Adress review

Co-authored-by: Olivier Grisel <[email protected]>
soda-inria · Jan 12, 2024 · 6bb0aa5 · 6bb0aa5
1 parent 0ce2013
commit 6bb0aa5
Show file tree

Hide file tree

Showing 3 changed files with 18 additions and 13 deletions.
diff --git a/benchmarks/ridge/consolidate_result_csv.py b/benchmarks/ridge/consolidate_result_csv.py
@@ -36,7 +36,6 @@
 COMPUTE_DEVICE = "Compute device"
 COMPUTE_RUNTIME = "Compute runtime"
 DATA_RANDOM_STATE = "Data random state"
-DATA_SAMPLE_WEIGHTS = "Data sample weights"
 DTYPE = "Dtype"
 NB_DATA_FEATURES = "Nb data features"
 NB_DATA_SAMPLES = "Nb data samples"
@@ -66,7 +65,6 @@
     NB_DATA_SAMPLES,
     NB_DATA_FEATURES,
     NB_DATA_TARGETS,
-    DATA_SAMPLE_WEIGHTS,
     WALLTIME,
     BACKEND_PROVIDER,
     COMPUTE_DEVICE,
@@ -94,7 +92,6 @@
     NB_DATA_SAMPLES: np.int64,
     NB_DATA_FEATURES: np.int64,
     NB_DATA_TARGETS: np.int64,
-    DATA_SAMPLE_WEIGHTS: str,
     WALLTIME: np.float64,
     BACKEND_PROVIDER: str,
     COMPUTE_DEVICE: str,
@@ -116,7 +113,7 @@
     COMMENT: str,
 }
 
-COLUMNS_WITH_NONE_STRING = [DATA_SAMPLE_WEIGHTS]
+COLUMNS_WITH_NONE_STRING = []
 
 # If all those fields have equal values for two given benchmarks, then the oldest
 # benchmark (given by RUN_DATE) will be discarded
@@ -126,7 +123,6 @@
     NB_DATA_SAMPLES,
     NB_DATA_FEATURES,
     NB_DATA_TARGETS,
-    DATA_SAMPLE_WEIGHTS,
     BACKEND_PROVIDER,
     SOLVER,
     COMPUTE_DEVICE,
@@ -146,7 +142,6 @@
     (NB_DATA_SAMPLES, False),
     (NB_DATA_FEATURES, False),
     (NB_DATA_TARGETS, False),
-    (DATA_SAMPLE_WEIGHTS, True),
     (WALLTIME, True),
     (BACKEND_PROVIDER, True),
     (COMPUTE_DEVICE, True),
@@ -180,7 +175,6 @@
     objective_dataset_param_dtype=DTYPE,
     objective_dataset_param_random_state=DATA_RANDOM_STATE,
     objective_objective_param_random_state=SOLVER_RANDOM_STATE,
-    objective_objective_param_sample_weight=DATA_SAMPLE_WEIGHTS,
     objective_objective_param_solver=SOLVER,
     objective_solver_param___name=BACKEND_PROVIDER,
     objective_solver_param_device=COMPUTE_DEVICE,

diff --git a/benchmarks/ridge/datasets/simulated_blobs.py b/benchmarks/ridge/datasets/simulated_blobs.py
@@ -10,13 +10,12 @@ class Dataset(BaseDataset):
 
     parameters = {
         "n_samples, n_features": [
-            (20_000_000, 100),
+            (10_000_000, 100),
             (5000, 5000),
-            (20_000_000, 10),
+            (10_000_000, 10),
             (2_000_000, 100),
-            (1500, 1500),
         ],
-        "n_targets": [1, 5, 20],
+        "n_targets": [1, 10],
         "dtype": ["float32"],
         "random_state": [123],
     }

diff --git a/benchmarks/ridge/objective.py b/benchmarks/ridge/objective.py
@@ -21,7 +21,7 @@ class Objective(BaseObjective):
         "alpha": [1.0],
         "fit_intercept": [True],
         "solver, max_iter, tol": [("svd", None, 0)],
-        "sample_weight": ["None", "random"],
+        "sample_weight": ["None"],  # NB: add "random" to test non None weights
         "random_state": [123],
     }
 
@@ -86,7 +86,19 @@ def evaluate_result(self, weights, intercept, **solver_parameters):
         )
 
     def get_one_result(self):
-        return dict(objective=1)
+        n_features = self.dataset_parameters["n_features"]
+        n_targets = self.dataset_parameters["n_targets"]
+        if n_targets == 1:
+            weights = np.ones((n_features,))
+        else:
+            weights = np.ones(
+                (
+                    n_targets,
+                    n_features,
+                )
+            )
+
+        return dict(weights=weights, intercept=np.ones((n_targets,)))
 
     def get_objective(self):
         # Copy the data before sending to the solver, to ensure that no unfortunate