Fix pandas dependency in CI (#3818)

This commit is contained in:
Lianmin Zheng
2025-02-24 05:56:57 -08:00
committed by GitHub
parent 1a6e97577a
commit c9745ee082
2 changed files with 8 additions and 8 deletions

View File

@@ -124,8 +124,8 @@ class ServerArgs:
speculative_draft_model_path: Optional[str] = None speculative_draft_model_path: Optional[str] = None
speculative_algorithm: Optional[str] = None speculative_algorithm: Optional[str] = None
speculative_num_steps: int = 5 speculative_num_steps: int = 5
speculative_num_draft_tokens: int = 64
speculative_eagle_topk: int = 8 speculative_eagle_topk: int = 8
speculative_num_draft_tokens: int = 64
# Double Sparsity # Double Sparsity
enable_double_sparsity: bool = False enable_double_sparsity: bool = False
@@ -719,12 +719,6 @@ class ServerArgs:
help="The number of steps sampled from draft model in Speculative Decoding.", help="The number of steps sampled from draft model in Speculative Decoding.",
default=ServerArgs.speculative_num_steps, default=ServerArgs.speculative_num_steps,
) )
parser.add_argument(
"--speculative-num-draft-tokens",
type=int,
help="The number of token sampled from draft model in Speculative Decoding.",
default=ServerArgs.speculative_num_draft_tokens,
)
parser.add_argument( parser.add_argument(
"--speculative-eagle-topk", "--speculative-eagle-topk",
type=int, type=int,
@@ -732,6 +726,12 @@ class ServerArgs:
choices=[1, 2, 4, 8], choices=[1, 2, 4, 8],
default=ServerArgs.speculative_eagle_topk, default=ServerArgs.speculative_eagle_topk,
) )
parser.add_argument(
"--speculative-num-draft-tokens",
type=int,
help="The number of token sampled from draft model in Speculative Decoding.",
default=ServerArgs.speculative_num_draft_tokens,
)
# Double Sparsity # Double Sparsity
parser.add_argument( parser.add_argument(

View File

@@ -20,7 +20,7 @@ pip install flashinfer_python==0.2.1.post2 --find-links ${FLASHINFER_REPO} --for
pip install torch_memory_saver --force-reinstall pip install torch_memory_saver --force-reinstall
pip install transformers==4.45.2 sentence_transformers accelerate peft pip install transformers==4.45.2 sentence_transformers accelerate peft pandas datasets
# For compling xgrammar kernels # For compling xgrammar kernels
pip install cuda-python nvidia-cuda-nvrtc-cu12 pip install cuda-python nvidia-cuda-nvrtc-cu12