xob | lightweight overlay volume bar for the X | Video Utils library

This [allocatable] is not standard Fortran.

Instead of

You declared i and length as char which cannot represent numbers above 127. They will overflow with bigger than that values. Make them short or int for example.

This problem is caused by the clipping in the function definition. The two minimization methods work fundamentally different and, therefore, react very differently to this clipping. Here minimize is used with Nelder-Mead, which is a gradient free method. The algorithm, hence is not calculating numerical gradients and not estimating any Jacobians. In contrasts the least-squares, which is eventually called by curve_fit, does exactly this. However, approximating a gradient and from this any Jacobian is somewhat questionable if the function is not continuous. As mentioned before, this discontinuity is introduced by the np.clip. When removed, one can easily see, that the P0 guess is not as good as it seems with clipping included. The curve_fit does converge with increased maxfev=5000, though, while the minimize immediately fails when changing the method to method='CG'. To see the algorithms difficulties one may try to provide manually the jac.

Some notes: 1) Concerning the clipping it might be a good idea to remove data that is clipped, such the according problem is avoided. 2) Looking at the covariance matrix the error of n and the correlation with the other values is extremely high.

So here is what I get from

In your function, xobs seems to be a vector, you are using xobs[nrow(xobs)] but class vector do not have rows, try:

simplexml_load_[string|file]() will consume the wrapper tag, in your case the cteProc tag.

This means that you don't need to look inside that object to access the other objects within.

To get to CNPJ, it would look like:

$cnpj = $xmldata->CTe->infCte->emit->CNPJ;

The denominator has a misplaced parenthesis:

NME=((np.sum(abs(xdiff)))/(np.sum(abs(np.mean(xobs))-(xobs)))) should be NME=((np.sum(abs(xdiff)))/(np.sum(abs(np.mean(xobs)-(xobs)))))

The absolute value was only affecting the mean value, and thus, the denominator was wrong (very close to zero in your case).

The posted code works fine. It reads 527 bytes from the file.

If you run this PHP script in the browser, the part that you think it misses () looks like an HTML element and it is not displayed by the browser.

Check the Page Source (find in in the menu or in the Developer Tools) to see the exact output produced by the script.

Or, even better, use the PHP header() function to tell the browser the content generated by the script is text, not HTML: