fixes from copilot review and standardize backend package to backends for plural

Signed-off-by: Mark Kurtz <[email protected]>

Author: markurtz

src/guidellm/__main__.py

@@ -6,7 +6,7 @@
 import click
 from pydantic import ValidationError
 
-from guidellm.backend import BackendType
+from guidellm.backends import BackendType
 from guidellm.benchmark import (
     ProfileType,
     reimport_benchmarks_report,

src/guidellm/backend/__init__.py renamed to src/guidellm/backends/__init__.py

@@ -16,7 +16,7 @@
 from abc import abstractmethod
 from typing import Literal
 
-from guidellm.backend.objects import (
+from guidellm.backends.objects import (
     GenerationRequest,
     GenerationResponse,
 )

src/guidellm/backend/backend.py renamed to src/guidellm/backends/backend.py

@@ -23,8 +23,8 @@
 from PIL import Image
 from pydantic import dataclasses
 
-from guidellm.backend.backend import Backend
-from guidellm.backend.objects import (
+from guidellm.backends.backend import Backend
+from guidellm.backends.objects import (
     GenerationRequest,
     GenerationRequestTimings,
     GenerationResponse,
@@ -351,8 +351,8 @@ async def resolve(
 
             if usage_stats is not None:
                 request_info.request_timings.request_end = time.time()
-                response.request_output_tokens = usage_stats.output_tokens
-                response.request_prompt_tokens = usage_stats.prompt_tokens
+                response.response_output_tokens = usage_stats.output_tokens
+                response.response_prompt_tokens = usage_stats.prompt_tokens
 
             yield response, request_info
 
@@ -602,7 +602,7 @@ def _get_body(
         **kwargs,
     ) -> dict[str, Any]:
         # Start with endpoint-specific extra body parameters
-        extra_body = self.extra_body.get(endpoint_type, self.extra_body)
+        extra_body: dict = self.extra_body.get(endpoint_type, self.extra_body)
 
         body = copy.deepcopy(extra_body)
         body.update(request_kwargs or {})
@@ -622,14 +622,22 @@ def _get_body(
             if max_output_tokens:
                 body.update({"stop": None, "ignore_eos": True})
 
+            if self.remove_from_body:
+                for key in self.remove_from_body:
+                    body.pop(key, None)
+
         return {key: val for key, val in body.items() if val is not None}
 
     def _get_completions_text_content(self, data: dict) -> Optional[str]:
         if not data.get("choices"):
             return None
 
-        choice = data["choices"][0]
-        return choice.get("text") or choice.get("delta", {}).get("content")
+        choice: dict = data["choices"][0]
+        return (
+            choice.get("text")
+            or choice.get("delta", {}).get("content")
+            or choice.get("message", {}).get("content")
+        )
 
     def _get_completions_usage_stats(self, data: dict) -> Optional[UsageStats]:
         if not data.get("usage"):

src/guidellm/backend/objects.py renamed to src/guidellm/backends/objects.py

@@ -12,7 +12,7 @@
 
 from pydantic import Field
 
-from guidellm.backend import ResponseSummary
+from guidellm.backends import ResponseSummary
 from guidellm.benchmark.benchmark import (
     BenchmarkArgs,
     BenchmarkRunStats,

src/guidellm/backend/openai.py renamed to src/guidellm/backends/openai.py

@@ -14,7 +14,7 @@
 from pydantic import Field
 from transformers import PreTrainedTokenizerBase  # type: ignore  # noqa: PGH003
 
-from guidellm.backend import Backend, ResponseSummary
+from guidellm.backends import Backend, ResponseSummary
 from guidellm.benchmark.aggregator import (
     AggregatorT,
     BenchmarkT,

src/guidellm/benchmark/aggregator.py

@@ -7,7 +7,7 @@
     PreTrainedTokenizerBase,
 )
 
-from guidellm.backend import Backend, BackendType
+from guidellm.backends import Backend, BackendType
 from guidellm.benchmark.benchmarker import GenerativeBenchmarker
 from guidellm.benchmark.output import (
     GenerativeBenchmarksConsole,

src/guidellm/benchmark/benchmarker.py

@@ -9,7 +9,7 @@
     PreTrainedTokenizerBase,
 )
 
-from guidellm.backend.backend import BackendType
+from guidellm.backends.backend import BackendType
 from guidellm.benchmark.profile import ProfileType
 from guidellm.objects.pydantic import StandardBaseModel
 from guidellm.scheduler.strategies import StrategyType

src/guidellm/benchmark/entrypoints.py

@@ -12,8 +12,8 @@
 
 import pytest
 
-from guidellm.backend.backend import Backend, BackendType
-from guidellm.backend.objects import (
+from guidellm.backends.backend import Backend, BackendType
+from guidellm.backends.objects import (
     GenerationRequest,
     GenerationRequestTimings,
 )
@@ -244,7 +244,7 @@ class TestBackendRegistry:
     @pytest.mark.smoke
     def test_openai_backend_registered(self):
         """Test that OpenAI HTTP backend is registered."""
-        from guidellm.backend.openai import OpenAIHTTPBackend
+        from guidellm.backends.openai import OpenAIHTTPBackend
 
         # OpenAI backend should be registered
         backend = Backend.create("openai_http", target="http://test")
@@ -262,7 +262,7 @@ def test_backend_create_invalid_type(self):
     @pytest.mark.smoke
     def test_backend_registry_functionality(self):
         """Test that backend registry functions work."""
-        from guidellm.backend.openai import OpenAIHTTPBackend
+        from guidellm.backends.openai import OpenAIHTTPBackend
 
         # Test that we can get registered backends
         openai_class = Backend.get_registered_object("openai_http")
@@ -327,6 +327,6 @@ def test_backend_registered_objects(self):
         assert len(registered) > 0
 
         # Check that openai backend is in the registered objects
-        from guidellm.backend.openai import OpenAIHTTPBackend
+        from guidellm.backends.openai import OpenAIHTTPBackend
 
         assert OpenAIHTTPBackend in registered