Comparing
Abineshabee:fix/array-api-reduction-returns-0d-array
(335edc9
) withmain
(3589a7c
)-21%
- Improvement
- 1
- Regression
- 1
- Untouched
- 338
Benchmarks
340 total
test_index_slice[side=100-rank=2-format='gcxs']
benchmarks/test_benchmark_coo.py-21%
1.7 ms2.2 ms
test_index_fancy[side=100-rank=1-format='coo']
benchmarks/test_benchmark_coo.py+24%
1.4 ms1.1 ms
test_elemwise[side=1000-rank=2-format='coo'-add]
benchmarks/test_benchmark_coo.py+3%
16 ms15.5 ms
test_index_fancy[side=100-rank=3-format='coo']
benchmarks/test_benchmark_coo.py+3%
3.1 ms3 ms
test_elemwise[side=100-rank=3-format='coo'-mul]
benchmarks/test_benchmark_coo.py+2%
7.6 ms7.5 ms
test_index_fancy[side=1000-rank=2-format='coo']
benchmarks/test_benchmark_coo.py+2%
4.1 ms4 ms
test_tensordot[left_index=1-right_index=1-left_format='coo'-right_format='dense'-m=10-n=10-p=50-q=50-COO]
benchmarks/test_tensordot.py+1%
2.6 ms2.6 ms
test_tensordot[left_index=1-right_index=2-left_format='coo'-right_format='coo'-m=50-n=10-p=50-q=50-ndarray]
benchmarks/test_tensordot.py+1%
6 ms5.9 ms
test_tensordot[left_index=1-right_index=1-left_format='coo'-right_format='dense'-m=50-n=10-p=50-q=50-COO]
benchmarks/test_tensordot.py+1%
6.6 ms6.5 ms
test_matmul[m=1000-n=1000-p=200-format='coo']
benchmarks/test_benchmark_coo.py+1%
9.7 ms9.6 ms
test_tensordot[left_index=1-right_index=2-left_format='coo'-right_format='coo'-m=50-n=10-p=50-q=50-COO]
benchmarks/test_tensordot.py+1%
6.1 ms6.1 ms
test_matmul[m=1000-n=500-p=500-format='coo']
benchmarks/test_benchmark_coo.py+1%
13 ms12.9 ms
test_elemwise_broadcast[side=1000-format='coo'-mul]
benchmarks/test_benchmark_coo.py0%
78.4 ms78 ms
test_elemwise_broadcast[side=100-format='gcxs'-add]
benchmarks/test_benchmark_coo.py0%
18.8 ms18.7 ms
test_matmul[m=200-n=1000-p=500-format='coo']
benchmarks/test_benchmark_coo.py0%
6.4 ms6.3 ms
test_matmul[m=1000-n=200-p=1000-format='gcxs']
benchmarks/test_benchmark_coo.py0%
14.5 ms14.5 ms
test_matmul[m=1000-n=1000-p=1000-format='coo']
benchmarks/test_benchmark_coo.py0%
41 ms40.9 ms
test_elemwise[mul-side=500-SciPy]
benchmarks/test_elemwise.py0%
477.5 µs475.9 µs
test_tensordot[left_index=1-right_index=1-left_format='coo'-right_format='dense'-m=50-n=20-p=50-q=50-COO]
benchmarks/test_tensordot.py0%
11.4 ms11.4 ms
test_matmul[m=200-n=200-p=1000-format='gcxs']
benchmarks/test_benchmark_coo.py0%
4 ms3.9 ms
test_matmul[m=500-n=200-p=500-format='coo']
benchmarks/test_benchmark_coo.py0%
4.9 ms4.9 ms
test_index_slice[side=1000-rank=2-format='gcxs']
benchmarks/test_benchmark_coo.py0%
3.6 ms3.6 ms
test_elemwise[side=1000-rank=2-format='gcxs'-mul]
benchmarks/test_benchmark_coo.py0%
11.3 ms11.2 ms
test_tensordot[left_index=1-right_index=1-left_format='coo'-right_format='dense'-m=10-n=20-p=20-q=50-COO]
benchmarks/test_tensordot.py0%
2.3 ms2.3 ms
test_matmul[m=500-n=1000-p=500-format='coo']
benchmarks/test_benchmark_coo.py0%
11.9 ms11.9 ms
Commits
Click on a commit to change the comparison rangeBase
main
3589a7c
-54.75%
fix: return 0-D array instead of scalar in reduce() and var()
9304ca7
2 months ago
by Abineshabee+12.1%
fix: update mean() docstring and fix einsum performance regression
4efc908
1 month ago
by Abineshabee+19.04%
fix: use fill_value check instead of todense() in nanmax/nanmin NaN detection
694ff82
1 month ago
by Abineshabee+2.8%
Update fixes and benchmark analysis
e059a6c
1 month ago
by Abineshabee-14.21%
fix: resolve doctest failures and ensure 0-D array reduction outputs
e124b76
1 month ago
by Abineshabee+14.12%
refactor: extract _contains_nan() helper, simplify nanmax/nanmin NaN check
335edc9
1 month ago
by Abineshabee