-2
$\begingroup$

From easy to difficult , which csv data sets can test ability of learning algorithm?

I find circular curl into inside

Where can download these data sets?

$\endgroup$
  • 4
    $\begingroup$ Your question is quite unclear. Can you elaborate a little more? $\endgroup$ – DuttaA Dec 8 '19 at 3:07
  • $\begingroup$ I think MNIST might be something you're looking for. You can't download this as a csv, but you can extract into a python list for ease of use yann.lecun.com/exdb/mnist $\endgroup$ – Recessive Dec 8 '19 at 7:46
  • $\begingroup$ @Recessive MNIST is not csv and its actually quite a bad dataset, accuracy is very high for almost all kinds of NNs to make any significant comment about it's effectiveness. $\endgroup$ – DuttaA Dec 8 '19 at 8:30
  • $\begingroup$ Expect to use csv data set in maple instead of python $\endgroup$ – Prince Martin Dec 8 '19 at 11:06
  • $\begingroup$ I find some tutorial using a recursive curl into center and a line fit through curl into center , where can find this csv data set or how to generate this difficult data to test? $\endgroup$ – Prince Martin Dec 8 '19 at 11:08
0
$\begingroup$

Not sure about where you can find datasets by difficulty, but I will concentrate on how you can generate your own spiral dataset.

To generate a spiral you just need to create a time vector $t$ (e.g., a column in excel with numbers from 0 to $T_{max}$) and then use the following formulas (for the next two columns):

$$x(t) = (r_0+v_rt)\cos(\omega t),\hspace{5pt} y(t) = (r_0+v_rt)\sin(\omega t),$$

where $v_r$ is the outward velocity of the spiral, and $\omega$ its angular velocity. You can choose these parameters as you like.

If you ignore the factor $r_0+v_rt$, you can notice that the formulas for the point $(x,y)$ correspond to the parametric equation of a circle. So, the factor simply changes linearly the size of this circle.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.