我当时经常说:“2的幂是有魔力的”。
当指定一个缓冲区大小时,该值必须是2的幂。
When you specify a buffer size, the value must be a power of 2.
传输较小的数据必须是2的幂,并是自然对齐的。
Transfers smaller than that must be a power of two and be naturally aligned.
它的确有一个隐藏的假设:ALIGN是2的幂。
It does have the hidden assumption that ALIGN is a power of two.
并由此得出幂和问题的一些新的性质。
Consequently we obtain some new proposition for the power sum problem.
我们需要知道求幂运算的逆运算的难度。
We need to know how hard it is to reverse the exponentiation.
两试样熔体的剪切流动大体上服从幂律。
The shear flow of two melts approximately obeyed the power law.
某一数或量的三次幂。
九的六次幂是多少?。
计算结果表明,幂律方程对本体系最为适用。
The results show that index-law equation is most suitable for this system.
最后还证明了幂等算子和它的伴随是相似的。
In the end we prove that idempotent operator is similar to its adjoint.
可以看到,对于比较小的2的幂次方,浮点数是非常精确的。
As you can see, floats are pretty accurate for small powers of two.
目的研究一类重要的幂等元半环,即乘法带半环。
Aim To study a very important class of idempotent semiring, so-called multiplicative band semirings.
幂等元与本原幂等元在环中有非常重要的地位与作用。
Idempotents and primitive idempotents have very important station in the ring.
GET方法应该是等幂的,意味着它可以安全地重复。
The GET method should also be idempotent, meaning that it can be safely repeated.
我们证明了该算法的收敛定理并把它与幂方法作了比较。
We prove the convergence theory for the algorithm and compare the method with Power method.
如果它是有限和非空的,则它必须包含至少一个幂等元。
If it is finite and nonempty, then it must contain at least one idempotent.
根据相位噪声功率谱的幂律模型,近似地模拟了相位噪声。
Flicker noise is approximately simulated by power law model of phase noise spectral density.
基于遗传算法求得矩阵的最大特征值,并且跟幂法进行比较,得到了较好的结果。
With genetic algorithms compared with the power method, a better result is gained.
基于遗传算法求得矩阵的最大特征值,并且跟幂法进行比较,得到了较好的结果。
With genetic algorithms compared with the power method, a better result is gained.
应用推荐