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

[CI/Build] Update Ruff version (#8469)

Browse Source 
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
This commit is contained in:
Aaron Pham
2024-09-18 07:00:56 -04:00
committed by GitHub
parent 6ffa3f314c
commit 9d104b5beb
27 changed files with 50 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/worker/test_encoder_decoder_model_runner.py
View File

@@ -453,8 +453,7 @@ def test_prepare_decode(batch_size):
# each sequence) in the decode phase
expected_selected_token_indices = []
selected_token_start_idx = 0
for seq_len in seq_lens:
for selected_token_start_idx, seq_len in enumerate(seq_lens):
# Compute the index offset of the final token in each
# sequence's decoded outputs; since a single token is
# decoded per iteration per sequence, then the length
@@ -463,7 +462,6 @@ def test_prepare_decode(batch_size):
# generated tokens is 0 (i.e. the expected sampling index
# for a given sequence is just `selected_token_start_idx`)
expected_selected_token_indices.append(selected_token_start_idx)
selected_token_start_idx += 1
sampling_metadata = model_input.sampling_metadata
actual = sampling_metadata.selected_token_indices