Summary of grad mentor/research group meeting:

• We had our first technical grad mentor meeting and our preparation evidently showed that we did not know enough about what we are doing yet.
  • We tried explaining the text classification model in PyTorch’s tutorials section that we have been studying
  • However, Yujian’s deconstruction of our explanation showed us where the many flaws in our knowledge were, especially pertaining to the mathematical intuition behind our model
• Goals from here on are to develop our own model from bottom up rather than top down, and understand the underlying math

Mentor Research Meeting #2

Summary of team meeting:

• Friday:
  • Worked on discussing our understandings of the text classification model and then worked on slides to present at research meeting
• Monday:
  • Discussed understandings of bag of words, tf - idf, and Naive Bayes’
  • We came up with a prototype outline of how to code a straightforward text classifier using tf - idf and pseudo Naive Bayes’ predictions

Summary of meeting with Chinmay:

<Starts in the second half of the Fall quarter> <Any new insights from discussions with Chinmay, and changes in the task distribution, challenges that Chinmay helped you resolved, refined versions of your questions for your mentors, anything that is still blocking you, any other concerns>

Weekly goals:

• [x] Attend research team meeting and record attendance
• [ ] Attend CS165B
• [x] Complete weekly assignments
• [x] Work on HG News text classifier model
  • [x] Understand how we should represent our words numerically
  • [x] Find an algorithm or model that can accomplish our task
  • [ ] Code prototype of news article text classifier
  • [ ] Code final draft of news article text classifier
  • [ ] Understand deeply the mathematics behind our model

Accomplishments:

<List of things other than the above goals that you completed in the week>

Details of how you accomplished your goals:

Wednesday(2 hours) for assignments:

Reflection 3