Merge branch 'lab/precommit' into 'export/v1-2-0'

Add pre-commit hooks for local development

See merge request onecomp/onecomp-lab!75

Author: aki916f

.pre-commit-config.yaml

@@ -0,0 +1,33 @@
+repos:
+  - repo: https://github.com/pycqa/isort
+    rev: 6.0.1
+    hooks:
+      - id: isort
+
+  - repo: https://github.com/psf/black
+    rev: 25.1.0
+    hooks:
+      - id: black
+        args: [--line-length=99]
+
+  - repo: local
+    hooks:
+      - id: no-japanese
+        name: Forbid Japanese characters
+        language: system
+        entry: bash scripts/run_venv_python_for_precommit.sh scripts/check_no_japanese.py
+        exclude: '\.md$|\.gitignore$'
+        types: [text]
+
+      - id: copyright-header
+        name: Check Fujitsu copyright header
+        language: system
+        entry: bash scripts/run_venv_python_for_precommit.sh scripts/check_copyright_header.py
+        types: [python]
+        exclude: '(__init__|__main__)\.py$|^tests/|^benchmark/'
+
+      - id: no-email-address
+        name: Forbid email addresses
+        language: pygrep
+        entry: '[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}'
+        types: [python]

CHANGELOG.md

@@ -2,6 +2,9 @@
 
 ## [v1.2.0] 2026-06-04 (WIP)
 
+### for Developer: pre-commit
+
+- Added `.pre-commit-config.yaml` with `black`, `isort`, and local hooks (`no-japanese`, `copyright-header`, `no-email-address`); install with `uv sync --extra dev` then `pre-commit install` (see README)
 
 ## [v1.1.1] 2026-05-21
 

README.md

@@ -130,7 +130,7 @@ The `--extra cu128` option installs the CUDA-enabled version of PyTorch (along w
 Replace `cu128` with the appropriate variant for your environment: `cpu`, `cu118`, `cu121`, `cu124`, `cu126`, `cu128`, or `cu130`.
 PyTorch will be automatically downloaded by `uv`, so you do not need to install it beforehand.
 
-Adding `--extra dev` installs development tools (black, pytest, pylint).
+Adding `--extra dev` installs development tools (black, pre-commit, pytest, pylint).
 Adding `--extra visualize` installs matplotlib for visualization features.
 Adding `--extra hydra` installs `hydra-core` for the example scripts and `model_validation/` runners that use Hydra-based configuration.
 
@@ -179,6 +179,40 @@ pip install -e ".[dev]"
 
 Replace `cu128` with the appropriate variant for your environment: `cpu`, `cu118`, `cu121`, `cu124`, `cu126`, `cu128`, or `cu130`.
 
+#### Pre-commit
+
+After installing development dependencies (`--extra dev` with uv, or `pip install -e ".[dev]"` with pip), register the Git hooks once:
+
+```bash
+pre-commit install
+```
+
+On every `git commit`, the following checks run automatically:
+
+| Hook | Description |
+|------|-------------|
+| **black** | Code formatting (line length 99) |
+| **isort** | Import sorting |
+| **no-japanese** | Forbid Japanese characters in text files (`.md` and `.gitignore` are excluded) |
+| **copyright-header** | Verify the Fujitsu copyright header in Python files |
+| **no-email-address** | Forbid email addresses in Python files |
+
+Common commands:
+
+```bash
+# Run hooks on staged files only
+pre-commit run
+
+# Run hooks on all files (useful after first install or config changes)
+pre-commit run --all-files
+
+# Run a specific hook
+pre-commit run black --all-files
+
+# With uv (no activation needed)
+uv run pre-commit run --all-files
+```
+
 
 ### Building Documentation Locally
 

benchmark/llama3-8b-gptq/quant_benchmark.py

@@ -15,7 +15,7 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import CalibrationConfig, GPTQ, ModelConfig, Runner
+from onecomp import GPTQ, CalibrationConfig, ModelConfig, Runner
 
 
 def create_quantizers(cfg: DictConfig):
@@ -80,9 +80,7 @@ def main(cfg: DictConfig):
 
     # Save results
     for q in quantizers:
-        runner.save_quantization_statistics(
-            f"quantization_statistics_{q.name}.json", quantizer=q
-        )
+        runner.save_quantization_statistics(f"quantization_statistics_{q.name}.json", quantizer=q)
 
     # Perplexity evaluation
     if cfg.calc_ppl:

benchmark/llama3-8b-jointq/quant_benchmark.py

@@ -15,7 +15,7 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import CalibrationConfig, GPTQ, JointQ, ModelConfig, Runner
+from onecomp import GPTQ, CalibrationConfig, JointQ, ModelConfig, Runner
 
 
 def create_quantizers(cfg: DictConfig):
@@ -93,9 +93,7 @@ def main(cfg: DictConfig):
     runner.run()
 
     for q in quantizers:
-        runner.save_quantization_statistics(
-            f"quantization_statistics_{q.name}.json", quantizer=q
-        )
+        runner.save_quantization_statistics(f"quantization_statistics_{q.name}.json", quantizer=q)
 
