Fixes for pip install, rename classes to standardize for requests and responses, restructure tests and begin working to get them passing

Author: markurtz

Makefile

@@ -0,0 +1,34 @@
+install:
+	pip install -r requirements.txt
+
+install-dev:
+	pip install -e .[dev]
+
+quality:
+	ruff check src tests
+	isort --check-only src tests
+	flake8 src tests --max-line-length 88
+	mypy src
+
+style:
+	ruff format src tests
+	isort src tests
+	flake8 src tests --max-line-length 88
+
+# test:
+#     pytest tests
+
+build:
+	python setup.py sdist bdist_wheel
+
+clean:
+	rm -rf __pycache__
+	rm -rf build
+	rm -rf dist
+	rm -rf *.egg-info
+	find . -type f -name "*.pyc" -delete
+	find . -type d -name "__pycache__" -exec rm -r {} +
+	rm -rf .mypy_cache
+	rm -rf .pytest_cache
+
+.PHONY: install install-dev quality style test test-unit test-integration test-e2e test-smoke test-sanity test-regression build clean

pyproject.toml

@@ -0,0 +1,28 @@
+[build-system]
+requires = ["setuptools", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.black]
+line-length = 88
+target-version = ['py38']
+
+[tool.isort]
+profile = "black"
+
+[tool.mypy]
+files = "src/guidellm"
+
+[tool.ruff]
+exclude = ["build", "dist", "env", ".venv"]
+lint.select = ["E", "F", "W"]
+
+[tool.flake8]
+max-line-length = 88
+
+[tool.pytest.ini_options]
+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"
+]
+

setup.py

@@ -0,0 +1,63 @@
+from setuptools import setup, find_packages
+from typing import Tuple
+
+
+def _setup_long_description() -> Tuple[str, str]:
+    return open("README.md", "r", encoding="utf-8").read(), "text/markdown"
+
+
+setup(
+    name='guidellm',
+    version='0.1.0',
+    author='Neuralmagic, Inc.',
+    description='Guidance platform for deploying and managing large language models.',
+    long_description=_setup_long_description()[0],
+    long_description_content_type=_setup_long_description()[1],
+    license="Apache",
+    url="https://github.com/neuralmagic/guidellm",
+    packages=find_packages(where='src'),
+    package_dir={'': 'src'},
+    include_package_data=True,
+    install_requires=[
+        'datasets',
+        'loguru',
+        'numpy',
+        'openai',
+        'requests',
+        'transformers',
+    ],
+    extras_require={
+        'dev': [
+            'pytest',
+            'sphinx',
+            'ruff',
+            'mypy',
+            'black',
+            'isort',
+            'flake8',
+            'pre-commit',
+        ],
+    },
+    entry_points={
+        'console_scripts': [
+            'guidellm=guidellm.main:main',
+        ],
+    },
+    python_requires=">=3.8.0",
+    classifiers=[
+        "Development Status :: 5 - Production/Stable",
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3 :: Only",
+        "Intended Audience :: Developers",
+        "Intended Audience :: Education",
+        "Intended Audience :: Information Technology",
+        "Intended Audience :: Science/Research",
+        "License :: OSI Approved :: Apache Software License",
+        "Operating System :: POSIX :: Linux",
+        "Topic :: Scientific/Engineering",
+        "Topic :: Scientific/Engineering :: Artificial Intelligence",
+        "Topic :: Scientific/Engineering :: Mathematics",
+        "Topic :: Software Development",
+        "Topic :: Software Development :: Libraries :: Python Modules",
+    ],
+)

src/guidellm/backend/__init__.py

@@ -1,2 +1,9 @@
-from .base import *
-from .openai import *
+from .base import Backend, BackendTypes, GenerativeResponse
+from .openai import OpenAIBackend
+
+__all__ = [
+    "Backend",
+    "BackendTypes",
+    "GenerativeResponse",
+    "OpenAIBackend",
+]

src/guidellm/backend/base.py

@@ -1,11 +1,13 @@
+import uuid
 from abc import ABC, abstractmethod
+from dataclasses import dataclass
 from enum import Enum
 from typing import Iterator, List, Optional, Type, Union
-from dataclasses import dataclass
-import uuid
+
 from loguru import logger
-from guidellm.core.result import BenchmarkResult
-from guidellm.core.request import BenchmarkRequest
+
+from guidellm.core.request import TextGenerationRequest
+from guidellm.core.result import TextGenerationResult
 
 __all__ = ["Backend", "BackendTypes", "GenerativeResponse"]
 
@@ -69,18 +71,18 @@ def create_backend(backend_type: Union[str, BackendTypes], **kwargs) -> "Backend
             raise ValueError(f"Unsupported backend type: {backend_type}")
         return Backend._registry[backend_type](**kwargs)
 
-    def submit(self, request: BenchmarkRequest) -> BenchmarkResult:
+    def submit(self, request: TextGenerationRequest) -> TextGenerationResult:
         """
-        Submit a benchmark request and populate the BenchmarkResult.
+        Submit a result request and populate the BenchmarkResult.
 
