-
Question: are you planning to implement variational inference? I.e. using a prior on each component to discover the correct number needed for a data set. As I understand this doesn't necessarily incre…
-
Hi,
I am using Mac OSX 10.10 and Python 2.7.3 and I am trying to install scikit-learn without success. I am getting an error message (during installation with sudo pip install -U scikit-learn), pleas…
-
After fitting, is there a way to evaluate the probability of a new datapoint having come from the model?
-
The COSMOS galaxy selection for Great3 included a lot of different cuts from the full catalog. The code where they do this is [here](https://github.com/barnabytprowe/great3-public/blob/master/great3s…
-
I am browsing Ivezic's book (Statistics, Data Mining and Machine Learning in Astronomy, pages 257-264) for density estimation options. K nearest neighbors and Gaussian Mixture Models are both implemen…
-
https://github.com/scikit-learn/scikit-learn
作業ログは、参加者が、このイベントを通して得るものをより増やすために重要になります。なぜなら作業ログがメンターからのフィードバックをより充実させるからです。
作業ログを元にメンターと「ふりかえり」をするタイミングがあります。「ふりかえり」では次のことをします。
- 参加者は作業ログの内容を順にメンターに説…
-
Right now, it seems to be the log of the standard deviation in the last 4 injection heats. The assumption would be that the last four injections are all the same, since they're dilution/mechanical hea…
-
Leaving all the parameters free to wander is useless and dangerous, sometimes LT crashes (through emcee or the optimizer) on problems where it shouldn't. I'd lie the following priors to be implemented…
-
**Reported by rcurtin on 16 Dec 41738404 07:47 UTC**
The current solution for just printing models in some random format to a file is not exactly optimal; we can do better.
It seems as though perhaps…
-