     if cfg.calc_ppl:
         runner.benchmark_perplexity(

benchmark/llama3-8b-lpcd-gptq/quant_benchmark.py

@@ -27,12 +27,12 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import GPTQ, ModelConfig, QEPConfig, LPCDConfig, Runner
+from onecomp import GPTQ, LPCDConfig, ModelConfig, QEPConfig, Runner
 
 
 def create_quantizer(cfg: DictConfig, task_id: int):
     """Create a single GPTQ quantizer for the given task_id."""
-    combinations = list(itertools.product(cfg.gptq.bits, ['none', 'res', 'qk', 'vo', 'ud', 'all']))
+    combinations = list(itertools.product(cfg.gptq.bits, ["none", "res", "qk", "vo", "ud", "all"]))
 
     if task_id < 0 or task_id >= len(combinations):
         raise ValueError(
@@ -43,7 +43,7 @@ def create_quantizer(cfg: DictConfig, task_id: int):
     bits, submodule = combinations[task_id]
 
     return GPTQ(
-        wbits=bits, 
+        wbits=bits,
         groupsize=128,
         name=f"lpcd_{submodule}_{bits}bit",
     )
@@ -57,11 +57,7 @@ def main(cfg: DictConfig):
 
     print(f"task_id={cfg.task_id} -> {quantizer.name}")
 
-    model_config = ModelConfig(
-        path=cfg.model_path, 
-        dtype="bfloat16",
-        device=cfg.model_device
-    )
+    model_config = ModelConfig(path=cfg.model_path, dtype="bfloat16", device=cfg.model_device)
 
     qep_config = QEPConfig(
         percdamp=cfg.qep.percdamp,
@@ -103,9 +99,7 @@ def main(cfg: DictConfig):
 
     runner.run()
 
-    runner.save_quantization_statistics(
-        f"quantization_statistics_{quantizer.name}.json"
-    )
+    runner.save_quantization_statistics(f"quantization_statistics_{quantizer.name}.json")
 
     # Perplexity evaluation
     if cfg.calc_ppl:

benchmark/llama3-8b-qep-gptq/quant_benchmark.py

@@ -21,7 +21,7 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import CalibrationConfig, GPTQ, ModelConfig, QEPConfig, Runner
+from onecomp import GPTQ, CalibrationConfig, ModelConfig, QEPConfig, Runner
 
 
 def create_quantizer(cfg: DictConfig, task_id: int):
@@ -86,9 +86,7 @@ def main(cfg: DictConfig):
 
     runner.run()
 
-    runner.save_quantization_statistics(
-        f"quantization_statistics_{quantizer.name}.json"
-    )
+    runner.save_quantization_statistics(f"quantization_statistics_{quantizer.name}.json")
 
     # Perplexity evaluation
     if cfg.calc_ppl:

benchmark/llama3-8b-various/quant_benchmark.py

@@ -12,7 +12,20 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import GPTQ, JointQ, CalibrationConfig, ModelConfig, Runner, ARB, CQ, DBF, QBB, QUIP, RTN, Onebit
+from onecomp import (
+    ARB,
+    CQ,
+    DBF,
+    GPTQ,
+    QBB,
+    QUIP,
+    RTN,
+    CalibrationConfig,
+    JointQ,
+    ModelConfig,
+    Onebit,
+    Runner,
+)
 
 
 def create_quantizers():
@@ -57,9 +70,7 @@ def main(cfg: DictConfig):
     runner.run()
 
     for q in quantizers:
-        runner.save_quantization_statistics(
-            f"quantization_statistics_{q.name}.json", quantizer=q
-        )
+        runner.save_quantization_statistics(f"quantization_statistics_{q.name}.json", quantizer=q)
 
     if cfg.calc_ppl:
         runner.benchmark_perplexity(

benchmark/qwen3-14b-gptq/quant_benchmark.py

@@ -15,7 +15,7 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import CalibrationConfig, GPTQ, ModelConfig, Runner
+from onecomp import GPTQ, CalibrationConfig, ModelConfig, Runner
 
 
 def create_quantizers(cfg: DictConfig):
@@ -75,9 +75,7 @@ def main(cfg: DictConfig):
     runner.run()
 
     for q in quantizers:
-        runner.save_quantization_statistics(
-            f"quantization_statistics_{q.name}.json", quantizer=q
-        )
+        runner.save_quantization_statistics(f"quantization_statistics_{q.name}.json", quantizer=q)
 
     if cfg.calc_ppl:
         runner.benchmark_perplexity(

benchmark/qwen3-14b-jointq/quant_benchmark.py

@@ -15,7 +15,7 @@
 import hydra
 from omegaconf import DictConfig, OmegaConf
 
-from onecomp import CalibrationConfig, GPTQ, JointQ, ModelConfig, Runner
+from onecomp import GPTQ, CalibrationConfig, JointQ, ModelConfig, Runner
 
 
 def create_quantizers(cfg: DictConfig):
@@ -93,9 +93,7 @@ def main(cfg: DictConfig):
     runner.run()
 
     for q in quantizers:
-        runner.save_quantization_statistics(
-            f"quantization_statistics_{q.name}.json", quantizer=q
-        )
+        runner.save_quantization_statistics(f"quantization_statistics_{q.name}.json", quantizer=q)
 
     if cfg.calc_ppl:
         runner.benchmark_perplexity(