huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-21 23:12:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix the warmup issue of prefill batch_size (#268)

Browse Source 
Signed-off-by: yuanwu <yuan.wu@intel.com>
This commit is contained in:
Yuan Wu 2025-01-24 00:26:17 +08:00 committed by GitHub
parent 63c64bb307
commit fe7594e369
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
server/text_generation_server/models/causal_lm.py
View File

@ -1197,7 +1197,7 @@ class CausalLM(Model):
# Warmup prefill batch_size # Warmup prefill batch_size
max_input_length = request.max_input_length max_input_length = request.max_input_length
prefill_batch_size_list = [batch for batch in range(BATCH_BUCKET_SIZE, max_prefill_batch_size, BATCH_BUCKET_SIZE)] prefill_batch_size_list = [batch for batch in range(PREFILL_BATCH_BUCKET_SIZE, max_prefill_batch_size, PREFILL_BATCH_BUCKET_SIZE)]
prefill_batch_size_list.append(max_prefill_batch_size) prefill_batch_size_list.append(max_prefill_batch_size)
prefill_seqlen_list = [seq for seq in range(PAD_SEQUENCE_TO_MULTIPLE_OF, max_input_length, PAD_SEQUENCE_TO_MULTIPLE_OF)] prefill_seqlen_list = [seq for seq in range(PAD_SEQUENCE_TO_MULTIPLE_OF, max_input_length, PAD_SEQUENCE_TO_MULTIPLE_OF)]
prefill_seqlen_list.append(max_input_length) prefill_seqlen_list.append(max_input_length)