-
Notifications
You must be signed in to change notification settings - Fork 47
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Create a template and directory for algorithms (#28)
* Create a template and directory for algorithm See #27 * Add notebook nomenclature example
- Loading branch information
Showing
3 changed files
with
1,357 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,19 @@ | ||
# Notebooks implementing various sklearn classifiers | ||
|
||
This directory hosts notebooks created in response to [Issue #27](https://github.com/cognoma/machine-learning/issues/27#issuecomment-238255454 "cognoma/machine-learning: Claim an sklearn algorithm to implement and troubleshoot") where team members can claim and implement an algorithm. | ||
|
||
The recommended approach is to copy `SGDClassifier-template.ipynb` to a file named `algorithm-username.ipynb` and modify the newly created file. For example, @brankaj who's [implementing](https://github.com/cognoma/machine-learning/issues/27#issuecomment-238132718) `sklearn.linear_model.LassoCV` would name her notebook `LassoCV-brankaj.ipynb`. | ||
|
||
Implementations are encouraged to use grid search to find optimal algorithm hyperparameters. Finally, make sure you're using the `cognoma-machine-learning` conda environment described in the [README](https://github.com/cognoma/machine-learning/tree/5eadf0309a409b08fc17d81b36b9c4bcd7acb916#environment). | ||
|
||
Once you've finished your notebook, you should run (from this directory): | ||
|
||
```sh | ||
# Restart and run all cells in your notebook | ||
jupyter nbconvert --inplace --execute --ExecutePreprocessor.timeout=-1 algorithm-username.ipynb | ||
|
||
# Export notebook to a .py script for diff viewing | ||
jupyter nbconvert --to=script --FilesWriter.build_directory=scripts algorithm-username.ipynb | ||
``` | ||
|
||
Then you can submit a pull request. |
Oops, something went wrong.