[mlpack-git] (blog) master: Keon, Week 1 (f99683c)

[gitdub at mlpack.org]

Repository : https://github.com/mlpack/blog
On branch  : master
Link       : https://github.com/mlpack/blog/compare/af0fd90961a51d6e99588ab0d604b710c5f8073b...f99683c01b5a3aaa1c2f5bf86d7358da37890674

>---------------------------------------------------------------

commit f99683c01b5a3aaa1c2f5bf86d7358da37890674
Author: Keon Kim <kwk236 at gmail.com>
Date:   Mon May 30 10:17:41 2016 +0900

    Keon, Week 1


>---------------------------------------------------------------

f99683c01b5a3aaa1c2f5bf86d7358da37890674
 content/blog/KeonWeekOne.md | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/content/blog/KeonWeekOne.md b/content/blog/KeonWeekOne.md
new file mode 100644
index 0000000..441fe3a
--- /dev/null
+++ b/content/blog/KeonWeekOne.md
@@ -0,0 +1,15 @@
+Title: Dataset and Experimentation Tools : Week-1 Highlights
+Date: 2016-05-26 12:09:00
+Tags: gsoc
+Author: Keon Kim
+
+The goal of the Dataset and Experimentation Tools project is to create utilities for dataset management in mlpack.
+The first week of GSoC went really quick. Here are some highlights I did:
+
+- Started implementing proof-of-concept for managing missing values. (more should be discussed)
+- Implemented CLI executable wrapper for data split and added split function for unlabeled data.
+- With Tham, discussed and planned about APIs for dataset management functions. (such as binarize, encoding, normalization, etc)
+
+Since I've been coding in JavaScript lately, it took me some time to quickly recap C++ and get used to the new environment, but it was really enjoyable.
+Next week, I am going to finalize working on missing data feature and analyze more into working with categorical features.
+