The use of np.allclose introduced a severe performance penalty,
caused by the creation of two `np.array`s for each check.
Instead create and use a similar check which maintains tolerance
to floating point rounding, but operates only on scalars.
Adds a check to see if the s_squared value is near 0.
When the number was very near 0, a very small negative floating
point, the sqrt throws a 'math domain error', this prevents that
case.
So that algorithms that specifiy market_aware, days and delta
as args can be transitioned to just specifying a window_length kwarg.
Moving towards removing market_aware and delta completely.
Though the addition of tracking mulitple values in the window
is powerful, the changes broke behavior of existing algorithms
by changing method signatures and names.
So temporarily reverting these changes, to be pulled back in when
a way to have the multiple fields tracked with the existing API
is written, or a cutover of the API is figured out and determined.
When run over large amounts of data the use of ndict's gets and sets
become a large bottleneck, around 1/5th of the CPU time is spent
in ndict's __setattr__, __getattr__, etc.
By switching to an object for an event,
we reduce the penalty significantly.
Removes asserts that check for event being an ndict, as well as those
that assume a certain behavior of the __contains__ method for events.
The latest flake8 release in now 1.5, which pulls in pep8: 1.3.4a0
The upgrade pep8 has changes to what it picks up as lint.
Making code base compatible, so that new devs can install pep8
from PyPI and not have friction over the version difference.
Currently using these ignores in the config file:
```
[pep8]
ignore = E124,E125,E126
```
Ignoring these since they are difficult to squash while maintaining
an 80 char line length, and appear spurious.
Should address later.
Updates Travis config, README, and pip requirements to reflect change.
Eddie Hebert