Recent content by xeon123

I am having some difficulty in duplicating @haruspex's example. I really need some help.

I understand your example because you use numbers easy to calculate. I don't know how to answer my example. Can you tell me if my answers are correct?

My question is not related to processes per second. I have a process X which is faster than Y. I want to know how much X is faster than Y in %, and how much Y is slower than X in %. Eg., X is faster 10% than Y, and Y is slower something in % than X.

I have a really simple question that I can't figure it out. I have a process X that took 905 seconds to finish. Process Y took 950 seconds to finish. I want to know a) how much in % process X is faster than Y, and b) how much in % process Y is slower than X? I think that the answer is the...

I don't know where to put this question. So, I am trying in this forum. I have this dataset [1] that contains the duration of an execution with an input size from 1.5GB to 6GB. This dataset also contains the percentile 5%, 95%, and the 1st and 3rd quartile. I am thinking in creating a gnuplot...

I don't understand this sentence: The weights are the coefficients. I have named wrongly the coefficients. It is: time = (wbytesread*params[0]) + (wqueueacapacity*params[1]) + (wmaps*params[2]) + (wcpu_info*params[3]) + (wmem_info*params[4]) + coeffs[5] I think that the 6th param...

The full units that I am using is: ## seconds = c_1 * \% + c_2 * number + c_3 * bytes + c_4 * MHz + c_5 * MBytes ## So, the set that I use to get the coefficients contains, total time it took to run a process (seconds) cpu usage at the time (%) number of tasks to perform. That tasks are the...

if I have ## t(x,y,z)=c_1 x + c_2y + c_3z ##, where ##c_3## is the total memory of the system, and ##c_1## and ##c_2## are the same. Can I say that ##(t,y,z)## is seconds? The ##t(x,y,z)## that I calculate is around 150000 (some unit, maybe seconds), and the set that I use to get the...

I am trying to understand multivariate linear regression. I have a list of time that it took running processes based on several params, like % of cpu usage, and data read. Eg, I have a process that took 50 seconds to run, with a cpu usage of 70%, and the process read 10bytes of data. I have...

I mean that I use 8 independent variables to get y. y = Beta1*x1 + Beta2*x2 + Beta3*x3 + Beta4*x4 + Beta5*x5 + Beta6*x6 + Beta7*x7 + Beta8*x8 And when I calculate the Betas to get a predicted y, \hat{y}, some of them are negative, making \hat{y} negative.

I am using linear regression to predict 'y' based on 8 variables. With my example, most the Betas that I got are negative. So, y, the value to predict, is negative. To my data, y is a time in seconds, so I think it shouldn't be negative. I my example in python, and I want to know if y...

I have 2 measures that I am using to rank terms that I get them by prediction (using linear regression). They are the time to transmit X bytes and the time to compute the X bytes. I do the prediction if I execute in host A, B, and C. I add the 2 measures and rank the hosts. I think adding these...

In a linear regression with 1 independent variable, if X is always the same (let's call I am unlucky), but Y present different values for the same X, I still can find the coefficient of the straight line equation?

@chiro I thought the error (residual variance?) means the same in the simple or multiple linear regression? @Stephen I am using simple and linear regression to predict the time that a task will take before executing it. I know that there is an error between the estimation and the real value...

So maybe the question can be, how the error can be useful in the linear regression?