Hi,
I had been going through HMM description and found the Gaussian HMM example
very helpful. However, the documentation does not provide an example the
discrete variant. I faced certain errors while writing my own and that led
me to known bugs and their work arounds. While the Rain/Sun/Cloudy exam
Someone on Cross-validated was able to give me an answer, and it is
something I should have seen. classify uses a uniform prior, while
ClassificationDisciminant uses an empirical prior. Problem solved.
On Mon, Feb 18, 2013 at 3:31 PM, David Reed wrote:
> Ahh, I have used Gist, but am absent m
Ahh, I have used Gist, but am absent minded today it seems.
https://gist.github.com/dvreed77/4980426
On Mon, Feb 18, 2013 at 3:23 PM, Andreas Mueller
wrote:
> On 02/18/2013 09:07 PM, David Reed wrote:
> > So I came up with this small test script to show the difference
> > between Matlab's class
On 02/18/2013 09:07 PM, David Reed wrote:
> So I came up with this small test script to show the difference
> between Matlab's classify function and Matlab's
> ClassificationDiscriminant function. It appears that for balanced
> classes, they perform identically, but when you unbalance them you
So I came up with this small test script to show the difference between
Matlab's classify function and Matlab's ClassificationDiscriminant
function. It appears that for balanced classes, they perform identically,
but when you unbalance them you get radically different answers. This
problem become
