Merge branch 'features/refactor/core' into features/refactor/base-draft

[GuideLLM Refactor] core updates for pyproject.toml and settings.py #352

Author: sjmonson

pyproject.toml

@@ -10,9 +10,6 @@ include = ["*"]
 [tool.setuptools.package-data]
 "guidellm.data" = ["*.gz"]
 
-[tool.pdm]
-distribution = true
-
 
 # ************************************************
 # ********** Project Metadata **********
@@ -24,9 +21,8 @@ name = "guidellm"
 description = "Guidance platform for deploying and managing large language models."
 readme = { file = "README.md", content-type = "text/markdown" }
 requires-python = ">=3.9.0,<4.0"
-license = "Apache-2.0"
-license-files = ["LICENSE"]
-authors = [ { name = "Red Hat" } ]
+license = { text = "Apache-2.0" }
+authors = [{ name = "Red Hat" }]
 keywords = [
     "ai",
     "benchmarking",
@@ -47,25 +43,27 @@ keywords = [
 ]
 dependencies = [
     "click>=8.0.0,<8.2.0",
+    "culsans~=0.9.0",
     "datasets",
+    "eval_type_backport",
+    "faker",
     "ftfy>=6.0.0",
     "httpx[http2]<1.0.0",
     "loguru",
+    "msgpack",
     "numpy",
     "pillow",
     "protobuf",
     "pydantic>=2.11.7",
     "pydantic-settings>=2.0.0",
     "pyyaml>=6.0.0",
     "rich",
+    "sanic",
     "transformers",
+    "uvloop>=0.18",
 ]
 
 [project.optional-dependencies]
-recommended = [
-    "tiktoken>=0.11.0",  # For OpenAI tokenizer
-    "blobfile>=3.1.0",   # For OpenAI tokenizer
-]
 dev = [
     # build
     "build>=1.0.0",
@@ -81,7 +79,7 @@ dev = [
     # testing
     "lorem~=0.1.1",
     "pytest~=8.2.2",
-    "pytest-asyncio~=0.23.8",
+    "pytest-asyncio~=1.1.0",
     "pytest-cov~=5.0.0",
     "pytest-mock~=3.14.0",
     "pytest-rerunfailures~=14.0",
@@ -106,9 +104,6 @@ dev = [
     "mkdocs-linkcheck~=1.0.6",
 ]
 
-[dependency-groups]
-dev = [ "guidellm[dev]" ]
-
 [project.urls]
 homepage = "https://github.com/vllm-project/guidellm"
 source = "https://github.com/vllm-project/guidellm"
@@ -143,11 +138,17 @@ exclude = ["venv", ".tox"]
 follow_imports = 'silent'
 
 [[tool.mypy.overrides]]
-module = ["datasets.*", "transformers.*", "setuptools.*", "setuptools_git_versioning.*"]
-ignore_missing_imports=true
+module = [
+    "datasets.*",
+    "transformers.*",
+    "setuptools.*",
+    "setuptools_git_versioning.*",
+]
+ignore_missing_imports = true
 
 
 [tool.ruff]
+target-version = "py39"
 line-length = 88
 indent-width = 4
 exclude = ["build", "dist", "env", ".venv"]
@@ -158,82 +159,83 @@ indent-style = "space"
 
 [tool.ruff.lint]
 ignore = [
-    "PLR0913",
-    "TC001",
-    "COM812",
-    "ISC001",
-    "TC002",
+    "COM812",  # ignore trailing comma errors due to older Python versions
+    "PD011",   # ignore .values usage since ruff assumes it's a Pandas DataFrame
+    "PLR0913", # ignore too many arguments in function definitions
     "PLW1514", # allow Path.open without encoding
-    "RET505", # allow `else` blocks
-    "RET506", # allow `else` blocks
-    "PD011", # ignore .values usage since ruff assumes it's a Pandas DataFrame
+    "RET505",  # allow `else` blocks
+    "RET506",  # allow `else` blocks
+    "S311",    # allow standard pseudo-random generators
+    "TC001",   # ignore imports used only for type checking
+    "TC002",   # ignore imports used only for type checking
+    "TC003",   # ignore imports used only for type checking
 ]
 select = [
     # Rules reference: https://docs.astral.sh/ruff/rules/
 
