A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 23:15 UTC
[CI] Upgrade codespell version. (#5381)
SangBin Cho committed
Jun 13, 2024 at 02:06 UTC
847cdcca1c94b12e6c118dbf863e4b111d1b4fd2
4 files changed
+4
-4
.github/workflows/ruff.yml
+1
-1
@@ -25,7 +25,7 @@ jobs:
25
- name: Install dependencies
26
run: |
27
python -m pip install --upgrade pip
28
- pip install ruff==0.1.5 codespell==2.2.6 tomli==2.0.1 isort==5.13.2
28
+ pip install ruff==0.1.5 codespell==2.3.0 tomli==2.0.1 isort==5.13.2
29
- name: Analysing the code with ruff
30
run: |
31
ruff .
requirements-dev.txt
+1
-1
@@ -3,7 +3,7 @@ yapf==0.32.0
3
toml==0.10.2
4
tomli==2.0.1
5
ruff==0.1.5
6
-codespell==2.2.6
6
+codespell==2.3.0
7
isort==5.13.2
8
clang-format==18.1.5
9
tests/core/test_chunked_prefill_scheduler.py
+1
-1
@@ -149,7 +149,7 @@ def test_complex():
149
# Only the first seq group has a new token appended.
150
append_new_token(running[0], 1)
151
152
- # Add 2 more requsets.
152
+ # Add 2 more requests.
153
for i in range(2, 4):
154
_, seq_group = create_dummy_prompt(str(i), prompt_length=60)
155
scheduler.add_seq_group(seq_group)
tests/test_sharded_state_loader.py
+1
-1
@@ -39,7 +39,7 @@ def test_filter_subtensors():
39
filtered_state_dict = ShardedStateLoader._filter_subtensors(state_dict)
40
assert tuple(filtered_state_dict.keys()) == ("a", "b", "c")
41
for key, tensor in filtered_state_dict.items():
42
- # NOTE: don't use `euqal` here, as the tensor might contain NaNs
42
+ # NOTE: don't use `equal` here, as the tensor might contain NaNs
43
assert tensor is state_dict[key]
44
45