use descriptive method names

Author: mayank31398

inference_server/model_handler/deployment.py

@@ -132,16 +132,16 @@ def _initialize_grpc_client(self):
             self.stubs.append(stub)
 
     # runs task in parallel and return the result from the first task
-    async def _query_in_tensor_parallel(self, text: List[str], generate_kwargs: dict):
+    async def generate_in_tensor_parallel(self, text: List[str], generate_kwargs: dict):
         responses = []
         for i in range(self.num_gpus):
-            responses.append(self.asyncio_loop.create_task(self._request_async_response(i, text, generate_kwargs)))
+            responses.append(self.asyncio_loop.create_task(self.generate_async(i, text, generate_kwargs)))
 
         await responses[0]
         return responses[0]
 
-    async def _request_async_response(self, stub_id: int, text: List[str], generate_kwargs: dict):
-        req = generation_pb2.GenerationRequest(texts=text, generate_kwargs=generate_kwargs)
+    async def generate_async(self, stub_id: int, text: List[str], generate_kwargs: dict):
+        req = generation_pb2.GenerationRequestProto(texts=text, generate_kwargs=generate_kwargs)
         response = await self.stubs[stub_id].Generate(req)
         return response
 
@@ -157,7 +157,7 @@ def generate(self, **kwargs) -> GenerateResponse:
             generate_kwargs = self.dict_to_proto(generate_kwargs)
 
             response = self.asyncio_loop.run_until_complete(
-                self._query_in_tensor_parallel(text, generate_kwargs)
+                self.generate_in_tensor_parallel(text, generate_kwargs)
             ).result()
 
             if response.error:

inference_server/model_handler/grpc_utils/pb/generation_pb2.py

@@ -16,8 +16,8 @@ def __init__(self, channel):
         """
         self.Generate = channel.unary_unary(
             "/generation.GenerationService/Generate",
-            request_serializer=generation__pb2.GenerationRequest.SerializeToString,
-            response_deserializer=generation__pb2.GenerationResponse.FromString,
+            request_serializer=generation__pb2.GenerationRequestProto.SerializeToString,
+            response_deserializer=generation__pb2.GenerationResponseProto.FromString,
         )
 
 
@@ -35,8 +35,8 @@ def add_GenerationServiceServicer_to_server(servicer, server):
     rpc_method_handlers = {
         "Generate": grpc.unary_unary_rpc_method_handler(
             servicer.Generate,
-            request_deserializer=generation__pb2.GenerationRequest.FromString,
-            response_serializer=generation__pb2.GenerationResponse.SerializeToString,
+            request_deserializer=generation__pb2.GenerationRequestProto.FromString,
+            response_serializer=generation__pb2.GenerationResponseProto.SerializeToString,
         ),
     }
     generic_handler = grpc.method_handlers_generic_handler("generation.GenerationService", rpc_method_handlers)
@@ -64,8 +64,8 @@ def Generate(
             request,
             target,
             "/generation.GenerationService/Generate",
-            generation__pb2.GenerationRequest.SerializeToString,
-            generation__pb2.GenerationResponse.FromString,
+            generation__pb2.GenerationRequestProto.SerializeToString,
+            generation__pb2.GenerationResponseProto.FromString,
             options,
             channel_credentials,
             insecure,

inference_server/model_handler/grpc_utils/pb/generation_pb2_grpc.py

@@ -2,7 +2,7 @@ syntax = "proto3";
 package generation;
 
 service GenerationService {
-    rpc Generate (GenerationRequest) returns (GenerationResponse) {}
+    rpc Generate (GenerationRequestProto) returns (GenerationResponseProto) {}
 }
 
 message Value {
@@ -14,12 +14,12 @@ message Value {
     }
 }
 
-message GenerationRequest {
+message GenerationRequestProto {
     repeated string texts = 1;
     map<string,Value> generate_kwargs = 2;
 }
 
-message GenerationResponse {
+message GenerationResponseProto {
     repeated string texts = 1;
     repeated int32 num_generated_tokens = 2;
     string error = 3;