▶ code
▼ output
▶ uv-logs
|
Cell: combine | 4.44s
|
Raw
# /// script
# requires-python = ">=3.10"
# dependencies = [
# "numpy",
# "torch==2.8.0",
# "kernels-benchmark-tools",
# "matplotlib",
# ]
#
# [tool.uv.sources]
# kernels-benchmark-tools = { path = "../../../../../tools", editable = true }
# ///
from kernels_benchmark_tools.core.visuals import generate_combined_results
# Map display names to uvnote environment variables
cache_env_map = {
"PyTorch LayerNorm": "UVNOTE_FILE_TORCH_LAYER_NORM_BENCHMARK",
"HF Kernels LayerNorm": "UVNOTE_FILE_HF_KERNELS_LAYER_NORM_BENCHMARK",
}
# Generate combined results with visualization
generate_combined_results(
cache_env_map=cache_env_map,
output_filename="layer_norm.jsonl",
svg_filename="latency.svg"
)
======================================================================
LOADING BENCHMARK DATA
======================================================================
✓ PyTorch LayerNorm : /__w/kernels-benchmarks/kernels-benchmarks/benches/layer_norm/impls/.uvnote/cache/4403c31e9bef6e648597b4fcc9cfdc402678aaa4f90636b74325f12d334214a3
✓ HF Kernels LayerNorm : /__w/kernels-benchmarks/kernels-benchmarks/benches/layer_norm/impls/.uvnote/cache/bd278151199f29b397d85857b87922edaa39a62623fb28e0465de47d6a3bac74
✓ Found PyTorch LayerNorm
Path: /__w/kernels-benchmarks/kernels-benchmarks/benches/layer_norm/impls/.uvnote/cache/4403c31e9bef6e648597b4fcc9cfdc402678aaa4f90636b74325f12d334214a3/layer_norm.jsonl
✓ Found HF Kernels LayerNorm
Path: /__w/kernels-benchmarks/kernels-benchmarks/benches/layer_norm/impls/.uvnote/cache/bd278151199f29b397d85857b87922edaa39a62623fb28e0465de47d6a3bac74/layer_norm.jsonl
======================================================================
Summary: 2 found, 0 skipped, 0 missing
======================================================================
COMBINED BENCHMARK SUMMARY
impl wl p50(ms) ok
hf_kernels_layer_norm LN_B16_S2048_D4096 0.83 True
hf_kernels_layer_norm LN_B16_S2048_D8192 1.65 True
hf_kernels_layer_norm LN_B16_S4096_D4096 1.65 True
hf_kernels_layer_norm LN_B16_S4096_D8192 3.25 True
torch_layer_norm LN_B16_S2048_D4096 0.83 True
torch_layer_norm LN_B16_S2048_D8192 1.68 True
torch_layer_norm LN_B16_S4096_D4096 1.61 True
torch_layer_norm LN_B16_S4096_D8192 3.33 True
GENERATING COMBINED VISUALIZATION
Loaded 8 records
✓ Visualization saved as latency.svg
Saved latency.png
✓ Visualization saved as latency.svg
✓ SVG visualization ready!
ANALYSIS COMPLETE
Total implementations analyzed: 2
Implementations included:
✓ PyTorch LayerNorm
✓ HF Kernels LayerNorm
▶ UV Install Logs