Skip to content

Improve performance of data processing module #135

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
Aug 29, 2025
Merged

Improve performance of data processing module #135

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
300c046
Standardizes configuration string formatting
LoserCheems Aug 29, 2025
f11ee36
Merge branch 'optimize-sparse-logic' of https://github.com/SmallDoges…
LoserCheems Aug 29, 2025
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
2 changes: 1 addition & 1 deletion benchmarks/backward_performance.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -864,7 +864,7 @@ def run_backward_performance_benchmark(test_type='all', num_runs=3, warmup_runs=

# Format configuration string
batch_size, num_heads, num_kv_heads, query_len, key_len, head_dim, keep_window_size, is_causal = config
config_str = f"B{batch_size}H{num_heads}K{num_kv_heads}Q{query_len}K{key_len}D{head_dim}W{keep_window_size}{'C' if is_causal else 'N'}"
config_str = f"B{batch_size} Hq{num_heads} Hkv{num_kv_heads} Q{query_len} K{key_len} D{head_dim} W{keep_window_size} {'C' if is_causal else 'N'}"
Copy link

Copilot AI Aug 29, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The parameter label 'K' is used twice in the configuration string - once for num_kv_heads and once for key_len. This creates ambiguity in the output. Consider using 'Kl' for key_len to differentiate it from the 'Hkv' label for num_kv_heads.

Suggested change
config_str = f"B{batch_size} Hq{num_heads} Hkv{num_kv_heads} Q{query_len} K{key_len} D{head_dim} W{keep_window_size} {'C' if is_causal else 'N'}"
config_str = f"B{batch_size} Hq{num_heads} Hkv{num_kv_heads} Q{query_len} Kl{key_len} D{head_dim} W{keep_window_size} {'C' if is_causal else 'N'}"

Copilot uses AI. Check for mistakes.

# Calculate averages and format results
sdpa_avg = f"{sum(results['sdpa_backward_times'])/len(results['sdpa_backward_times']):.2f}ms" if results['sdpa_backward_times'] else results['sdpa_backward_status']
Expand Down
6 changes: 4 additions & 2 deletions benchmarks/forward_performance.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -832,9 +832,11 @@ def run_performance_benchmark(test_type='all', num_runs=3, warmup_runs=2):
speedup_strs[impl_key] = "N/A"

# Format output with shorter config string
config_short = f" b{batch_size} h{num_heads} kv{num_kv_heads} q{query_len} k{key_len} d{head_dim} w{keep_window_size} "
config_short = f" B{batch_size} Hq{num_heads} Hkv{num_kv_heads} Q{query_len} K{key_len} D{head_dim} W{keep_window_size} "
if not is_causal:
config_short += "nc"
config_short += "N"
else:
config_short += "C"

# Add status icons
icons = ""
Expand Down