biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[V1] Make v1 more testable (#9888)

Browse Source 
Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>
This commit is contained in:
Joe Runde
2024-11-06 12:57:35 -07:00
committed by GitHub
parent 87bd7e0515
commit d58268c56a
75 changed files with 243 additions and 165 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/model_executor/models/granite.py
View File

@@ -42,7 +42,7 @@ from vllm.model_executor.layers.quantization.base_config import (
from vllm.model_executor.layers.quantization.compressed_tensors.utils import (
get_compressed_tensors_cache_scale)
from vllm.model_executor.layers.rotary_embedding import get_rope
from vllm.model_executor.layers.sampler import Sampler, SamplerOutput
from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
from vllm.model_executor.layers.vocab_parallel_embedding import (
DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead, VocabParallelEmbedding)
from vllm.model_executor.model_loader.weight_utils import (
@@ -411,7 +411,7 @@ class GraniteForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
config.vocab_size,
scale=logit_scale)
self.sampler = Sampler()
self.sampler = get_sampler()
else:
self.lm_head = PPMissingLayer()