Friends: The One With Deep Learning

Join us for our 3rd community coding adventure in Deep Learning! Just bring your curiosity and get ready to meet our growing community 😀 We are using Deep Learning and SocialNLP to detect emotions in many episodes of the popular TV show, Friends!

Join Zoom Meeting:
https://us02web.zoom.us/j/84402592502?pwd=d1lVSkxQZE1sSGljR3dXaEZwYmNEdz09

Phone: +1 929 205 6099 US
Meeting ID: 844 0259 2502

Agenda:

• Introductions and get to know our community

• Deep Learning YouTube recordings, feel free to share and subscribe 😀
  https://bit.ly/deep-learning-tf
  https://bit.ly/deep-learning-tf-coding

• Deep Learning Adventures - Coding Presentation:
  https://docs.google.com/presentation/d/1XXSLSTDOUnlYK1ksA4p3Kym-sDd7Nzj2r7__2fYqkxo/edit?usp=sharing

• Code repository for our Deep Learning Adventures 🎉
  https://github.com/georgezoto/Deep-Learning-Adventures

• Join us on Slack:
  https://join.slack.com/t/deeplearninga-nmk8930/shared_invite/zt-fupjj8pi-8Mb_D6dVRjTrrG9cktaehQ

• Spread the word about our meetup 🎉

• Coding session on real world data - Sequence models/Recurrent Neural Networks/Transformer models

• Step 1 😀
  Learn more about the award winning TV show Friends, an American sitcom television series, created by David Crane and Marta Kauffman, which aired on NBC from September 22, 1994, to May 6, 2004, lasting ten seasons. With an ensemble cast starring Jennifer Aniston, Courteney Cox, Lisa Kudrow, Matt LeBlanc, Matthew Perry and David Schwimmer, the show revolves around six friends in their 20s and 30s who live in Manhattan, New York City.
  https://en.wikipedia.org/wiki/Friends

• Step 2 😀
  Read more about the EmotionX 2019 challenge and its shared task of SocialNLP 2019 where participants are challenged to predict the emotions induced by dialogue utterances by classifying each utterance into one of seven label candidates: neutral, joy, sadness, fear, anger, surprise and disgust.
  https://sites.google.com/view/emotionx2019/home
  https://sites.google.com/view/emotionx2019/datasets

• Step 3 😀
  Learn more about EmotionLines, the first dataset with emotions labeling on all utterances in each dialogue only based on their textual content. Dialogues in EmotionLines are collected from Friends TV scripts and private Facebook messenger dialogues. Then one of seven emotions, six Ekman’s basic emotions plus the neutral emotion, is labeled on each utterance by 5 Amazon MTurkers. A total of 29,245 utterances from 2,000 dialogues are labeled in EmotionLines.http://doraemon.iis.sinica.edu.tw/emotionlines/index.html

• Step 4 😀
  Read the EmotionLines Paper: An Emotion Corpus of Multi-Party Conversations
  https://arxiv.org/pdf/1802.08379.pdf

• Step 5 😀
  Download and explore the Friends dataset which contains a training, validation and test set
  http://doraemon.iis.sinica.edu.tw/emotionlines/download.html

• Step 6 😀
  Build your NLP classification model using deep learning or any other ML or non-AI approach. Feel free to use any environment you feel comfortable: personal computer, Colab, Kaggle, GCP/AWS virtual machine. Feel free to use any model you are comfortable with and spend some time in hyper-paramater tuning to make sure you are not overfitting your training data while optimizing for train/validation accuracy and loss. We recommend you use TensorFlow so it's easier to comment on each other's code but you don't have to.

• Step 7 😀
  Have fun 🎉 and share your journey, findings, lessons learned, success or failures with us and be ready to take a deeper dive in our code. For us, it's the effort that counts and not the final result. Most importantly you should enjoy exploring this interesting dataset and learn something new 😀

Source:
https://en.wikipedia.org/wiki/Friends
https://www.kaggle.com/rezaghari/friends-series-dataset
http://doraemon.iis.sinica.edu.tw/emotionlines/index.html
https://sites.google.com/view/emotionx2019/home
https://affective-meld.github.io/
https://github.com/declare-lab/MELD