     # Code Style / Formatting
-    "E", # pycodestyle: checks adherence to PEP 8 conventions including spacing, indentation, and line length
-    "W", # pycodestyle: checks adherence to PEP 8 conventions including spacing, indentation, and line length
-    "A", # flake8-builtins: prevents shadowing of Python built-in names
-    "C", # Convention: ensures code adheres to specific style and formatting conventions
-    "COM", # flake8-commas: enforces the correct use of trailing commas
-    "ERA", # eradicate: detects commented-out code that should be removed
-    "I", # isort: ensures imports are sorted in a consistent manner
-    "ICN", # flake8-import-conventions: enforces import conventions for better readability
-    "N", # pep8-naming: enforces PEP 8 naming conventions for classes, functions, and variables
-    "NPY", # NumPy: enforces best practices for using the NumPy library
-    "PD", # pandas-vet: enforces best practices for using the pandas library
-    "PT", # flake8-pytest-style: enforces best practices and style conventions for pytest tests
-    "PTH", # flake8-use-pathlib: encourages the use of pathlib over os.path for file system operations
-    "Q", # flake8-quotes: enforces consistent use of single or double quotes
-    "TCH", # flake8-type-checking: enforces type checking practices and standards
-    "TID", # flake8-tidy-imports: enforces tidy and well-organized imports
+    "E",      # pycodestyle: checks adherence to PEP 8 conventions including spacing, indentation, and line length
+    "W",      # pycodestyle: checks adherence to PEP 8 conventions including spacing, indentation, and line length
+    "A",      # flake8-builtins: prevents shadowing of Python built-in names
+    "C",      # Convention: ensures code adheres to specific style and formatting conventions
+    "COM",    # flake8-commas: enforces the correct use of trailing commas
+    "ERA",    # eradicate: detects commented-out code that should be removed
+    "I",      # isort: ensures imports are sorted in a consistent manner
+    "ICN",    # flake8-import-conventions: enforces import conventions for better readability
+    "N",      # pep8-naming: enforces PEP 8 naming conventions for classes, functions, and variables
+    "NPY",    # NumPy: enforces best practices for using the NumPy library
+    "PD",     # pandas-vet: enforces best practices for using the pandas library
+    "PT",     # flake8-pytest-style: enforces best practices and style conventions for pytest tests
+    "PTH",    # flake8-use-pathlib: encourages the use of pathlib over os.path for file system operations
+    "Q",      # flake8-quotes: enforces consistent use of single or double quotes
+    "TCH",    # flake8-type-checking: enforces type checking practices and standards
+    "TID",    # flake8-tidy-imports: enforces tidy and well-organized imports
     "RUF022", # flake8-ruff: enforce sorting of __all__ in modules
 
     # Code Structure / Complexity
-    "C4", # flake8-comprehensions: improves readability and performance of list, set, and dict comprehensions
+    "C4",  # flake8-comprehensions: improves readability and performance of list, set, and dict comprehensions
     "C90", # mccabe: checks for overly complex code using cyclomatic complexity
     "ISC", # flake8-implicit-str-concat: prevents implicit string concatenation
     "PIE", # flake8-pie: identifies and corrects common code inefficiencies and mistakes
-    "R", # Refactor: suggests improvements to code structure and readability
+    "R",   # Refactor: suggests improvements to code structure and readability
     "SIM", # flake8-simplify: simplifies complex expressions and improves code readability
 
