Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Allow batch ratio to be tuned #1648

Closed
wants to merge 1 commit into from
Closed

Allow batch ratio to be tuned #1648

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 15 additions & 9 deletions magic_pdf/model/doc_analyze_by_custom_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -174,20 +174,26 @@ def doc_analyze(
if torch_npu.npu.is_available():
npu_support = True

override_batch_ratio = int(os.getenv("MINERU_OVERRIDE_BATCH_RATIO", 0))

if torch.cuda.is_available() and device != 'cpu' or npu_support:
gpu_memory = int(os.getenv("VIRTUAL_VRAM_SIZE", round(get_vram(device))))
if gpu_memory is not None and gpu_memory >= 8:
batch_ratio = 0

if 8 <= gpu_memory < 10:
batch_ratio = 2
elif 10 <= gpu_memory <= 12:
batch_ratio = 4
elif 12 < gpu_memory <= 16:
batch_ratio = 8
elif 16 < gpu_memory <= 24:
batch_ratio = 16
if override_batch_ratio > 0:
batch_ratio = override_batch_ratio
else:
batch_ratio = 32
if 8 <= gpu_memory < 10:
batch_ratio = 2
elif 10 <= gpu_memory <= 12:
batch_ratio = 4
elif 12 < gpu_memory <= 16:
batch_ratio = 8
elif 16 < gpu_memory <= 24:
batch_ratio = 16
else:
batch_ratio = 32

if batch_ratio >= 1:
logger.info(f'gpu_memory: {gpu_memory} GB, batch_ratio: {batch_ratio}')
Expand Down
Loading