Merge remote-tracking branch 'origin/master' into 3313-enable-auto-ea…

…rly-stopping

.ci/test.sh

@@ -146,8 +146,8 @@ if [[ $TASK == "check-docs" ]] || [[ $TASK == "check-links" ]]; then
     make -C docs html || exit 1
     if [[ $TASK == "check-links" ]]; then
         # check docs for broken links
-        pip install linkchecker
-        linkchecker --config=.linkcheckerrc ./docs/_build/html/*.html || exit 1
+        pip install 'linkchecker>=10.5.0'
+        linkchecker --config=./docs/.linkcheckerrc ./docs/_build/html/*.html || exit 1
         exit 0
     fi
     # check the consistency of parameters' descriptions and other stuff

.pre-commit-config.yaml

@@ -17,15 +17,9 @@ repos:
     hooks:
       - id: end-of-file-fixer
       - id: trailing-whitespace
-  - repo: https://github.com/pycqa/isort
-    rev: 5.13.2
-    hooks:
-      - id: isort
-        name: isort (python)
-        args: ["--settings-path", "python-package/pyproject.toml"]
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.7.0
+    rev: v0.8.3
     hooks:
       # Run the linter.
       - id: ruff
@@ -40,7 +34,7 @@ repos:
     hooks:
       - id: shellcheck
   - repo: https://github.com/crate-ci/typos
-    rev: v1.23.2
+    rev: v1.28.3
     hooks:
       - id: typos
         args: ["--force-exclude"]

.vsts-ci.yml

@@ -69,15 +69,17 @@ jobs:
       # check disk usage
       print-diagnostics
       # remove old containers, container images, volumes
-      # ref: https://stackoverflow.com/a/32723127/3986677)
+      # ref: https://stackoverflow.com/a/32723127/3986677
+      # ref: https://depot.dev/blog/docker-clear-cache#removing-everything-with-docker-system-prune
       echo "---- running 'docker system prune' ----"
       /tmp/docker system prune \
         --all \
         --force \
+        --volumes \
         --filter until=720h
       # check disk usage again
       print-diagnostics
-    displayName: clean
+    displayName: Clean
 ###########################################
 - job: Linux
 ###########################################

R-package/tests/testthat/test_basic.R

@@ -2345,7 +2345,7 @@ test_that("early stopping works with lgb.cv()", {
   # never changes, its first iteration was the best oone
   expect_equal(bst$best_iter, 1L)
 
-  # best_score should be taken from the first metri
+  # best_score should be taken from the first metric
   expect_equal(bst$best_score, 0.2)
 
   # early stopping should have happened, since constant_metric was the first

docs/.linkcheckerrc

@@ -1,9 +1,9 @@
 [checking]
-maxrequestspersecond=1
+maxrequestspersecond=0.1
 recursionlevel=1
 anchors=1
 sslverify=0
-threads=1
+threads=4
 
 [filtering]
 ignore=

examples/python-guide/notebooks/interactive_plot_example.ipynb

@@ -30,7 +30,7 @@
     "try:\n",
     "    # To enable interactive mode you should install ipywidgets\n",
     "    # https://github.com/jupyter-widgets/ipywidgets\n",
-    "    from ipywidgets import interact, SelectMultiple\n",
+    "    from ipywidgets import SelectMultiple, interact\n",
     "\n",
     "    INTERACTIVE = True\n",
     "except ImportError:\n",

python-package/lightgbm/basic.py

@@ -2504,13 +2504,13 @@ def _compare_params_for_warning(
         compare_result : bool
           Returns whether two dictionaries with params are equal.
         """
-        for k in other_params:
+        for k, v in other_params.items():
             if k not in ignore_keys:
-                if k not in params or params[k] != other_params[k]:
+                if k not in params or params[k] != v:
                     return False
-        for k in params:
+        for k, v in params.items():
             if k not in ignore_keys:
-                if k not in other_params or params[k] != other_params[k]:
+                if k not in other_params or v != other_params[k]:
                     return False
         return True
 

python-package/lightgbm/callback.py

@@ -73,15 +73,13 @@ class CallbackEnv:
 
