delta/python-packages/numpy.git/numpy/lib/polynomial.py, branch v1.12.0rc1

DOC: Fix erroneous return type description for np.roots.

2016-09-23T10:46:23+00:00

np.roots() does not always return complex roots, the
roots of a polynomial depend its coefficients and
therefore may be in the real or complex domain.

e.g.
```
>>> (np.roots([1,2,3])).dtype
dtype('complex128')
>>> (np.roots([1,2,])).dtype
dtype('float64')
```

ENH: Add stacklevel to all (or almost all) our function calls

2016-09-02T14:39:17+00:00

DOC: Fixed documented dimension of return value

2016-08-26T16:21:02+00:00

The first dimension of the return value depends on the degree of the polynomial, not the number of elements being used in the fit.

TST: Test that polyfit raises if not enough data for cov estimate.

2016-06-15T18:04:39+00:00

Also a slight refactoring of np.polyfit.

BUG: Make sure we don't divide by zero in np.polyfit.

2016-06-15T17:23:20+00:00

This should fix the issue discussed at
https://mail.scipy.org/pipermail/numpy-discussion/2013-July/067076.html

Without the ValueError added here, polyfit can (and does) return
negative or nan variances with no warning.

Merge pull request #4073 from endolith/patch-2

2016-06-15T14:34:56+00:00

BUG: change real output checking to test if all imaginary parts are zero

DOC: Fix some incorrect RST definition lists

2016-05-12T00:47:47+00:00

DOC: Fix poly_val description for 'x' input

2015-12-23T00:24:19+00:00

Closes gh-6849.

DOC: Use print only as function when print_function is imported from future

2015-12-20T00:50:09+00:00

Closes gh-6863.

add clarification of weights to documentation

2015-11-13T02:17:54+00:00