[Core] Streamline some structured output related code (#26737)
Signed-off-by: Nick Hill <nhill@redhat.com>
This commit is contained in:
@@ -30,7 +30,6 @@ from vllm.v1.kv_cache_interface import (
|
||||
from vllm.v1.outputs import DraftTokenIds, ModelRunnerOutput
|
||||
from vllm.v1.request import Request, RequestStatus
|
||||
from vllm.v1.structured_output import StructuredOutputManager
|
||||
from vllm.v1.structured_output.request import StructuredOutputRequest
|
||||
|
||||
from .utils import EOS_TOKEN_ID, create_requests, create_scheduler
|
||||
|
||||
@@ -335,10 +334,10 @@ def test_stop_via_update_from_output():
|
||||
requests[0].request_id: [],
|
||||
requests[1].request_id: [10],
|
||||
},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -383,10 +382,10 @@ def test_stop_via_update_from_output():
|
||||
requests[0].request_id: [10, 42],
|
||||
requests[1].request_id: [13],
|
||||
},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -429,10 +428,10 @@ def test_stop_via_update_from_output():
|
||||
requests[0].request_id: [10, 11],
|
||||
requests[1].request_id: [],
|
||||
},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -470,10 +469,10 @@ def test_stop_via_update_from_output():
|
||||
total_num_scheduled_tokens=3,
|
||||
scheduled_encoder_inputs={},
|
||||
scheduled_spec_decode_tokens={requests[0].request_id: [EOS_TOKEN_ID, 10]},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -1941,7 +1940,6 @@ def test_schedule_skip_tokenizer_init_structured_output_request():
|
||||
sampling_params=sampling_params,
|
||||
pooling_params=None,
|
||||
eos_token_id=EOS_TOKEN_ID,
|
||||
structured_output_request=StructuredOutputRequest(sampling_params),
|
||||
)
|
||||
scheduler.add_request(request)
|
||||
output = scheduler.schedule()
|
||||
|
||||
@@ -26,7 +26,7 @@ def _make_empty_scheduler_output():
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
kv_connector_metadata=SharedStorageConnectorMetadata(),
|
||||
)
|
||||
|
||||
@@ -89,10 +89,10 @@ def _schedule_new_request(*req_ids: str) -> SchedulerOutput:
|
||||
total_num_scheduled_tokens=total_num_scheduled_tokens,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -168,10 +168,10 @@ def test_update_states_request_finished(model_runner):
|
||||
total_num_scheduled_tokens=0,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids={req_id},
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -198,10 +198,10 @@ def test_update_states_request_resumed(model_runner):
|
||||
total_num_scheduled_tokens=0,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -225,10 +225,10 @@ def test_update_states_request_resumed(model_runner):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -256,10 +256,10 @@ def test_update_states_no_changes(model_runner):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -291,10 +291,10 @@ def test_update_states_request_unscheduled(model_runner):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
|
||||
@@ -146,10 +146,10 @@ def _schedule_new_request(*req_ids: str) -> SchedulerOutput:
|
||||
total_num_scheduled_tokens=total_num_scheduled_tokens,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -212,10 +212,10 @@ def test_update_states_request_finished(model_runner, dist_init):
|
||||
total_num_scheduled_tokens=0,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids={req_id},
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -244,10 +244,10 @@ def test_update_states_request_resumed(model_runner, dist_init):
|
||||
total_num_scheduled_tokens=0,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -273,10 +273,10 @@ def test_update_states_request_resumed(model_runner, dist_init):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -366,10 +366,10 @@ def test_update_states_no_changes(model_runner, dist_init):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
@@ -403,10 +403,10 @@ def test_update_states_request_unscheduled(model_runner, dist_init):
|
||||
total_num_scheduled_tokens=1,
|
||||
scheduled_spec_decode_tokens={},
|
||||
scheduled_encoder_inputs={},
|
||||
num_common_prefix_blocks=0,
|
||||
num_common_prefix_blocks=[],
|
||||
finished_req_ids=set(),
|
||||
free_encoder_mm_hashes=[],
|
||||
structured_output_request_ids={},
|
||||
structured_output_request_ids=[],
|
||||
grammar_bitmask=None,
|
||||
)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user