# Conner\_Liu - Machine Learning Pathway

**URL:** <http://staging.stemaway.com/t/conner-liu-machine-learning-pathway/5640>\
**Category:** Machine Learning Pipeline (NLP)Self-Assessments\
**Created:** [August 6, 2020, 6:35am UTC](http://staging.stemaway.com/t/conner-liu-machine-learning-pathway/5640 "2020-08-06T06:35:20Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Conner\_Liu](https://d3rzmptxz7fqzy.cloudfront.net/user_avatar/staging.stemaway.com/conner_liu/32/1422_2.png) [@Conner\_Liu](http://staging.stemaway.com/u/Conner_Liu)\
**Post date:** [August 6, 2020, 6:35am UTC](http://staging.stemaway.com/t/conner-liu-machine-learning-pathway/5640/1 "2020-08-06T06:35:20Z")

</div>

Technical Area  
Scraped thousands of articles  
Put articles into tf-idf model

Tools  
Jupyter notebook  
BeautifulSoup  
BERT  
Selenium

Soft Skills  
Broke down accomplishments in to presentable pieces

Meetings  
7/21/20  
7/31/20  
8/4/20

Goals for upcoming week  
Solve a couple last bugs for tf-idf  
Start the creation of BERT model

Tasks Done  
During this week, I scraped data from the forum provided and researched the general outline of the BERT model. Furthermore, I did comprehensive research on word embeddings and TF-IDF, deciding to try the implementation of TF-IDF on the scraped data. I struggled with the implementation of TF-IDF, but I think I am near crossing that hurdle. This week has taught me the value of self-motivation.

---

<div class="post-metadata">

**Author:** ![Conner\_Liu](https://d3rzmptxz7fqzy.cloudfront.net/user_avatar/staging.stemaway.com/conner_liu/32/1422_2.png) [@Conner\_Liu](http://staging.stemaway.com/u/Conner_Liu)\
**Post date:** [August 15, 2020, 11:12pm UTC](http://staging.stemaway.com/t/conner-liu-machine-learning-pathway/5640/2 "2020-08-15T23:12:31Z")

</div>

#DistilBert, #time-management, #data-mining, #BERT

Technical Area  
Learned the basics of BERT and its components  
Understood the variation of BERT, DistilBERT

Tools  
GitHub  
Jupyter Notebook  
Google Collab  
Python

Achievements:  
Used data scraped by the different groups to create a DistilBERT model  
Meetings attended:  
2 meetings during Week 4

Goals for the Upcoming Week  
Try to improve the BERT model  
Try other variations of BERT
