Improved Vivekn sentiment analysis training performance

[saif-ellafi]

Description

• Improved vivekn training performance
• Fixed dependency parser tests
• Separated wordCount from vivekn wordCount
• New params to customize tokens from corpus

Motivation and Context

Vivekn Sentiment analysis was causing a bottleneck on training and memory consumption. This helps there by reducing the amount of redudant calls during word count.

How Has This Been Tested?

This has been tested utilizing a third party sentiment analysis corpora, works best together with the broadcast and rdd object serialization.

Screenshots (if appropriate):

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• Code improvements with no or little impact
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING page.
• I have added tests to cover my changes.
• All new and existing tests passed.