-        :param request: The benchmark request to submit.
-        :type request: BenchmarkRequest
-        :return: The populated benchmark result.
-        :rtype: BenchmarkResult
+        :param request: The result request to submit.
+        :type request: TextGenerationRequest
+        :return: The populated result result.
+        :rtype: TextGenerationResult
         """
         logger.info(f"Submitting request with prompt: {request.prompt}")
         result_id = str(uuid.uuid4())
-        result = BenchmarkResult(result_id)
+        result = TextGenerationResult(result_id)
         result.start(request.prompt)
 
         for response in self.make_request(request):
@@ -98,12 +100,14 @@ def submit(self, request: BenchmarkRequest) -> BenchmarkResult:
         return result
 
     @abstractmethod
-    def make_request(self, request: BenchmarkRequest) -> Iterator[GenerativeResponse]:
+    def make_request(
+        self, request: TextGenerationRequest
+    ) -> Iterator[GenerativeResponse]:
         """
         Abstract method to make a request to the backend.
 
-        :param request: The benchmark request to submit.
-        :type request: BenchmarkRequest
+        :param request: The result request to submit.
+        :type request: TextGenerationRequest
         :return: An iterator over the generative responses.
         :rtype: Iterator[GenerativeResponse]
         """

src/guidellm/backend/openai.py

@@ -1,17 +1,19 @@
+from typing import Any, Iterator, List, Optional
+
 import openai
-from typing import Iterator, List, Optional, Dict, Any
-from transformers import AutoTokenizer
 from loguru import logger
+from transformers import AutoTokenizer
+
 from guidellm.backend import Backend, BackendTypes, GenerativeResponse
-from guidellm.core.request import BenchmarkRequest
+from guidellm.core.request import TextGenerationRequest
 
 __all__ = ["OpenAIBackend"]
 
 
 @Backend.register_backend(BackendTypes.OPENAI_SERVER)
 class OpenAIBackend(Backend):
     """
-    An OpenAI backend implementation for the generative AI benchmark.
+    An OpenAI backend implementation for the generative AI result.
 
     :param target: The target URL string for the OpenAI server.
     :type target: str
@@ -58,15 +60,18 @@ def __init__(
             self.model = self.default_model()
 
         logger.info(
-            f"Initialized OpenAIBackend with target: {self.target} and model: {self.model}"
+            f"Initialized OpenAIBackend with target: {self.target} "
+            f"and model: {self.model}"
         )
 
-    def make_request(self, request: BenchmarkRequest) -> Iterator[GenerativeResponse]:
+    def make_request(
+        self, request: TextGenerationRequest
+    ) -> Iterator[GenerativeResponse]:
         """
         Make a request to the OpenAI backend.
 
-        :param request: The benchmark request to submit.
-        :type request: BenchmarkRequest
+        :param request: The result request to submit.
+        :type request: TextGenerationRequest
         :return: An iterator over the generative responses.
         :rtype: Iterator[GenerativeResponse]
         """
@@ -84,7 +89,10 @@ def make_request(self, request: BenchmarkRequest) -> Iterator[GenerativeResponse
             request_args.update(self.request_args)
 
         response = openai.Completion.create(
-            engine=self.model, prompt=request.prompt, stream=True, **request_args,
+            engine=self.model,
+            prompt=request.prompt,
+            stream=True,
+            **request_args,
         )
 
         for chunk in response:

src/guidellm/core/__init__.py

@@ -1,3 +1,19 @@
-from .distribution import *
-from .request import *
-from .result import *
+from .distribution import Distribution
+from .request import TextGenerationRequest
+from .result import (
+    RequestConcurrencyMeasurement,
+    TextGenerationBenchmark,
+    TextGenerationBenchmarkReport,
+    TextGenerationError,
+    TextGenerationResult,
+)
+
+__all__ = [
+    "Distribution",
+    "TextGenerationRequest",
+    "TextGenerationResult",
+    "TextGenerationError",
+    "TextGenerationBenchmark",
+    "TextGenerationBenchmarkReport",
+    "RequestConcurrencyMeasurement",
+]