Using pytest fixtures to improve benchmarking tests

[prady0t]

What type of PR is this? (check all applicable)

• 💾 Refactor
• 🪄 Feature
• 🐞 Bug Fix
• 🔧 Optimization
• 📚 Documentation
• 🧪 Test
• 🛠️ Other

Related issues

• Related issue Tracking issue for benchmarking improvements. #889

Checklist

• Code follows style guide
• Tests added
• Documented the changes

---

Please explain your changes below.

[prady0t]

@hameerabbasi, These were some places where fixtures could be used.

[codspeed-hq]

CodSpeed Performance Report

Merging #892 will degrade performances by 31.97%

_{Comparing prady0t:adding-fixtures (2029adf) with main (32ab23d)}

Summary

⚡ 1 improvements
❌ 1 regressions
✅ 257 untouched benchmarks
🆕 81 new benchmarks
⁉️ 81 dropped benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	test_index_fancy[side=100-rank=1-format='coo']	913.3 µs	1,342.4 µs	-31.97%
⚡	test_index_slice[side=100-rank=2-format='gcxs']	2.5 ms	2.3 ms	+10.72%
⁉️	test_matmul[m=1000-n=1000-p=1000-coo]	39.2 ms	N/A	N/A
🆕	test_matmul[m=1000-n=1000-p=1000-format='coo']	N/A	39.2 ms	N/A
🆕	test_matmul[m=1000-n=1000-p=1000-format='gcxs']	N/A	27 ms	N/A
⁉️	test_matmul[m=1000-n=1000-p=1000-gcxs]	27 ms	N/A	N/A
⁉️	test_matmul[m=1000-n=1000-p=200-coo]	9.1 ms	N/A	N/A
🆕	test_matmul[m=1000-n=1000-p=200-format='coo']	N/A	9.1 ms	N/A
🆕	test_matmul[m=1000-n=1000-p=200-format='gcxs']	N/A	10.2 ms	N/A
⁉️	test_matmul[m=1000-n=1000-p=200-gcxs]	10.2 ms	N/A	N/A
⁉️	test_matmul[m=1000-n=1000-p=500-coo]	20.7 ms	N/A	N/A
🆕	test_matmul[m=1000-n=1000-p=500-format='coo']	N/A	20.5 ms	N/A
🆕	test_matmul[m=1000-n=1000-p=500-format='gcxs']	N/A	21.2 ms	N/A
⁉️	test_matmul[m=1000-n=1000-p=500-gcxs]	21.2 ms	N/A	N/A
⁉️	test_matmul[m=1000-n=200-p=1000-coo]	13 ms	N/A	N/A
🆕	test_matmul[m=1000-n=200-p=1000-format='coo']	N/A	12.9 ms	N/A
🆕	test_matmul[m=1000-n=200-p=1000-format='gcxs']	N/A	14.3 ms	N/A
⁉️	test_matmul[m=1000-n=200-p=1000-gcxs]	14.3 ms	N/A	N/A
⁉️	test_matmul[m=1000-n=200-p=200-coo]	3.5 ms	N/A	N/A
🆕	test_matmul[m=1000-n=200-p=200-format='coo']	N/A	3.5 ms	N/A
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

[hameerabbasi]

LGTM, thanks @prady0t!