Add verbose to load_fraud

[gsheni]

Pull Request Description

• Sometimes a user may not want to print verbose output while doing load_fraud. This PR makes the change to allow for optionally changing that.

[codecov]

Codecov Report

Merging #560 into master will not change coverage by %.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #560   +/-   ##
=======================================
  Coverage   98.83%   98.83%           
=======================================
  Files         115      115           
  Lines        4297     4297           
=======================================
  Hits         4247     4247           
  Misses         50       50           

Impacted Files	Coverage Δ
evalml/demos/fraud.py	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 79a2358...f2f1e5c. Read the comment docs.

[dsherry]

Looks good to me! Thanks.

A thought: should we also update the demos (and therefore the load_data verbose option) to use our logger? evalml/utils/logger.py.

[gsheni]

Looks good to me! Thanks.

A thought: should we also update the demos (and therefore the load_data verbose option) to use our logger? evalml/utils/logger.py.

Yah, we should. That way you can turn it to the level you want:

logging.basicConfig(level=your_level) #debug, info, warning, error, critical

[dsherry]

@gsheni cool, yeah. Our logger is very simple right now and doesn't support verbosity levels, but we plan to upgrade it to use the logging package as you mentioned: #460. At that point we could support all sorts of features including different verbosity. So while we can't do that logging level setting you mentioned yet, we'll be able to soon!