v0.5.0

[angela97lin]

v0.5.0 Oct. 29, 2019

Enhancements

• Added basic one hot encoding Added basic one-hot encoding #73
• Use enums for model_type Use enums for model_type instead of magic string. #110
• Support for splitting regression datasets Added support for splitting regression sets #112
• Auto-infer multiclass classification Automatically infer multiclass based off of objective #99
• Added support for other units in max_time Added support multiple units of time for max_time #125
• Detect highly null columns guardrail Detect highly null columns #121
• Added additional regression objectives Added Additional Regression Objectives #100

Fixes

• Reordered describe_pipeline to make main objective first column Reordered columns in describe_pipeline to make main objective first column #94
• Added type check for model_type adding type check for model_type #109
• Fixed s units when setting string max_time Fix for 's' units when setting string max_time #132
• Fix objectives not appearing in API documentation Fixed new objectives not appearing in API documentation #150

Changes

• Reorganized tests Reorganize tests #93
• Moved logging to its own module moving logging into its own util module #119
• Show progress bar history Show progress bar history and changes for docs #111
• Using cloudpickle instead of pickle to allow unloading of custom objectives Address loading in pipeline that uses unloaded custom objective class code #113
• Removed render.py Removed render.py #154

Documentation Changes

• Update release instructions Update release.md with correct instructions #140
• Include additional_objectives parameter Updated documentation to include additional_objectives param #124
• Added Changelog Added Changelog #136

Testing Changes

• Code coverage Added codecov #90
• Added CircleCI tests for other Python versions Adding CircleCI tests for other versions of Python #104
• Added doc notebooks as tests Added notebooks as tests #139
• Test metadata for CircleCI and 2 core parallelism Added test metadata and 2 core parallelism #137

[codecov]

Codecov Report

Merging #163 into master will not change coverage.
The diff coverage is 100%.

@@           Coverage Diff           @@
##           master     #163   +/-   ##
=======================================
  Coverage   95.52%   95.52%           
=======================================
  Files          57       57           
  Lines        1498     1498           
=======================================
  Hits         1431     1431           
  Misses         67       67

Impacted Files	Coverage Δ
evalml/__init__.py	100% <100%> (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 8a9d550...3fdcda1. Read the comment docs.

[angela97lin]

Note: just followed the release instructions, but do we want to condense / group the PRs in a sensible way instead of just listing them as is?

[kmax12]

LGTM