Added details to quickstart documentation on missing r-squared value #473
Conversation
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
Syncing from original
Syncing latest from DistrictDataLabs head.
bbengfort
reviewed
Jun 11, 2018
docs/quickstart.rst
Outdated
|
|
||
| Finally the residuals are colored by training and test set. This helps us identify errors in creating train and test splits. If the test error doesn't match the train error then our model is either overfit or underfit. Otherwise it could be an error in shuffling the dataset before creating the splits. | ||
|
|
||
| Because our coefficient of determination for this model is 0.328, let's see if we can fit a better model using *regularization*, and explore another visualizer at the same time. | ||
| Along with generating the residuals plot, we also measured the peformance, or "scored" our model on the test data above: ``visualizer.score(X_test, y_test)``. Because we used a Linear Regression model, the `scoring consists of finding the R-squared value of the data <http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html#sklearn.linear_model.LinearRegression.score>`_, which is a statistical measure of how close the data are to the fitted regression line. The R-squared value of any model may vary slightly between prediction/test runs, however it should generally be comparable. In our case, the R-squared value for this model was only 0.328, suggestion that linear correlation may not be the most appropriate to use for fitting this data. Let's see if we can fit a better model using *regularization*, and explore another visualizer at the same time. |
There was a problem hiding this comment.
minor typos:
peformance --> performance
Linear Regression --> LinearRegression or "linear regression".
was only 0.328, suggestion --> suggesting
also, how about "we also measured the performance by "scoring" our model on the test data, e.g. the code snippet visualizer.score(X_test, y_test)."?
There was a problem hiding this comment.
Thanks for the review! I'll make the changes and re-commit.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR is in response to #406
Added some more details and explanation on where the value was coming from based on which we are determining which model to use. Linked to the source sklearn .score() documentation rather than diving into details on the quickstart page.