Added Mean Squared Logarithmic Error (MSLE) Loss Function

[ank426]

Describe your change:

I have added MSLE loss function which is very useful when dealing with regression problems involving skewed or large-value targets.

• Add an algorithm?
• Fix a bug or typo in an existing algorithm?
• Documentation change?

Checklist:

• I have read CONTRIBUTING.md.
• This pull request is all my own work -- I have not plagiarized.
• I know that pull requests will not be merged if they fail the automated tests.
• This PR only changes one algorithm file. To ease review, please open separate PRs for separate algorithms.
• All new Python files are placed inside an existing directory.
• All filenames are in all lowercase characters with no spaces or dashes.
• All functions and variable names follow Python naming conventions.
• All function parameters and return values are annotated with Python type hints.
• All functions have doctests that pass the automated testing.
• All new algorithms include at least one URL that points to Wikipedia or another similar explanation.
• If this pull request resolves one or more open issues then the description above includes the issue number(s) with a closing keyword: "Fixes #ISSUE-NUMBER".

[cclauss]

This directory of loss functions is less compelling than I thought because each is three lines or less of algorithmic code.

The files are about 100 lines long with comments, doctests, validation code, and other boilerplate. It takes a lot of clicking in and out of files to review all of the various loss functions.

Would we be better off making a single loss_functions.py that contains all these functions so that they can be studied side-by-side without all the opening and closing of files?

[tianyizheng02]

Would we be better off making a single loss_functions.py that contains all these functions so that they can be studied side-by-side without all the opening and closing of files?

Sounds good to me. Once we're done handling the existing loss function PRs we can open a new PR to consolidate all of the loss functions.

[ank426]

Would we be better off making a single loss_functions.py that contains all these functions so that they can be studied side-by-side without all the opening and closing of files?

Sounds good to me. Once we're done handling the existing loss function PRs we can open a new PR to consolidate all of the loss functions.

Hey so could you please accept the PR, or are there any changes I have to make?