 def _format_eval_result(value: _EvalResultTuple, show_stdv: bool) -> str:
     """Format metric string."""
-    if len(value) == 4:
-        return f"{value[0]}'s {value[1]}: {value[2]:g}"
-    elif len(value) == 5:
-        if show_stdv:
-            return f"{value[0]}'s {value[1]}: {value[2]:g} + {value[4]:g}"  # type: ignore[misc]
-        else:
-            return f"{value[0]}'s {value[1]}: {value[2]:g}"
-    else:
-        raise ValueError("Wrong metric value")
+    dataset_name, metric_name, metric_value, *_ = value
+    out = f"{dataset_name}'s {metric_name}: {metric_value:g}"
+    # tuples from cv() sometimes have a 5th item, with standard deviation of
+    # the evaluation metric (taken over all cross-validation folds)
+    if show_stdv and len(value) == 5:
+        out += f" + {value[4]:g}"
+    return out
 
 
 class _LogEvaluationCallback:

python-package/lightgbm/engine.py

@@ -765,7 +765,7 @@ def cv(
     train_set._update_params(params)._set_predictor(predictor)
 
     results = defaultdict(list)
-    cvfolds = _make_n_folds(
+    cvbooster = _make_n_folds(
         full_data=train_set,
         folds=folds,
         nfold=nfold,

python-package/pyproject.toml

@@ -84,17 +84,6 @@ minimum-version = "build-system.requires"
 
 # end:build-system
 
-[tool.isort]
-include_trailing_comma = true
-line_length = 120
-# "vertical hanging indent", to match what ruff-format does
-# ref: https://pycqa.github.io/isort/docs/configuration/multi_line_output_modes.html#3-vertical-hanging-indent
-multi_line_output = 3
-skip_glob = [
-    "*/external_libs/*",
-    "*/lightgbm-python/*",
-]
-
 [tool.mypy]
 disallow_untyped_defs = true
 exclude = 'build/*|compile/*|docs/*|examples/*|external_libs/*|lightgbm-python/*|tests/*'
@@ -140,7 +129,7 @@ ignore = [
     "PLR1714",
     # (pylint) Magic value used in comparison
     "PLR2004",
-    # (pylint) for loop veriable overwritten by assignment target
+    # (pylint) for loop variable overwritten by assignment target
     "PLW2901",
     # (pylint) use 'elif' instead of 'else' then 'if', to reduce indentation
     "PLR5501"
@@ -152,10 +141,12 @@ select = [
     "C4",
     # pydocstyle
     "D",
-    # pycodestyle
+    # pycodestyle (errors)
     "E",
     # pyflakes
     "F",
+    # isort
+    "I",
     # NumPy-specific rules
     "NPY",
     # pylint
@@ -166,11 +157,13 @@ select = [
     "SIM401",
     # flake8-print
     "T",
+    # pycodestyle (warnings)
+    "W",
 ]
 
 [tool.ruff.lint.per-file-ignores]
 "docs/conf.py" = [
-    # (flake8-bugbear) raise exceptions with "raise ... from errr"
+    # (flake8-bugbear) raise exceptions with "raise ... from err"
     "B904",
     # (flake8-print) flake8-print
     "T"
@@ -196,3 +189,6 @@ select = [
 
 [tool.ruff.lint.pydocstyle]
 convention = "numpy"
+
+[tool.ruff.lint.isort]
+known-first-party = ["lightgbm"]

src/objective/rank_objective.hpp

@@ -204,7 +204,7 @@ class LambdarankNDCG : public RankingObjective {
     }
     const double worst_score = score[sorted_idx[worst_idx]];
     double sum_lambdas = 0.0;
-    // start accmulate lambdas by pairs that contain at least one document above truncation level
+    // start accumulate lambdas by pairs that contain at least one document above truncation level
     for (data_size_t i = 0; i < cnt - 1 && i < truncation_level_; ++i) {
       if (score[sorted_idx[i]] == kMinScore) { continue; }
       for (data_size_t j = i + 1; j < cnt; ++j) {

src/treelearner/kernels/histogram_16_64_256.cu

@@ -150,7 +150,7 @@ __global__ void KERNEL_NAME(const uchar* feature_data_base,
     // size of threads that process this feature4
     const unsigned int subglobal_size = lsize * (1 << power_feature_workgroups);
 
-    // equavalent thread ID in this subgroup for this feature4
+    // equivalent thread ID in this subgroup for this feature4
     const unsigned int subglobal_tid  = gtid - feature_id * subglobal_size;