[mlpack-git] [mlpack] Perceptron documentation (#594)

[Ryan Curtin]

Hi Arman,

Thanks for the kind words about mlpack.  But it sounds like maybe our documentation is not good enough.  Do you think that if I wrote a tutorial called "Data formats in mlpack" which detailed the types of input that mlpack will expect, this would clarify the issue?

As a quicker response to the problem at hand, `training_data.csv` should be a CSV with one point per row.  So the three-point dataset with the points (1, 0), (2, 2), and (-1, -1) would be

```
1, 0
2, 2
-1, -1
```

and then `training_labels.csv` should contain one label per line corresponding to each training point.  So if, e.g., the label of the first two points was 0 and the label of the second point was 1, then the `training_labels.csv` file should be

```
0
0
1
```

The last file will be an output file that saves the parameters of the perceptron.

It looks like you are using an old version of mlpack (1.0.12?).  I think maybe the documentation is a little clearer in the newest release (2.0.1), so I'd recommend upgrading.  Also, the formats that you can save the model in are expanded (you can save as xml, bin, or txt).

I hope this is helpful... let me know if I can clarify further.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/issues/594#issuecomment-201270071
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160325/976bb4c9/attachment.html>