pre-commit: replace flake8 by ruff, rerun on all python

.pre-commit-config.yaml

@@ -4,20 +4,21 @@ default_language_version:
 exclude: '^tools/(build_utils/fypp)'
 fail_fast: false
 repos:
+- repo: https://github.com/charliermarsh/ruff-pre-commit
+  rev: 'v0.0.261'
+  hooks:
+  - id: ruff
+    args: [ --fix, --exit-non-zero-on-fix ]
+    exclude: >-
+      (?x)^(
+        .cp2k/.*|
+      )$
 - repo: https://github.com/ambv/black
   rev: 22.6.0
   hooks:
   - id: black
     name: Reformat Python files with the black code formatter
     files: '^.*(/PACKAGE)|(\.py)$'
-- repo: https://github.com/pycqa/flake8
-  rev: 5.0.4
-  hooks:
-  - id: flake8
-    exclude: >-
-      (?x)^(
-        .cp2k/.*|
-      )$
 - repo: https://github.com/pre-commit/pre-commit-hooks
   rev: v4.3.0
   hooks:

.ruff.toml

@@ -0,0 +1,3 @@
+select = ["E", "F", "B"]
+line-length = 128
+ignore = ["B905"]

src/acc/libsmm_acc/kernels/smm_acc.py

@@ -83,7 +83,7 @@ def compatible_mnk(algo, m, n, k):
             compatible = False
     else:
         if algo != "medium":
-            assert False, f"Cannot identify algorithm:{str(algo)}"
+            raise AssertionError(f"Cannot identify algorithm:{str(algo)}")
 
     return compatible
 

src/acc/libsmm_acc/kernels/smm_acc_predict.py

@@ -247,14 +247,14 @@ def get_baseline_performances_per_mnk(data, algorithm, gpu, autotuning):
                 & (data.v == baseline_pars["v"])
             ].index.tolist()
         else:
-            assert False, f"Cannot recognize algorithm: {algorithm}"
+            raise AssertionError(f"Cannot recognize algorithm: {algorithm}")
 
         if len(idx_baseline) == 1:
             idx_baseline = idx_baseline[0]
             baseline_perf[mnk] = data["perf (Gflop/s)"][idx_baseline]
         elif len(idx_baseline) > 1:
-            assert False, "Found more than one corresponding index: " + str(
-                idx_baseline
+            raise AssertionError(
+                "Found more than one corresponding index: " + str(idx_baseline)
             )
         else:
             pass  # if none were found, they're in another data chunk. Do nothing.

src/acc/libsmm_acc/notebooks/nb_helper.py

@@ -57,7 +57,7 @@ def get_folders_to_read(to_read, autotuning_data_path):
             if to_read.match(f) is not None
         ]
     else:
-        assert False, "Cannot recognize option: " + to_read
+        raise AssertionError("Cannot recognize option: " + to_read)
 
     num_folders_to_read = len(folders_to_read)
     assert num_folders_to_read > 0

src/acc/libsmm_acc/predict/predict_train.py

@@ -582,7 +582,7 @@ def get_hyperparameter_grid(algo, model_name, n_features):
             "n_estimators": list(n_estimators),
         }
     else:
-        assert False, "Cannot recognize model: " + model_name
+        raise AssertionError("Cannot recognize model: " + model_name)
 
     return param_grid
 
@@ -700,7 +700,9 @@ def get_model(model_to_train, algo, njobs, ntrees):
     elif model_to_train == "xgb-RF":
         model, model_name = get_xgb_RandomForest_model(algo, njobs, ntrees)
     else:
-        assert False, "Cannot recognize model: " + model_to_train + ". Options: DT, RF"
+        raise AssertionError(
+            "Cannot recognize model: " + model_to_train + ". Options: DT, RF"
+        )
     return model, model_name
 
 
@@ -849,7 +851,7 @@ def train_model(X, X_mnk, Y, algo, model_options, folder, log):
         # Hyperparameter Optimization
         param_grid = get_hyperparameter_grid(algo, model_name, n_features)
         if param_grid is None:
-            assert False, "param_grid object is None. Please implement!"
+            raise AssertionError("param_grid object is None. Please implement!")
 
         # At this point, we "cheat"/"take a shortcut" in 2 ways:
         # - we split into train/test partitions using the simple default splitter, not one that is aware of mnk-groups

src/acc/libsmm_acc/predict/prepare_training_data.py

@@ -113,10 +113,11 @@ def get_performance_closest_to_baseline(
             if len(idx_baseline) > 0:
                 break
         else:
-            assert False, (
-                'Could not find closest baseline for mnk=({}x{}x{}) and for algorithm "{}".'
-                "\nLast baseline parameters searched:\n{}"
-                "\nParameter sets searched:\n".format(m, n, k, algorithm, baseline_pars)
+            raise AssertionError(
+                (
+                    'Could not find closest baseline for mnk=({}x{}x{}) and for algorithm "{}".\n'
+                    "Last baseline parameters searched:\n{}\nParameter sets searched:\n"
+                ).format(m, n, k, algorithm, baseline_pars)
             )
 
     idx_baseline = idx_baseline[0]
@@ -142,7 +143,7 @@ def process_chunk(data_chunk, algorithm, gpu_properties, autotuning_properties):
     # For each (mnk), ...
     baseline_performances = dict()
     max_performances = dict()
-    for i, mnk in enumerate(mnks):
+    for _i, mnk in enumerate(mnks):
 
         data_mnk = data_chunk[data_chunk["mnk"] == mnk]
         m, n, k = mnk_pattern.match(mnk).groups()
@@ -433,7 +434,7 @@ def plot_baseline(baseline_perfs_by_algo, data_path, algorithms):
 
     # Get all mnks
     mnk_sequences = list()
-    for algo, baseline_dic in baseline_perfs_by_algo.items():
+    for _algo, baseline_dic in baseline_perfs_by_algo.items():
         mnk_sequences += list(baseline_dic.keys())
     all_mnks = list(set.union(set(mnk_sequences)))
 
@@ -451,9 +452,9 @@ def plot_baseline(baseline_perfs_by_algo, data_path, algorithms):
                 baseline_perfs[mnk] = baseline_perfs_by_algo[algo][mnk]
                 break
         else:
-            assert (
-                False
-            ), "NOOOO this is actually impossible by def of all_mnks, isn't it?"
+            raise AssertionError(
+                "NOOOO this is actually impossible by def of all_mnks, isn't it?"
+            )
 
     # Sort
     mnks = list()

src/acc/opencl/smm/tune_multiply.py

@@ -741,7 +741,7 @@ def handle_sigint(self, signum, frame):
     )
     args = argparser.parse_args()
     # OPENCL_LIBSMM_SMM_xx=tune|enabled|on must be given to permit tuning)
-    if not os.getenv("OPENCL_LIBSMM_SMM_WS") in {"tune", "enabled", "on"}:
+    if os.getenv("OPENCL_LIBSMM_SMM_WS") not in {"tune", "enabled", "on"}:
         os.environ["OPENCL_LIBSMM_SMM_WS"] = "{}".format(args.ws)
     # if not os.getenv("OPENCL_LIBSMM_SMM_AL") in {"tune", "enabled", "on"}:
     # os.environ["OPENCL_LIBSMM_SMM_AL"] = "{}".format(args.al)