[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Author: pre-commit-ci[bot]

codes/benchmark/bench_fcts.py

@@ -460,7 +460,9 @@ def evaluate_iterative_predictions(
         # We predict steps 1..(chunk_len-1) relative to the provided init state (index 0).
         # Map these to global indices [start+1 .. end] inclusively.
         if i == 0:
-            iterative_preds[:, start : end + 1, :] = preds_chunk[:, : model.n_timesteps, :].detach().cpu().numpy()
+            iterative_preds[:, start : end + 1, :] = (
+                preds_chunk[:, : model.n_timesteps, :].detach().cpu().numpy()
+            )
         iterative_preds[:, start + 1 : end + 1, :] = (
             preds_chunk[:, 1 : model.n_timesteps, :].detach().cpu().numpy()
         )

codes/benchmark/bench_utils.py

@@ -261,7 +261,9 @@ def count_trainable_parameters(model: torch.nn.Module) -> int:
     return sum(p.numel() for p in model.parameters() if p.requires_grad)
 
 
-def measure_memory_footprint(model: torch.nn.Module, inputs: tuple, device: torch.device) -> dict:
+def measure_memory_footprint(
+    model: torch.nn.Module, inputs: tuple, device: torch.device
+) -> dict:
     """
     Measure peak GPU memory usage for forward/backward passes.
 

codes/train/__init__.py

@@ -1,10 +1,10 @@
 from .train_fcts import (
+    DummyLock,
+    create_task_list_for_surrogate,
     parallel_training,
     sequential_training,
     train_and_save_model,
-    create_task_list_for_surrogate,
     worker,
-    DummyLock,
 )
 
 __all__ = [

datasets/_data_analysis/__init__.py

@@ -4,5 +4,5 @@
     "get_data_subset",
     "create_dataset",
     "normalize_data",
-    "download_data"
+    "download_data",
 ]

test/test_training_pipeline.py

@@ -1,20 +1,20 @@
 import threading
 from queue import Queue
-import pytest
 from unittest.mock import Mock, patch
 
+import pytest
+
 # import *from* the module that actually defines them
 from codes.train.train_fcts import (
     DummyLock,
     create_task_list_for_surrogate,
-    train_and_save_model,
-    worker,
     parallel_training,
     sequential_training,
+    train_and_save_model,
+    worker,
 )
 from codes.utils import load_task_list, save_task_list
 
-
 # — fixtures —
 
 

test/test_utils.py

@@ -1,26 +1,27 @@
 import os
-import time
 import random
-import yaml
-import torch
+import time
+
 import numpy as np
 import pytest
+import torch
+import yaml
 
 from codes.utils import (
-    read_yaml_config,
-    time_execution,
+    batch_factor_to_float,
+    check_training_status,
     create_model_dir,
+    determine_batch_size,
     get_progress_bar,
     load_and_save_config,
-    set_random_seeds,
-    nice_print,
+    load_task_list,
     make_description,
-    worker_init_fn,
+    nice_print,
+    read_yaml_config,
     save_task_list,
-    load_task_list,
-    check_training_status,
-    determine_batch_size,
-    batch_factor_to_float,
+    set_random_seeds,
+    time_execution,
+    worker_init_fn,
 )