|Hw-3+ final Grades are updated on GR++|
Final grade = Round(Max(Avg(Hw_1,Hw_2,Hw_3),0.35*Hw_1+0.35*Hw_3+0.3*Hw_2))
For Hw-3 remarks see: http://www.cs.technion.ac.il/~ronbeg/236756/Hw3/<your id number>.pdf
Grades at September
|Grades will be published at the beginning of September|
Enjoy your summer vacation!
Length of HW 3 document
|Recall the length of your solution is limited to 5 pages.|
However, you may add an appendix section (beyond this limitation) for presenting complementary materials.
|Hw-3 new due date: August 9th (one week postponement)|
Note you have a full control on the experimental settings; you may consider the assignment's time limit when defining the experiment protocol.
Active learning slides + Hw3
|handout - on site.|
hw.3 - plz make sure your experiment protocol is *valid* (in the sense discussed in the recitation)
Hw3 - updated "matlab_to_weka" script
Supporting logical labels (data_5) with the following code:
Y = +Y;
Hw3 text - new clarifying remarks
|Hw3 text has been updated with a diagram and a few clarifying remarks|
Hw3 - remarks
| - Start working on this assignment! (You should write some code + Experiments consumes time)|
- Exceptionally good works will gain a bonus (you may gain more than 100 points)
- Added a script for translating a matlab dataset to weka textual format (ARFF)
No tutorials at 12,19/07 + Hw2-grades
| - Next two tutorials are canceled (12 and 19th to July) due to studying in "MATKONET" Sunday and Monday|
- Hw2 grades are published (see GR++).
The relevant remarks are located at:
(replace XXX with your id number)
NOTE: Determining the quality of C4.5 classifier = 15 points.
Hw3 + updated version of "Linear models" handout
Hw2 + "Linear models" handout
Empirical Evaluation slides + First assignment
|Empirical Evaluation slides: on site.|
- use GR++ to view your grade
- read the relevant remarks at:
http://www.cs.technion.ac.il/~ronbeg/236756/<your id number>.pdf
Statistical Machine Learning Handouts - on site
"Find a Partner" Button
|is enabled (on course's website)|
Matlab open source clones: 2 links
|Added links to two major Matlab "clones" (see Links).|
Note they do not offer 100% compatibility with Matlab's "m-files".
Essentially, both clones are interpreted, matrix-based programming
languages. "They share with Matlab:
1. The use of matrices as a fundamental data type.
2. Built-in support for complex numbers.
3. Powerful built-in math functions and extensive function libraries.
4. Extensibility in the form of user-defined functions."
"Statistical Machine Learning" Handout
|Will be published before the upcoming tutorial (after Passover). Send Ron an email|
for a copy of the current (preliminary) version.
|on site; with most relevant questions (and few answers); will be|
updated "on demand"; plz. read it before emailing a question.
|on site; see "course material."|
|On site's 'syllabus'+ 'general info.' have been updated.|
We wish you all a happy new semester :)