pd.DataFrame,
                         fold_num: int,
                         train_fn: LearnerFnType,
                         eval_fn: EvalFnType,
-                        predict_oof: bool = False) -> LogType:
+                        predict_oof: bool = False,
+                        verbose: bool = False) -> LogType:
     """
     Perform an iteration of train test split, training and evaluation.
 
@@ -65,7 +67,10 @@ def validator_iteration(data: pd.DataFrame,
 
     eval_results = []
     oof_predictions = []
-    for test_index in test_indexes:
+
+    if verbose:
+        print(f"Running validation for {fold_num} fold.")
+    for test_index in (tqdm(test_indexes) if verbose else test_indexes):
         test_predictions = predict_fn(data.iloc[test_index])
         eval_results.append(eval_fn(test_predictions))
         if predict_oof:
@@ -85,7 +90,8 @@ def validator(train_data: pd.DataFrame,
               eval_fn: EvalFnType,
               perturb_fn_train: PerturbFnType = identity,
               perturb_fn_test: PerturbFnType = identity,
-              predict_oof: bool = False) -> ValidatorReturnType:
+              predict_oof: bool = False,
+              verbose: bool = False) -> ValidatorReturnType:
     """
     Splits the training data into folds given by the split function and
     performs a train-evaluation sequence on each fold by calling
@@ -122,6 +128,9 @@ def validator(train_data: pd.DataFrame,
     predict_oof : bool
         Whether to return out of fold predictions on the logs
 
+    verbose: bool
+        Whether to show more information about the cross validation or not
+
     Returns
     ----------
     A list of log-like dictionary evaluations.
@@ -134,7 +143,8 @@ def validator(train_data: pd.DataFrame,
 
     def fold_iter(fold: Tuple[int, Tuple[pd.Index, pd.Index]]) -> LogType:
         (fold_num, (train_index, test_indexes)) = fold
-        return validator_iteration(train_data, train_index, test_indexes, fold_num, train_fn, eval_fn, predict_oof)
+        return validator_iteration(train_data, train_index, test_indexes, fold_num,
+                                   train_fn, eval_fn, predict_oof, verbose)
 
     zipped_logs = pipe(folds,
                        enumerate,
@@ -168,11 +178,12 @@ def parallel_validator_iteration(train_data: pd.DataFrame,
                                  fold: Tuple[int, Tuple[pd.Index, pd.Index]],
                                  train_fn: LearnerFnType,
                                  eval_fn: EvalFnType,
-                                 predict_oof: bool) -> LogType:
+                                 predict_oof: bool,
+                                 verbose: bool = False) -> LogType:
     (fold_num, (train_index, test_indexes)) = fold
     train_fn = cloudpickle.loads(train_fn)
     eval_fn = cloudpickle.loads(eval_fn)
-    return validator_iteration(train_data, train_index, test_indexes, fold_num, train_fn, eval_fn, predict_oof)
+    return validator_iteration(train_data, train_index, test_indexes, fold_num, train_fn, eval_fn, predict_oof, verbose)
 
 
 @curry
@@ -181,7 +192,8 @@ def parallel_validator(train_data: pd.DataFrame,
                        train_fn: LearnerFnType,
                        eval_fn: EvalFnType,
                        n_jobs: int = 1,
-                       predict_oof: bool = False) -> ValidatorReturnType:
+                       predict_oof: bool = False,
+                       verbose: bool = False) -> ValidatorReturnType:
     """
     Splits the training data into folds given by the split function and
     performs a train-evaluation sequence on each fold. Tries to run each
@@ -213,6 +225,9 @@ def parallel_validator(train_data: pd.DataFrame,
     predict_oof : bool
         Whether to return out of fold predictions on the logs
 
+    verbose: bool
+        Whether to show more information about the cross validation or not
+
     Returns
     ----------
     A list log-like dictionary evaluations.
@@ -223,7 +238,7 @@ def parallel_validator(train_data: pd.DataFrame,
     dumped_eval_fn = cloudpickle.dumps(eval_fn)
 
     result = Parallel(n_jobs=n_jobs, backend="threading")(
-        delayed(parallel_validator_iteration)(train_data, x, dumped_train_fn, dumped_eval_fn, predict_oof)
+        delayed(parallel_validator_iteration)(train_data, x, dumped_train_fn, dumped_eval_fn, predict_oof, verbose)
         for x in enumerate(folds))
     gc.collect()
 

tests/training/test_transformation.py

@@ -56,17 +56,17 @@ def test_capper():
 
     expected2 = pd.DataFrame({"feat1": [7, 9], "feat2": [75, None]})
 
-    pred_fn1, data1, log = capper(input_df, ["feat1", "feat2"], {"feat1": 9.0})
+    pred_fn1, data1, log = capper(input_df, ["feat1", "feat2"], {"feat1": 9})
     pred_fn2, data2, log = capper(
-        input_df, ["feat1", "feat2"], {"feat1": 9.0}, suffix="_suffix"
+        input_df, ["feat1", "feat2"], {"feat1": 9}, suffix="_suffix"
     )
     pred_fn3, data3, log = capper(
-        input_df, ["feat1", "feat2"], {"feat1": 9.0}, prefix="prefix_"
+        input_df, ["feat1", "feat2"], {"feat1": 9}, prefix="prefix_"
     )
     pred_fn4, data4, log = capper(
         input_df,
         ["feat1", "feat2"],
-        {"feat1": 9.0},
+        {"feat1": 9},
         columns_mapping={"feat1": "feat1_raw", "feat2": "feat2_raw"},
     )