Skip to content

Update code and README #122

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

Merged
merged 2 commits into from
Jun 3, 2024
Merged

Update code and README #122

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6d3dab4
Update code and README
yanbing-j May 29, 2024
9e2e345
Add the new column to track the device
yanbing-j Jun 3, 2024
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
6 changes: 3 additions & 3 deletions experiments/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,9 +55,9 @@ For GPU,
- $ pip install https://download.pytorch.org/whl/nightly/cu121/torch-2.2.0.dev20231117%2Bcu121-cp310-cp310-linux_x86_64.whl
- $ pip install https://download.pytorch.org/whl/nightly/cu121/torchvision-0.17.0.dev20231117%2Bcu121-cp310-cp310-linux_x86_64.whl
For CPU,
- $ pip install https://download.pytorch.org/whl/nightly/cpu/torch-2.4.0.dev20240509%2Bcpu-cp310-cp310-linux_x86_64.whl
- $ pip install https://download.pytorch.org/whl/nightly/cpu/torchvision-0.19.0.dev20240509%2Bcpu-cp310-cp310-linux_x86_64.whl
- $ pip install triton
- $ pip install https://download.pytorch.org/whl/nightly/cpu/torch-2.4.0.dev20240530%2Bcpu-cp310-cp310-linux_x86_64.whl
- $ pip install https://download.pytorch.org/whl/nightly/cpu/torchvision-0.19.0.dev20240530%2Bcpu-cp310-cp310-linux_x86_64.whl
- $ install triton based on https://github.com/triton-lang/triton?tab=readme-ov-file#quick-installation

$ git clone https://github.com/cpuhrsch/segment-anything.git
$ cd segment-anything
Expand Down
25 changes: 18 additions & 7 deletions experiments/eval_combo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import math
import segment_anything_fast
import time
import resource

torch._dynamo.config.cache_size_limit = 50000

Expand Down Expand Up @@ -257,7 +258,10 @@ def profile_top_runner(fn, *args, **kwargs):
torch.profiler.ProfilerActivity.CUDA],
record_shapes=True) as prof:
result = fn(*args, **kwargs)
print(prof.key_averages().table(sort_by="self_cuda_time_total", row_limit=-1))
if torch.cuda.is_available():
print(prof.key_averages().table(sort_by="self_cuda_time_total", row_limit=-1))
else:
print(prof.key_averages().table(sort_by="self_cpu_time_total", row_limit=-1))
return result


Expand Down Expand Up @@ -444,15 +448,22 @@ def run(
batch_ms_batch_size = (avg_ms_per_img * num_images) / num_batches / batch_size

mIoU = calculate_miou(results, mask_debug_out_dir, True, cat_id_to_cat)
max_memory_allocated_bytes = torch.cuda.max_memory_allocated()
_, total_memory = torch.cuda.mem_get_info()
max_memory_allocated_percentage = int(100 * (max_memory_allocated_bytes / total_memory))
max_memory_allocated_bytes = max_memory_allocated_bytes >> 20
if torch.cuda.is_available():
max_memory_allocated_bytes = torch.cuda.max_memory_allocated()
_, total_memory = torch.cuda.mem_get_info()
max_memory_allocated_percentage = int(100 * (max_memory_allocated_bytes / total_memory))
max_memory_allocated_bytes = max_memory_allocated_bytes >> 20
else:
import psutil
total_memory = psutil.virtual_memory().total
max_memory_allocated_bytes = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
max_memory_allocated_percentage = int(100 * (max_memory_allocated_bytes / (total_memory >> 10)))
max_memory_allocated_bytes = max_memory_allocated_bytes >> 10

if print_header:
print(",".join(["sam_model_type", "batch_size", "memory(MiB)", "memory(%)", "img_s(avg)", "batch_ms(avg)/batch_size", "mIoU", "use_compile",
print(",".join(["device", "sam_model_type", "batch_size", "memory(MiB)", "memory(%)", "img_s(avg)", "batch_ms(avg)/batch_size", "mIoU", "use_compile",
"use_half", "compress", "epilogue_fusion_first", "use_compile_decoder", "use_nested_tensor", "use_rel_pos", "pad_input_image_batch", "num_workers", "num_batches", "num_images", "profile_path", "memory_path"]))
print(",".join(map(str, [sam_model_type, batch_size, max_memory_allocated_bytes, max_memory_allocated_percentage, img_s, batch_ms_batch_size, mIoU, use_compile,
print(",".join(map(str, [device, sam_model_type, batch_size, max_memory_allocated_bytes, max_memory_allocated_percentage, img_s, batch_ms_batch_size, mIoU, use_compile,
use_half, compress, epilogue_fusion_first, use_compile_decoder, use_nested_tensor, use_rel_pos, pad_input_image_batch, num_workers, num_batches, num_images, profile_path, memory_path])))


Expand Down
Loading