     # Code Security / Bug Prevention
-    "ARG", # flake8-unused-arguments: detects unused function and method arguments
+    "ARG",   # flake8-unused-arguments: detects unused function and method arguments
     "ASYNC", # flake8-async: identifies incorrect or inefficient usage patterns in asynchronous code
-    "B", # flake8-bugbear: detects common programming mistakes and potential bugs
-    "BLE", # flake8-blind-except: prevents blind exceptions that catch all exceptions without handling
-    "E", # Error: detects and reports errors in the code
-    "F", # Pyflakes: detects unused imports, shadowed imports, undefined variables, and various formatting errors in string operations
-    "INP", # flake8-no-pep420: prevents implicit namespace packages by requiring __init__.py
-    "PGH", # pygrep-hooks: detects deprecated and dangerous code patterns
-    "PL", # Pylint: comprehensive source code analyzer for enforcing coding standards and detecting errors
-    "RSE", # flake8-raise: ensures exceptions are raised correctly
-    "S", # flake8-bandit: detects security issues and vulnerabilities in the code
-    "SLF", # flake8-self: prevents incorrect usage of the self argument in class methods
-    "T10", # flake8-debugger: detects the presence of debugging tools such as pdb
-    "T20", # flake8-print: detects print statements left in the code
-    "UP", # pyupgrade: automatically upgrades syntax for newer versions of Python
-    "W", # Warning: provides warnings about potential issues in the code
-    "YTT", # flake8-2020: identifies code that will break with future Python releases
+    "B",     # flake8-bugbear: detects common programming mistakes and potential bugs
+    "BLE",   # flake8-blind-except: prevents blind exceptions that catch all exceptions without handling
+    "E",     # Error: detects and reports errors in the code
+    "F",     # Pyflakes: detects unused imports, shadowed imports, undefined variables, and various formatting errors in string operations
+    "INP",   # flake8-no-pep420: prevents implicit namespace packages by requiring __init__.py
+    "PGH",   # pygrep-hooks: detects deprecated and dangerous code patterns
+    "PL",    # Pylint: comprehensive source code analyzer for enforcing coding standards and detecting errors
+    "RSE",   # flake8-raise: ensures exceptions are raised correctly
+    "S",     # flake8-bandit: detects security issues and vulnerabilities in the code
+    "SLF",   # flake8-self: prevents incorrect usage of the self argument in class methods
+    "T10",   # flake8-debugger: detects the presence of debugging tools such as pdb
+    "T20",   # flake8-print: detects print statements left in the code
+    "UP",    # pyupgrade: automatically upgrades syntax for newer versions of Python
+    "W",     # Warning: provides warnings about potential issues in the code
+    "YTT",   # flake8-2020: identifies code that will break with future Python releases
 
     # Code Documentation
     "FIX", # flake8-fixme: detects FIXMEs and other temporary comments that should be resolved
 ]
 
 [tool.ruff.lint.extend-per-file-ignores]
 "tests/**/*.py" = [
-    "S101", # asserts allowed in tests
-    "ARG", # Unused function args allowed in tests
+    "S101",    # asserts allowed in tests
+    "ARG",     # Unused function args allowed in tests
     "PLR2004", # Magic value used in comparison
-    "TCH002", # No import only type checking in tests
-    "SLF001", # enable private member access in tests
-    "S105", # allow hardcoded passwords in tests
-    "S311", # allow standard pseudo-random generators in tests
-    "PT011", # allow generic exceptions in tests
-    "N806", # allow uppercase variable names in tests
-    "PGH003", # allow general ignores in tests
-    "S106", # allow hardcoded passwords in tests
+    "TCH002",  # No import only type checking in tests
+    "SLF001",  # enable private member access in tests
+    "S105",    # allow hardcoded passwords in tests
+    "S311",    # allow standard pseudo-random generators in tests
+    "PT011",   # allow generic exceptions in tests
+    "N806",    # allow uppercase variable names in tests
+    "PGH003",  # allow general ignores in tests
+    "S106",    # allow hardcoded passwords in tests
     "PLR0915", # allow complext statements in tests
 ]
 
@@ -246,5 +248,5 @@ addopts = '-s -vvv --cache-clear'
 markers = [
     "smoke: quick tests to check basic functionality",
     "sanity: detailed tests to ensure major functions work correctly",
-    "regression: tests to ensure that new changes do not break existing functionality"
+    "regression: tests to ensure that new changes do not break existing functionality",
 ]

src/guidellm/__init__.py

@@ -20,7 +20,8 @@
     hf_logging.set_verbosity_error()
     logging.getLogger("transformers").setLevel(logging.ERROR)
 
-from .config import (
+from .logger import configure_logger, logger
+from .settings import (
     DatasetSettings,
     Environment,
     LoggingSettings,
@@ -30,7 +31,6 @@
     reload_settings,
     settings,
 )
-from .logger import configure_logger, logger
 
 __all__ = [
     "DatasetSettings",

src/guidellm/__main__.py

@@ -13,9 +13,9 @@
 )
 from guidellm.benchmark.entrypoints import benchmark_with_scenario
 from guidellm.benchmark.scenario import GenerativeTextScenario, get_builtin_scenarios
-from guidellm.config import print_config
 from guidellm.preprocess.dataset import ShortPromptStrategy, process_dataset
 from guidellm.scheduler import StrategyType
+from guidellm.settings import print_config
 from guidellm.utils import DefaultGroupHandler
 from guidellm.utils import cli as cli_tools
 

src/guidellm/backend/backend.py

@@ -7,7 +7,7 @@
 from PIL import Image
 
 from guidellm.backend.response import ResponseSummary, StreamingTextResponse
-from guidellm.config import settings
+from guidellm.settings import settings
 
 __all__ = [
     "Backend",

src/guidellm/backend/openai.py

@@ -16,7 +16,7 @@
     ResponseSummary,
     StreamingTextResponse,
 )
-from guidellm.config import settings
+from guidellm.settings import settings
 
 __all__ = [
     "CHAT_COMPLETIONS",

src/guidellm/backend/response.py

@@ -2,8 +2,8 @@
 
 from pydantic import computed_field
 
-from guidellm.config import settings
 from guidellm.objects.pydantic import StandardBaseModel
+from guidellm.settings import settings
 
 __all__ = [
     "RequestArgs",

src/guidellm/benchmark/aggregator.py

@@ -21,7 +21,6 @@
     GenerativeTextErrorStats,
     GenerativeTextResponseStats,
 )
-from guidellm.config import settings
 from guidellm.objects import (
     RunningStats,
     StandardBaseModel,
@@ -40,6 +39,7 @@
     SchedulerRequestResult,
     WorkerDescription,
 )
+from guidellm.settings import settings
 from guidellm.utils import check_load_processor
 
 __all__ = [

src/guidellm/benchmark/output.py

@@ -20,7 +20,6 @@
     SweepProfile,
     ThroughputProfile,
 )
-from guidellm.config import settings
 from guidellm.objects import (
     DistributionSummary,
     StandardBaseModel,
@@ -29,6 +28,7 @@
 from guidellm.presentation import UIDataBuilder
 from guidellm.presentation.injector import create_report
 from guidellm.scheduler import strategy_display_str
+from guidellm.settings import settings
 from guidellm.utils import Colors, split_text_list_by_length
 from guidellm.utils.dict import recursive_key_update
 from guidellm.utils.text import camelize_str

src/guidellm/benchmark/profile.py

@@ -4,7 +4,6 @@
 import numpy as np
 from pydantic import Field, computed_field
 
-from guidellm.config import settings
 from guidellm.objects import StandardBaseModel
 from guidellm.scheduler import (
     AsyncConstantStrategy,
@@ -15,6 +14,7 @@
     SynchronousStrategy,
     ThroughputStrategy,
 )
+from guidellm.settings import settings
 
 __all__ = [
     "AsyncProfile",

src/guidellm/logger.py

@@ -41,7 +41,7 @@
 
 from loguru import logger
 
-from guidellm.config import LoggingSettings, settings
+from guidellm.settings import LoggingSettings, settings
 
 __all__ = ["configure_logger", "logger"]
 
@@ -72,7 +72,7 @@ def configure_logger(config: LoggingSettings = settings.logging):
         sys.stdout,
         level=config.console_log_level.upper(),
         format="<green>{time:YY-MM-DD HH:mm:ss}</green>|<level>{level: <8}</level> \
-        |<cyan>{name}:{function}</cyan>:<cyan>{line}</cyan> - <level>{message}</level>"
+        |<cyan>{name}:{function}</cyan>:<cyan>{line}</cyan> - <level>{message}</level>",
     )
 
     if config.log_file or config.log_file_level: